hacker news with inline top comments    .. more ..    4 Jun 2011 Best
home   ask   best   8 years ago   
Enough Is Enough avc.com
1007 points by ssclafani  2 days ago   185 comments top 46
jasonkester 2 days ago 6 replies      
Maybe somebody needs to move this issue to someplace more visible.

I'd like to introduce the concept of Movie Patents. I'd register a patent for the plot device of having the bad guy be the main character's best friend, with additional clauses for being his boss, advisor, or partner. I wouldn't ever make any movies. I'd just sit back and collect royalties from everybody who used my idea.

The reason behind this is that maybe then people who matter would actually notice how silly this concept is. It's so plainly obvious that it's a bad idea that regular people can finally understand what's going on. Eventually, people with the ability to make changes might actually find themselves forced to make changes.

Software, as big as it is, is just not that big. This issue can piss every single one of us off, and it won't make the NBC Nightly News. It can't piss off the people it needs to piss off in order to get it fixed, so chances are it will stay unfixed indefinitely.

DrJokepu 2 days ago  replies      
I really don't think the problem is software patents, it seems to me the real problem is that most software patents seem to be ridiculously trivial.

Here's an example: my employer, a rather small company has invested an enormous amount of resources in developing a specific algorithm. It took years of research and development, lots of money and brainpower, it's very far from being trivial (in fact it's so complicated I have to admit I don't understand some parts of it) and it's light years ahead (in terms of usefulness and speed) of any other algorithm with similar applications.

Obviously the last thing we want now is simply handing it over to our competitors, after all this work. If the patent system didn't guarantee us monopoly on using the fruits of our own R&D, this algorithm probably would not have been developed for a very long time, if ever, by anyone. It would have been innovation that never happened.

My point is that the threshold of what can patentable should by way higher, but I think we need a (software) patent system none the less.

pkteison 2 days ago 2 replies      
I spent a week vacation with 2 friends who work at the patent office, so w e had time to discuss over beers. They don't believe there is a problem. They rely so heavily on prior art, and have trouble with the very concept that something was too obvious for the first 20 programmers to want to patent it, so prior art is very incomplete. I show them 1 click, they tell me that if it was that obvious someone would have patented it before. They refuse to consider the "skilled in the art" obviousness test because they say it's too easy to believe something is obvious after it's been explained to me.
tedjdziuba 2 days ago 4 replies      
Alright, Fred. We all agree with you here, the people you need to convince are the ones writing the laws. You're a wealthy and powerful person, who is good friends with other wealthy and powerful people, so why not leverage that? Hire a lobbying firm, get actively involved in the cause.

Whatever you do, don't count on Reddit and Hacker News to do it for you, that's just slacktivism.

statictype 2 days ago 2 replies      
Mostly agree, except:

Second, Lodsys didn't even "invent" the idea. They purchased the patent and are now using it like a cluster bomb on the entire mobile app developer community.

The fact that Lodsys bought the patent and didn't invent the idea themselves seems entirely irrelevant to this dilemma. Unless you also think that if you acquire someone else's software, you don't have the right to sell it.

ChuckMcM 2 days ago 2 replies      
"I believe that software patents should not exist. They are a tax on innovation. And software is closer to media than it is to hardware. Patenting software is like patenting music."

This is what I fear. I worry this idea will take hold and we'll get copyright protection for software 'design' instead of patent protection, and then that happens, the Lodsys crap will have a lifetime of 85 years not 20.

We need to be able to protect someone's innovation so that they aren't screwed over by some larger competitor taking their idea and running with it. That need is real and continues to exist, we need patents.

Let's argue for the real problem here, we also need a better system for dumping things that should not have been patented in the first place because they fail the 'novel' test. We need jurists that understand the technology they are being asked to evaluate, and we need to require at least three people 'skilled in the art' to sit in on juries in patent hearings. I might even go so far as to create special patent juries for these cases. We need a better way of reviewing a patent and validating it against the state of the art.

What we don't need is to convince policy makers that software is like 'music.' And we don't need people believing that we don't need any protection for people's work should they choose to take advantage of those protections.

tlrobinson 2 days ago 1 reply      
In a previous discussion someone suggested forming a reverse patent troll organization that accepted donations of patents from open source developers and whoever, and used them to counter sue patents trolls. The obvious problems are it's expensive to file for patents and lawsuits, and it doesn't protect against "pure" trolls.

Another idea: a simple community that accepts and publishes every random software idea anyone ever comes up with in order to establish prior art.

Perhaps include mechanisms to discover patent applications linked to relevant keywords, etc.

What would it take to prove the date of submissibon to the system? Is there some type of digital notary that we could send a daily batch of documents?

zhyder 2 days ago 1 reply      
We need a generic fix for the patent system, not something specific to software. I think the only way to do this is to change the process of suing for patent infringement. The patent holder must determine how much investment was made to develop the specific patent [1], and this number should be the upper bound on how much the patent holder can sue for [2].

This would effectively cover the pharmaceutical case where hundreds of millions are invested, as well as the worst software case where only a few thousand are invested.

[1] - This number would need to pass simple smell tests: the sum of these investment amounts across a patent portfolio can't exceed the total expenses in the company's books for R&D, etc.
[2] - If a patent holder sues multiple companies at once, each infringing company would owe only a fraction of the total.

rbanffy 2 days ago 3 replies      
Unfortunately, by choosing to license instead of crushing them early, companies like Google, Apple, IBM, Cisco and Accenture and government bodies like the IRS provided funding for the troll to grow and thrive.

It's relatively easy to say "enough is enough". Actually fighting back is harder and, at this point, almost impossible.

radu_floricica 2 days ago 0 replies      
Posting in this kind of threads is always bringing karma points and leaving me vaguely unsatisfied.

Lots of people speak about the "stupidity" of it all. I see no stupidity here. Just the normal, to-be-expected lean of a big government toward protecting the incumbents.

Also I think people should speak more often about drastically shortening copyright and patent terms. They are both very useful, even in controversial domains like software or biotech (patenting genes). The problem is that the situation is waay skewed one way. If you want to bring it closer to normal, start asking for 1 year patent terms and 3 year copyright.

laujen 2 days ago 1 reply      
An idea I will throw out there: I always wondered why anyone should be able to hold onto an idea (patent it) if they don't use it. I can't help but wonder if the best way to fix the patent system isn't to overhaul it -- which would be nearly impossible to do due to entrenched interests -- but instead invalidate all patents if they aren't used within a certain time period by the rights holder. In other words, use it or lose it.
iqster 2 days ago 1 reply      
I'm just fed up with the stupidity of the status quo. As a number of posts have pointed out, the lawyers, Patent Office, etc. generally don't see this as a problem. People like us who actually want to create new things ... we're the ones who are the losers.

This is another instance where it would have been beneficial if all software professionals belonged to a Guild or union. If we were organized, at least there is some sliver of a chance that we'd be able to have an influence on the legislative process. If that fails, we would just not work to create software patents.

tomelders 2 days ago 1 reply      
Can we trick a patent troll into patenting the Triforce and just let 4Chan deal with this?
lmarinho 2 days ago 1 reply      
If you are at the end of a hard day's work, struggling with a tough problem and making no progress at all, even regressing at some points, take solace in this: you still did a lot more for the world than the guys at Lodsys
brudgers 2 days ago 0 replies      
>"They [software patents] are a tax on innovation."

While I tend to agree in general, the Lodsys situation is not a particularly good example in support of the position from an intellectual standpoint, though it may be good for rallying the troops. This is because Apple's actions in regard to in app purchases (IAP's) have been explicitly anti-innovation, i.e. Apple has required a specific monetizable IAP protocol in lieu of allowing developers to innovate (one does not need to even get into questions regarding the innovation raised by Apple's patent portfolio regarding UX elements). Let us remember that developers are only vulnerable because they are complying with a technical mandate of their agreement with Apple and that this mandate is solely intended to produce uniformity within the IOS ecosystem. Given Apple's patent portfolio, they may have far more to gain in terms of IP protection by upholding Lodsys's position than by fighting it.

MatthewPhillips 2 days ago 0 replies      
I think there are 2 main problems here.

1) Patenting obvious stuff / stuff with prior art. "X, but on the internet"

2) Patenting stuff with no intention of releasing a product.

I see a lot of talk about the former and not enough about the latter. My question is this, could we put a time limit on releasing a product once a patent is awarded? Is it normal in other industries to patent something as soon as possible or do they wait until a product is ready to be released? What would be an acceptable time limit? In this case, the patent was issued in May 2007, 4 years ago. Is it reasonable for the inventor to not have a product using this on the market yet?

jaekwon 23 hours ago 0 replies      
I think a major potential for disruption is right here under our nose. Instead of approaching it from a political point of view (blame those politicians!) why don't someone with money create an alternative community-supported system of work attribution?

You can imagine each company keeping a private/internal log of innovation in Merkle trees (with timed signatures by some central authority). All innovation and production is assumed to be original works of art until proven otherwise. In the current patent system, all innovation is public, but there is only 1 winner. In the new system, everybody can win if we all arrive at innovations independently.

Problem solved, + you enforce good documentation practices.

This is an engineering problem. Why don't you fund a startup that builds this system? If we all start using it, that's when we have the power to change the status quo. If we don't know what alternatives are out there, we have no chance of disabling the current system.

I doubt that keeping the government out of software IP disputes altogether is a good idea, because realistically that won't work out -- it'll encourage stealing and the playing field will change for the worse.

stcredzero 2 days ago 0 replies      
The whole thing is nuts. I can't understand why our goverment [sic] allows this shit to go on

Because software and technology are like magic to a large part of the population, including a large segment of those involved in government and the judiciary.

Patents are just another specialized form of legalese. It's obvious we need better ways of challenging patents and filtering them in the first place.

SoftwareMaven 2 days ago 0 replies      
As much as a wonder if there is a point, it can't hurt to write your elected officials and tell them, in reasonable, cogent terms, why software patents cause problems and how they should be fixed.

You can find and write them here: http://capwiz.com/c-span/dbq/officials/

orijing 2 days ago 0 replies      
> I can't understand why our goverment allows this shit to go on.

It's because Congress has been embroiled in a bitter fight of brinkmanship over who can stall longer over raising the debt limit over the past year. Not much actual progress has come out since they decided that they will keep up the debt limit talks, knowing completely that the partisanship will yield no progress.

While they have debated raising the debt limit, the debt has risen by over half a trillion dollars. They need to fix that and move onto other things. The point of Congress isn't to debate solely over how to balance the budget. They have other responsibilities too, and they need to get to it.

LiveTheDream 2 days ago 0 replies      
Would there be any value in a large collection of prior art software, designs, "look and feels", etc? The idea being that since we feel that many software patents are trivial, why not actually implement the ideas and thus provide a line of defense against silly, trivial patents.

Very novel, complex software that might be worth a patent would be very unlikely to turn up in such a collection because of the effort required. Ideally, this would help create an ecosystem where patent-worthy software is awarded a patent, and trivial software is not (because of prior art).

This is an idea I've thought about for a long time. One could argue that much of this idea might already be implemented by websites like sourceforge, github, etc. A focus on explicitly being prior art would probably help, however, justifying the need for a separate archive of ideas.

vilya 2 days ago 0 replies      
The point of a patent is to ensure that ideas are made public, so that all can benefit from them. In exchange, the inventor gets their time-limited monopoly on the idea.

Unfortunately software patents are useless for this purpose: they don't fulfil the function of making ideas public. Who, in this day and age, refers to a patent for the details of an algorithm?

I would argue that open-source software, in many ways, succeeds where software patents fail.

drcube 2 days ago 0 replies      
I'm sorry, but I've got a patent for "outrage on the internet", so I'm going to need to see some royalties.
MatthewPhillips 2 days ago 1 reply      
I like the idea of losing patents if you don't have a product in the market within a year or so.
grimen 2 days ago 1 reply      
Somethng like this scares the hell out of me:


Now I'm in Europe where this don't apply, but we actually had plans making our platform available in U.S. later on with something that has to do with tagging images - which itself is hardly any innovation (?!!)!. I mean, c'mon. We will now probably avoid that and look east instead. I'm still fascinated that they managed to get thorugh with this, almost like the 1-click-purchase that Amazon trying to paten. It's almost like a fairy tale - with no happy ending.

hnsmurf 2 days ago 0 replies      
What we need is a trade association to address this issue. One specifically focused on software startups would be nice. It'd probably have to be funded more by investors and recently successful startups. Maybe Fred should give it a shot.
stusmith1977 2 days ago 1 reply      
I'm not a fan of sotware patents, but perhaps there's a halfway house towards getting rid of them:

Make software patents only holdable by a real person (i.e. not a corporation), and not transferable. Any money received from a patent infringement lawsuit would be paid to that named person.

That way, they can be used for their original purpose (to protect small inventors and allow them to build a larger business), but can't be collected by the trolls.

(It would also give companies an incentive to keep 'their' patent owners happy).

Just an idea... I'm sure there's a glaring hole in it somewhere.

seats 2 days ago 0 replies      
I appreciate this thread and I think that the movie plot patent is a fantastic analogy so great topic starter submission.

I have to say though that I am disappointed by the blog post. Based on the title and the topic I kind of expected something, anything as a suggestion, instead of just a rant and a loose plan to spam every elected official you meet.

huherto 2 days ago 0 replies      
Other than patent trolls. Who is supporting software patents?

I understand that big software companies. (IBM, Oracle, Apple, etc) own a large portfolios of patents that they can use to defend themselves. But, do they really support software patents or they just use them to play the games they have to play?

tttp 2 days ago 1 reply      
I hope that avc, ycombinator, Andreessen Horowitz and a lot of known VC and business angels introduce a new simple rule on their agreement:

To get funded, a company must not try to obtain software patent, and if they do have software patent(s), they will publicly say they will not try to license them and will not sue for infringement of one of their patent.

Beside clearing up the landscape, it will send a clear message, hopefully that is going to be heard by politicians and media: software patent doesn't help innovation, quite the opposite, the guys that fund innovation refuse to give money to companies (ab)using software patent.

Moreover, as VCs seem to be able to work on templates for the paperwork to lower the cost of the creation and founding of a start-up, couldn't they work on standard responses to lower the cost of defending against a patent troll as well ?

drostan 2 days ago 0 replies      
Does it really promote innovation to advantage 2nd movers? Sure it is possible to add to the software and improve it, but often, the innovation is simply execution, buzz, sales and other non-trivial business things. How can there be a reward for the equally important talents of having the vision for new solutions and ability to create them if there is no patent or licensing protection?

Also, it scares me a bit when someone as influential as Fred takes this to his blog - read by so many startups. VCs make money from execution and have an (or another) incentive to suggest that patents are not an acceptable tool for startups.

I agree with the need for patent reform, actually, but a) I'm not sure eliminating software patents is the answer and b) would encourage each startup to make their own decision on how to use the current laws to give them every advantage they can get.

joeburke 2 days ago 0 replies      
This article is much more about the evilness of patent trolls than about software patents.

My thoughts:

- Prevent the existence of trolls (for example by mandating that only patents that are central to your own business can be enforced).

- Raise the bar for the acceptability of software patents.

- Reduce the lifetime of software patents to something more in line with the software industry (a few years maybe).

But by all means, preserve software patents, they do have some value if used as they were designed for (protect companies that invest into R&D).

atirip 2 days ago 1 reply      
Why not incorporate in Europe? Ireland, like Google? Software patent problem solved.
zarify 1 day ago 0 replies      
Why not simply remove the ability to sell the patent to someone else?

Sunset the validity of the patent like the original scope of copyright, and limit the transferral to being able to license it for use so that the only financial benefit goes to the original R&D crowd. That way it still promotes research but limits the possibility of patent trolls since they'd have to actually do the research themselves.

generators 2 days ago 0 replies      
kleiba 2 days ago 0 replies      
Patenting software is like patenting music.

Oh god, please don't give them ideas...

levifig 2 days ago 0 replies      
And then there's iA Writer and their egotistic attempt to patent "Focus Mode"… Seriously: https://twitter.com/#!/iA/status/74588465953640448

Same problem: other apps used it because it's such a generic concept. They (always) say that it's for protection… Riiiight! :|

jtap 2 days ago 0 replies      
So all of technology including manufacturing has gotten quicker, faster, overall better, but the patent system hasn't changed. Also the entire cycle of a company is speeding up. An easy example the time companies that are listed on the s&p 500 are getting shorter. Instead of getting rid of patents can we just shorten the time that they are valid. Say 5 years.
jcarreiro 2 days ago 0 replies      
I don't understand why someone who invents a better combustion engine should deserve a patent while someone who invents a better computer algorithm should not.

That doesn't imply that I think that every bit of programming deserves the protection of a patent. But if someone invents a new algorithm, then why doesn't that deserve a patent?

imtyler 2 days ago 1 reply      
I think America's patent & copyright laws are in need of a serious rethinking in the digital age, but what are we the people doing about it? I would love to get involved (or donate) but I'm not aware of any credible groups or projects making strides to help bring about such a change. Any noteworthy efforts out there that I should know about?
iamben 2 days ago 1 reply      
Why not diminish the value of a patent once it's sold? That way you prevent companies like Lodsys buying and trolling, and if you put the money / time / resources into the development of something, you can reap the reward.

Just a thought, not the answer. But the system is totally ridiculous as it stands.

timedoctor 2 days ago 0 replies      
The only real solution for these companies is to incorporate offshore (or to have no money). This could be a significant long term issue for US companies, and makes me more hesitant about incorporating in the US where I am vulnerable to patent trolls.
mfn 2 days ago 0 replies      
Just wondering, would these still apply if the app developer was in a country other than the US, but the apps were being downloaded from an app store by US citizens?
mrkva 2 days ago 2 replies      
What does VC mean? Vinylchlorid?
meow 2 days ago 0 replies      
The cave man who invented 'wheel' was knocking on my door today.. apparently he was granted a 4k year patent for cart wheels by Unbelievably Sloppy Patent Troll Office (USPTO)...
tomelders 2 days ago 1 reply      
Sadly, Software® patents are here to stay, so long® as the patent trolls® and lobbyists®© can find enough people in the "corridors of power"® who'll shamelessly accept their cheques to maintain the status quo.

The problem is inherent in the fact that people®© are flawed. And by "flawed", I mean "cunts"®

Unless I'm wrong. In which case we can expect software patents to disappear next week.

How AirBnB Became a Billion Dollar Company davegooden.com
617 points by cabinguy  3 days ago   184 comments top 50
Mizza 3 days ago 4 replies      
Is anybody else slightly disturbed by all of the 'attaboy!' comments here on a post exposing unscrupulous and probably illegal business tactics?

Think the response here would be the same if it was MicroSoft or the RIAA caught in something like this and not a YC alum?

Anyway, great investigation and great analysis here, I think.

nhangen 3 days ago 1 reply      
This is the same thing spammers do to steal money from people, and it's the same thing that dating sites do to pretend to have women interested in meeting dudes. Either way, it's fraudulent.

I don't hate that they did this, but I hate that they tried to pretend they were anonymous women that happened to really like AirBnB. Why be so shady about it and instead why not just be honest?

"Hey, we noticed your rental and thought you might like something we built. It's helped a lot of people fill vacancies and make money; check it out - url"

tlrobinson 3 days ago 0 replies      
It's interesting that a lot of the recent billion dollar internet companies have some sort of gray/black hat streak. Off the top of my head:

Groupon: "we've already seen businesses complaining that the Groupons didn't make them money, or that Groupon sales people suggested they raise their prices substantially just before the Groupon 'discount'." http://www.theatlantic.com/business/archive/2011/05/why-does...

Zynga: "I Did Every Horrible Thing In The Book Just To Get Revenues" http://techcrunch.com/2009/11/06/zynga-scamville-mark-pinkus...

pgroves 3 days ago 1 reply      
I started my own company to get away from the do-anything-for-a-sale attitude of corporate america. I do not admire or respect AirBnb for this behavior. I don't care that it worked. I don't care if it was legal or not, it was dishonest. I know it's common, but it takes a fundamental lack of respect to lie directly to another human being.

I'll just continue to toil in obscurity; self-respect intact.

tomkarlo 3 days ago 1 reply      
Everyone's focused on the ethics here, but if this is indeed what AirBnB was doing, isn't it a massive (and repeated) violation of Federal laws on unsolicited emails (for one thing, you're required to identify yourself properly) and likely other rules as well? What this article alleges isn't just questionable marketing practices, it's potentially deceptive / illegal marketing.
rkon 2 days ago 1 reply      
These types of messages definitely appear to fit the CAN-SPAM Act, which covers "any electronic mail message the primary purpose of which is the commercial advertisement or promotion of a commercial product or service".

Airbnb may be spending that $100M sooner than they thought: "Each separate email in violation of the law is subject to penalties of up to $16,000, and more than one person may be held responsible for violations. For example, both the company whose product is promoted in the message and the company that originated the message may be legally responsible."


ChuckFrank 3 days ago 1 reply      
Clearly the unethical part of this discovery is that Airbnb used 'shell' gmail accounts to do their work. Instead they should have found a way to present themselves with honesty and integrity to future members.

What are the good reasons for Airbnb's success and valuation? Are there any stories about their actions that inspire both confidence and congratulations. (aside from their Obama O's and Capt'n McCain flash success).

Joakal 2 days ago 0 replies      
Seems to be happening for over a year so far.

> It never fails that I get at least 20 marketing emails a day from airbnb when I post a property on craigslist. I hope they do more of these PR stunts than filling up my inbox with unsolicited messages.


amirhhz 3 days ago 0 replies      
Is this something the the AirBnB guys would put on their YC app form if they were applying again? And then what would PG & co make of it? In fact, did pg know about this or encourage it in the first place?

Answers to these questions might be a good test of whether they should have done it or not.

BYurgi 2 days ago 0 replies      
> After harvesting email addresses (I only grabbed real email addresses, not anonymous craigslist addresses) I did one email blast to people that were advertising vacation rentals on craigslist...My results: 1,000+ vacation rental owners signed up and listed their properties on my test site.

He's missing a few steps...
How many email addresses did this dude harvest and blast to get 1,000+ people to sign up for his special vacation rental site?
What did the email say that he sent out?
Was he also sending them from throwaway gmail accounts?
What were his tactics?

This guy needs to follow up with more details. All he's shown is that at some point he posted 4 vacation rentals to Craigslist and he got 5 emails linking to AirBNB.

His findings rely on the fact that he got 1,000+ people to sign up their vacation rentals on his site after spamming some number of emails he admits to harvesting.

The "test" site he made (he's since edited it out), but google cached it http://bit.ly/iOPWi5 is www.mimbeo.com.

A quick look at the press page (http://mimbeo.com/vacation-rentals/press) and they have press releases announcing they got 1,000+ properties in one month: http://www.prlog.org/10408565-mimbeocom-reaches-1000-vacatio...

Here's a quote:

> "It's a no-brainer" said one of the founders. "We are offering owners and managers the same great service they receive from the pay-for-placement sites like VRBO & Homeaway. Not only that, but we broadcast our members listings to Facebook, Twitter, Myspace and more - and we do it all at no cost. We like to think of ourselves as the 'craigslist' of vacation rentals - with the added bonus that the listings do not expire - which is the main complaint we have heard from craigslist users."

Seems odd to go through all that for a weekend project "test" website. What's this guy's deal? This seems weird...

marcamillion 3 days ago 3 replies      
Is this 'grey' ? In the sense that it can be interpreted as 'shady' ? Hell yes. But is it "wrong" ? I don't think so.

I always thought that getting that first set of customers that will take you to product/market fit was going to be easy. What with the Twitterverse, Facebook and all and sundry. But let me tell you, from experience, it is DAMN HARD!

I guess the true test of whether or not this is 'unethical' is whether those people that signed up felt scammed after using the service.

If I were one of those people that got that email, sure I might feel a bit 'weird' that they presented themselves as 'Jill D' and not AirBnB, but after going through what I have been doing the last few weeks - trying to get customers - I can't say I would be upset.

allanscu 3 days ago 2 replies      
Good research. I don't think this is anything new. I'm familiar with a few YC companies that use CL as a method to get their supply (in this case vacation rentals).
euroclydon 3 days ago 0 replies      
One day someone will come along and scrape the data from that "clunky and expensive AirBnB" to start an XX million dollar startup.
techcofounder 3 days ago 1 reply      
can we please get a response from one of the AirBnB founders?
nhangen 3 days ago 1 reply      
If you were an investor, would you feel lied to, or is this a cause and not an effect?
Lucadg 2 days ago 0 replies      
I lead a group of guys trying make money online with the rental business, so they can travel.
Each one is responsible for an area and it's hard to get new owners and apartments, believe me, but we never resorted to spamming.
Instead we contact the Owners directly and openly, then we go to meet them personally and see the apartments, when possible.
This way is much slower, but at the end we have much better listings for our customers.
We'll never be a billion dollars company of course.
Zakuzaa 2 days ago 0 replies      
Curious to know PG's opinion on this.
kcurtin 3 days ago 2 replies      
i think people would have a much different view if this wasn't airbnb and was a less reputable company. The things described in the post definitely constitute as spam and violate CAN-SPAM...but these types of tactics do work (if they didn't we wouldn't have spammers) and there wasn't anything malicious about it.
dreamdu5t 3 days ago 1 reply      
AirBnB is not a "Billion Dollar Company." Investment is not revenue or profit. The revenue numbers I heard are in the 5-10 million range.

When AirBnB posts a billion dollars in revenue, then it will be a "Billion Dollar Company."

buildorfail 3 days ago 2 replies      
I do not want to argue if this type of stuff is good or bad. But I do want to point out to you all that list of tech gods is full of similar stories. Off the top of my head:

MySpace started out purely through spam...Tim Ferris started out in online supplement sales...People complained about Plaxo doing black hat stuff... they sold ok. People complained about Zynga ripping off other games and about breaking every Facebook platform rule there was... look at Mark Pincus now. People complained about Facebook's Mark Zuckerberg stealing their idea... he does not care.

aresant 3 days ago 4 replies      
"It is a story of pure unadulterated hustle. And I love it." - Fred Wilson talking about the AirBNB team's Obama cereal hustle.

I'd like to borrow his words in regard to this post.

Fred's original post - http://www.avc.com/a_vc/2011/03/airbnb.html

brown9-2 3 days ago 1 reply      
When someone says "Craigslist spam!", is the first you think of a) spam posted to craigslist as listings, or b) emails advertising something (surreptitiously) to people that have listed their property on craigslist?

Not that the latter isn't "spam" of a sort, but it doesn't feel quite like "craigslist spam".

jmtame 3 days ago 0 replies      
In Startups Open Sourced, he says Airbnb "tried everything" but he says that press is the single best source of their growth. Having the happy customers came in second, but would be interested to hear if this helped them much. It's hard to believe that Craigslist is the reason why they'd be so successful when the product is so easy to spread by word of mouth and press.
rvanniekerk 3 days ago 3 replies      
Good read, although, I wouldn't necessarily consider advertising your services to people already listing vacation rentals as "blackhat".

Blackhat insinuates that someone is being gamed, in this case, both the home owners and AirBnB are benefiting (maybe craigslist is losing out?).

wickedchicken 3 days ago 1 reply      
For the record, I don't think this is that bad of a thing. If all of this is actually true (I'm not 100% sure it is), then it's certainly not on the level "actual spammers" perform at. Spam is basically unwanted and harmful advertising, and nobody complains when someone gets e-mailed an actual useful relevant thing out of nowhere. If the thing you're being e-mailed is more relevant than the ads GMail is showing you then I'm not sure you have a right to complain too much. Sure it may feel slightly off, but calling them "completely unethical" is going slightly too far.
SkyMarshal 3 days ago 0 replies      
"Some very famous investors have alluded to the fact that they look for a dangerous streak in the entrepreneurs they invest in…and while those investors will never come out and tell you what they mean, this kind of thing is probably what they mean."

Eg, be naughty, not evil. Whether they crossed the line or not, dunno. It's a far cry from Microsoft intentionally leveraging their OS monopoly to destroy companies and corner industries back in the day.

tzm 3 days ago 1 reply      
Unethical? possibly. Does it piss me off? Sometimes. Do I care? No.

I generally expect to receive spam from CL when posting or responding to ads.

ivankirigin 2 days ago 0 replies      
I hadnt heard about anything like this from the airbnb folks, but I do know they have had successful advertising campaigns since over a year ago. So the basic premise of this article, that their growth source is a big mystery, is false.
Hisoka 1 day ago 0 replies      
Whether or not it's wrong or right doesn't matter.. what matters is are there consequences? If not, then the guys at AirBnB probably could care less, and are laughing and enjoying cocktails in rooftop bars while the rest of us are toiling away at jobs...

Remember, we only got 1 life to live.. think about it. Arguing right or wrong makes little sense

dhbanes 2 days ago 0 replies      
I wonder what nickb thinks of this.
djloche 2 days ago 0 replies      
Supposing Dave Gooden is correct...

The method seems risky at first, but when you think about the risk and how they solved the problem, it is perfectly acceptable.

To catch danger from unsolicited email (from the end recipient), the mail must first be reported. A complaint must be made. If the recipient doesn't report it, there is no action taken. Most of the time, it doesn't take but a few reports and then there are people digging into the scheme trying to figure out who is abusing the end users. This is because most people sending unsolicited email are not able to directly target their market. Viagra spammers have to send out boat loads of messages just to hit their market of old men with erectile disfunction who are too embarrassed to get the pills through traditional means AND not web savvy enough to realize that giving out their credit card information to a fly by night company selling drugs isn't the smartest idea.

Simply put - if you can directly target your market AND you have something that is game changing to those people - eg a superior product / experience / service compared to their current business efforts, you won't get flagged for spam. They'll check out your site, think it's the greatest, and become your customer because you make them money.

The problem for AirBnB was that Craigslist has evolved to try and prevent unsolicited mail of all kind - because most people or companies are NOT able to hit the sweet spot where they are sure that what they have to offer is exactly what the recipient wants, but doesn't know it until the email is received.

Having interns, family, friends, or other people who are genuinely excited about your business send email recommending it to exactly who will want it, without revealing their relationship to the business, is an acceptable, if temporary solution.

But you only need to build momentum, because if your product is amazing, the recipients of the previously unsolicited email will do all the word of mouth marketing for you.

So before you go and try and repeat this alleged successful route, make sure you have all the pieces to the puzzle before giving the mission a green light.

[cite: personal past experience at an ESP watching and doing the digging in resolving spam/fraud complaints]

temphn 3 days ago 1 reply      
AirBnB raided Craigslist for users. Those users made more money and had a far better user experience on AirBnB to boot. Not exactly your traditional Viagra/Nigeria type spam.

When the end doesn't justify the means, sure, complain about the means. But this is not such a case.

gigantor 3 days ago 1 reply      
Blackhat or not, this is a great example of how there are very viable alternatives to SEO in terms of internet marketing. The ever changing rules and laws of search engines unfortunately make or break a vast majority of online businesses.
forgotmyuser 3 days ago 1 reply      
Glad you did all that testing to find something so damn obvious. Look at how they rank on google - they don't. It was really obvious they were spamming craigslist.
bproper 3 days ago 0 replies      
Does the new "illegal hotels" law have any effect on AirBnB?


throwaway61802 2 days ago 2 replies      
Nobody here knows how much of Airbnb's growth is driven by Craigslist, it's purely speculative. Until Airbnb confirms, you'll never know but you'd have to be naive to believe that you can build a company up to $1 billion by firing off a bunch of emails. Nobody here has even suggested the possibility that "Airbnb became a billion dollar company" by suggesting how many Craigslist rentals are posted each day, along with a standardized conversion rate.

I'm surprised to see how negative the comments are on this. This is like Yelp paying people to write reviews. Shady? Sure. Illegal? The only part of the CAN-SPAM Act Airbnb broke was (1) not including a physical mailing address, (2) not mentioning it was an ad, and (3) not including an unsubscrive link. Otherwise, CAN-SPAM is laughable at best. It's hardly enforced. Don't expect Craigslist to come at Airbnb with a lawsuit, it'd probably not hold up in court.

I see this instead as one of Airbnb's many tactics to try and reach out to users. If you look at this logically, they've been covered by the press a lot. The amount of growth they might (or might not) have reached doing this would hail in comparison to how much they obtained by press and simple product and customer development.

theklub 2 days ago 0 replies      
It really shouldn't shock anyone that very few people follow the rules.
hnsmurf 3 days ago 2 replies      
"It is a literal gold mine for black hats that learn how to exploit its millions of users and curtail its terms of service."

Either he doesn't know what "literal" means or I need to get a pickaxe and head on over to Craigslist.

mbertrand 2 days ago 0 replies      
Interesting amount of negative feelings towards AirBnB, and as a disclaimer I am a big proponent of "Do, and ask for forgiveness". This is what startups are all about, using available resources to their advantage and getting out their and hustling (even if it involves bending the rules a bit). Obviously rental owners (their target) were not to upset about being told of another resource to list their site on otherwise they would not have continued this approach. I can see where some of you are coming from in feeling violated/disturbed by this but given that a.) it appears to have worked b.) they informed users of something they wanted c.) no one complained (until now). I this that it was a very effective/creative way to solve their supply problem and it obviously worked.
tuckbuck 2 days ago 0 replies      
great writeup. AirBnB may not be wrong, but what this means, it that anyone else can do the same thing .

this definitely reduces their valuation.

bakbak 3 days ago 1 reply      
just wondering what s/w do you use for mass mailing and craiglist email harvesting? without getting sandboxed / blacklisted ...
johnx123 2 days ago 0 replies      
I believe, PG spammed more than what the two ladies did in CL.
adrianwaj 3 days ago 1 reply      
Why doesn't and didn't craigslist check for the almost identical emails if they are going through their servers? They could likely dig them out if they keep archives by searching on airbnb, nicest, largest, page, views. Airbnb have the funds now to cough up.
gorm 3 days ago 1 reply      
Welcome to The Mesh!
andrewcross 2 days ago 0 replies      
While this is clearly not the most ethical behavior, how many of us wouldn't do the exact same? You gotta do what you gotta do.
savrajsingh 3 days ago 0 replies      
alluded, not eluded
andrewhillman 3 days ago 0 replies      
Sounds like a lot of people on here are jealous. Obviously they weren't being ethical, but I think just about every popular web startup does what they need to do to solve the chicken and the egg problem. I think every single dating site puts up fake profiles, sends fake msgs, winks and flirts to get started.
orenmazor 3 days ago 1 reply      
The author uses the term "black hat" way too much and very loosely.
wickedchicken 3 days ago 1 reply      
Do you know what else is scummy? Writing a smear webpage about a competitor complete with intrigue involving leet "black hat software" and linking that page to your for-profit clone (mimbeo.com). For all we know this dude falsified the AirBnB e-mails himself just to make them look bad.
MatthewB 3 days ago 6 replies      
I love this. Being scrappy is one of the best qualities among entrepreneurs. The attitude of "do anything" to succeed usually results in success not to mention it is one of the characteristics that PG looks for in founders.
Poll: Should HN display comment scores?
602 points by pg  6 days ago   280 comments top 135
mdasen 6 days ago  replies      
This wasn't my idea (someone else had posted it), but I think it would be nice to see the usernames fade from grey to green as the comment score increases (maybe capping at 10 or 15). That way, one can quickly scan and see the best comments, but there isn't such a direct relationship as a numerical score. For example, the username could go through these steps:

. . .

That way one can even faster pick out things that people have generally agreed is of good value, but it isn't immediately obvious what the exact score is - there's a fuzzyness to it - and there's also a cap on the displayed score (in that you can't get more green than 00FF00 and starting at 004400 would cap it at 12).

It fits my use cases which are skimming for good comments and figuring out if something I don't know much about has value (for example, if someone says, "they should have done it this way. . .", it can act as a barometer of the suggestion's value). It also seems like it would be easy enough to implement.

I'm not going to vote on the poll because I've actually found that the level of discussion has felt better without the votes shown. I enjoy the site more for the reason you've cited. Not to psycho-analyze too much, but the lack of vote scores removes the pressure to have the best comment or a better score than someone that has a different perspective. Even if two people are being totally respectful to start with, it can become psychologically difficult when the other person is getting more votes. One might try to make it a debate (that they're trying to win against the other person) more than a discussion (in which two parties are trying to figure out the truth together).

However, the site has become a bit less utilitarian and it sometimes does take me longer to weed through the information in the thread. Maybe varying the color of the usernames based on the comment score (and capping at 12) would add some limitations and fuzzyness to it that would meld the two. Capping at 12 would mean that both parties arguing might get the same public presentation of 00FF00 and that might quell the need for parties to prove that they're the winner of the argument by popular vote. Likewise, humans don't perceive colors exactly and that might add another layer that would diminish people comparing themselves so much and trying to score points. I guess I think it would be interesting to see if this would be a nice balance.

Example: A person posts that they think VPSs are better for hosting than renting a physical box and they talk about their reasoning (machine images that you can bring up more boxes of, launching new instances within minutes, whatnot) and it's a good comment. Someone replies espousing the virtues of physical hardware (a tad more speed, not sharing IO, whatnot). Now it becomes a bit of a competition between the two ideas (and the two posters). They were both good, valuable comments about different approaches. There is no right answer and there might not even be a better answer. The community knows this and both have comment scores above 12, but each person feels pressure to "win". With the green usernames, they're both at 00FF00 and have no idea if the community has given the other person more votes. There's no need to score points off each other or need to defend one's ego. You know you've made a valuable contribution and the other person has also, but for all you know they've gotten a good fewer votes than you. It still allows everyone reading the two comments to know that they're both generally good advice. So, there isn't a fight and an on-looker can see that both comments should be read and headed.

lionhearted 5 days ago 6 replies      
This was a reply I wrote to tptacek (first quoted part his) -

> Meanwhile, not having up-to-the-minute scores makes the site more pleasant to participate in; one isn't prodded to make statements in reaction to ludicrous (and likely ephemeral) voting swings.

I feel the same. The change makes the site a bit harder to consume - particularly, skimming-quickly-for-a-couple-good-points is harder.

But it makes the site much nicer to participate in. First, I don't feel the need to reply to mean, nasty, or incorrect replies to me unless I have something to add. Second - and I think this is really nice - voting has become less about promoting an argument or viewpoint and more about saying "thanks for this" or "less of this, please" - I find myself voting up thoughtful stuff I disagree with more often now, and voting less frequently overall.

Personally, I'd say the site is harder to consume efficiently now, but more pleasant to interact with.


It'd be interesting to see voting by people who are regular/frequent commentors vs. mostly lurkers, if there was a way to segment it out somehow. I bet the response rates are different between the two groups.

tokenadult 6 days ago 3 replies      
Hi, Paul, I see this poll is "official," because you are the participant asking the question. I don't think I ever saw the particular threads you noticed that prompted your thread-opening post with the title "Ask HN: How to stave off decline of HN?"


in which you said 55 days ago, "The problem has several components: comments that are (a) mean and/or (b) dumb that (c) get massively upvoted." I have never known exactly what the problem was previously. I suppose few people read this site exhaustively now that the message volume is so high.

Since you announced you would experiment with changes in the site,


I haven't noticed any inconvenience from comments not having visible scores. I have noticed several threads with much discussion of that meta-issue. And I think you have answered your own question here about the issue of site quality when you say, "I prefer HN without comment scores, because those fights really disturbed me, and they've practically gone away since I hid comment scores." I too like the new way of doing things. Some discussions that I participate in seem to have been better than they would have been three months ago.

But in one thread complaining about lack of visible comment karma scores that was fizzling out as these suggestions were made, several users wondered if a user who had accrued enough karma to upvote, downvote, and flag might be allowed to optionally view comment scores while giving up voting power. That's an intriguing idea. To some participants here, finding the comments with highest point values for skimming threads is perhaps more valued to them than voting. I have thought about that issue for a while. It seems to me that there are several possible responses to the suggestion that voting power be separated from visibility of comment karma scores. Perhaps the learned readers here can suggest other possible responses. In any event, I think it is your call to decide what to do.

If HN offered participants a choice of either voting (without comment karma scores visible) or seeing comment karma scores without having voting power, would that be a good idea? Possible responses include:

1) Yes, then some readers can skim threads for information, while others vote on comments, and everyone is happy.

2) Yes, because readers can skim threads for information, even if that inconveniences voters.

3) No, because comment karma scores are misleading as a guide to what to read. (See the post from just before when the experiment began


in which some kinds of high-karma, low-value comments were identified.)

4) No, because everyone should be able to vote, and everyone should have a clue about which comments have a high score (through colors or fonts or approximate scores).

5) No, because the interface should be like it was last month, when everyone could see comment karma scores and could vote based on personal karma.

6) No, because HN users will use sockpuppets to get around any such distinction between viewing karma and voting.

Personally, I haven't committed myself to any one position on this issue, or indeed even committed to whether this is a fitting way to look at how to improve the site. I appreciate you asking for user response, and I wish you all the best in setting site defaults that build a civil, thoughtful informative community on HN that helps innovation flourish.

cygwin98 5 days ago 1 reply      
I have a theory that showing comment scores benefits each individual member, especially lurkers (myself included); on the contrary, hiding them will be good for the HN community in the long run, however, each of us will have to sacrifice a bit more of our time.

Here is what I observed during the past two months:

1. I have been more and more inclined to participate conversations, without fearing being outgunned by those celebrities. In old days, almost every thread touched by one of top members quickly becomes a one-man show, other comments (may be more insightful) got somewhat ignored. The consequence is that a lot of us just don't bother to write a word. Now, they are still there, but the herding effect has been much less since, so their advantages are not as obvious as before. The game is more fair. For example, I saw one thread today patio11 was using his BCC as an example, a guy was asking what BCC was. I don't think this will happen in old days if the guy saw patio11's comment had say 300ish upvotes.

2. Those celebrities are not as thoughtful and knowledgeable as what I used to think. I have nothing against them. They gained their fame because they are experts in certain areas and they gave insightful comments there. However, the issue is that they may not be experts in other areas. Without the scores being shown, comments are evaluated in a more fair way, based on its intrinsic value rather than who their owners are. I saw a lot of comments that have spelling error here and there, may not be worded properly, but are really insightful. I appreciate those comments and gladly upvote them. I doubt I would have patience to read them if scores were being shown, more likely, I gloss over them and just skip them because of those minor issues.

In summary, showing scores brings short-term gains to the members, discourages participation and encourages lurking. HN will become cliques around a few key members.

patio11 5 days ago 3 replies      
I personally care most about comment scores for preserving the ability to use SearchYC on a keyword of interest and find good advice. This doesn't work if they're off.
andywood 6 days ago 3 replies      
I'll just point out that making this - or any other measure - into a vote, runs into the same problem that IMO is at the core of the overall "decline of HN" problem. Democratic decisions will tend to be compromised by the diluted community. In other words, I think it would be better if you either do what you think is best, or poll a group of users that you are familiar with and respect highly.
hooande 6 days ago 2 replies      
This has been debated in so many threads. I want to sum up the arguments for comment scores (as see them) and maybe someone else could sum up the arguments against them.

1. comment scores let you quickly scan to see the most interesting comments

2. comment scores provide an additional layer of information ("what did the community think about this?")

3. comment scores make it easier to assess comment quality in threaded discussions (the most popular top level comments will move to the top of the page, but little information is provided about the quality of replies)

As a hacker I'm generally on the side of getting as much information as I can, as fast as I can get it. I'd love to bring back the comment scores and think of other ways to address the problem of nasty arguments.

jeromec 6 days ago 0 replies      
I'm a convert. I last voted to display scores, but now I think hiding them is best.

The reason is because HN was (is) becoming more affected by the same cycle we've seen play out where content from a small set of users which reflect core site values makes the site good, but this draws more attention and users until content quality begins to suffer; quality posters leave, and find some new site which starts the cycle again.

I think leaving off the comment scores is the only way to check that unwanted behavior. We have to deal with the inconvenience of not relying on scores to skim comments, but I think that can actually be a good thing; I think that's just HN at a new (more sustainable) maturity.

JesseAldridge 6 days ago 4 replies      
How about switching comment scores on after a certain amount of time? I've found myself especially wanting them when looking through older threads.
joe_the_user 6 days ago 0 replies      
I liked it better when comment scores were displayed and it is better, now, when comment scores are not displayed.

It alternatively fed and chastised my ego with scores visible. It was marvelous showing who was right and who wrong on teh Interwebs...

But without that, the only thing that sustains me here is saying what actually think. I'm "starved", yet I'm better for it.

tptacek 5 days ago 0 replies      
Put it to a vote and a group of nerds is always going to vote for more bells, whistles, and knobs. All I can say is, the site is more pleasant to interact with without the scores.
credo 5 days ago 0 replies      
Here are some of the reasons why I voted to display comment scores

1. HN is a great community and imo members should be treated as responsible adults.

Comments that are "(a) mean and/or (b) dumb" are a problem. However, I think that the people who make these comments are going to make the comments regardless of whether comment scores are visible or not. After the visibility-change, I didn't see any positive change in comment quality (but I've wondered if comment quantity has increased)

Obviously, this is a subjective assessment. For everyone who believes as I do, I suspect that many others believe that they have seen a qualitative improvement in comments.

Regardless, I think that it will be great for HN to self-police itself as a community of responsible adults (and responsible teens) instead of assuming that HN is an immature community that needs to be protected from looking at comment scores.

2. Hiding comment scores skews the scores. Some people have said that they don't upvote comments because they don't know what the comment's score is.

OTH some populist comments (or comments from popular members) are going to get insanely high scores because upvoters don't realize that the banal comment's score is already much higher than what it should have been.

3. Visible comment scores are a simple and great indicator of what HN feels about the comment. imo this is a very useful indicator for everyone on HN.

ssp 5 days ago 1 reply      
It's not clear to me that "number of upvotes minus number of downvotes" is a meaningful measure of anything, whether its displayed or not. An upvote from a high-quality user should count for more than one from a low-quality user.

If you consider the graph of comments and users, where each vote is an edge from a user to a comment, and there is an edge from each comment to whoever wrote it, I bet there would be interesting information to be found in the eigenvectors.

ericb 6 days ago 0 replies      
If not comment scores, what about a tiered color grading system so people can see what the community thinks. If it was blunt enough, it might dull point-jousts.

For example, perhaps red = highly voted (based on averages), orange = moderately high, yellow = just a bit higher than average. Without an actual "score" perhaps the point-jousting would be reduced. If the jousts continue, you could reduce the resolution in the color grading system, so the incremental incentive for arguing is reduced.

The most important type of info that is lost in the current implementation, is that, as an observer, you can't currently be trained in what the culture of the site is by observation. Because of this, I think the current setting will cause the culture to eventually dilute entirely as it is no longer heritable. Posting comment like "Thanks" or "very interesting!" which, while polite, waste everyone's time is the type of cultural affectation I mean. It is no longer obvious to newbies that this is discouraged--and when newbies are the majority, it won't be discouraged.

The non-voting view comment scores option is very interesting though...

mixmax 5 days ago 1 reply      
Hiding scores to reduce arguments seems like a very crude and inelegant way of solving the problem. You solve one problem but effectively render the scoring system useless in the process.

Surely there is a better way of solving this particular problem, though I'll be the first to admit that I can't think of anything clever.

DanBlake 6 days ago 0 replies      
For threads created in less than 24 hours, why not only display comment scores for comments with over X points? Solves the issue of bickering back and forth but also lets comments which have significant votes get easily scanned.

On threads older than 24 hours there is no reason to hide, so just always display comments if older than 24hrs.

Just a thought, seems like it might work.

wickedchicken 5 days ago 0 replies      
This is going to get lost in the noise and will probably be a bit off-topic, but I've been enjoying HN as of late. I'm not sure what changes have been going on behind the scenes (I hope it's not placebo) but the overall site (comments + submissions) have seemed better in the past few weeks. I'd like to think the lack of comment scores is part of this. Thanks Paul for trying to keep the site well-tended.
mgkimsal 5 days ago 1 reply      
Why is it not simply an option for users? Perhaps only an option for logged in users, but an option nonetheless, would balance both worlds.

For the 'point players egging each other on', that's only going to work when you know that everyone else is seeing the scores too, and it won't be apparent.

msg 5 days ago 1 reply      
For better or worse, comment scores are useful for reading. I know how to ignore flamewars: I look for indentation and skip it. My virgin eyes somehow are not offended.

So I'm much less likely to read threads now unless I'm really interested. I guess that has made me less involved.

nkh 5 days ago 0 replies      
How about showing the comment scores after 24 hours?

I personally like the not having the comment scores while the discussion is active. I do think it makes thinks feel more intimate, and you can not just scan the comments and move on.

However, I really miss the ability to save articles and see which comments where the most valuable at a glance. It would also make sites like "searchyc.com" a valuable resource for comments once again.

jemfinch 6 days ago 0 replies      
Perhaps a more reasonable middle ground can be found: only show the scores after a person votes, a la Slashdot's "you can't comment and moderate in the same story/thread". You avoid groupthink and pissing contests, but you allow someone to see the community response by contributing to the community first.
akamaka 5 days ago 0 replies      
I'm not in favour of bringing back comment scores -- the quality of discussions was definitely declining when scores were removed, and bringing them back isn't going to change that original problem.

If we need more changes, let's try something new.

For example, would encouraging people to vote more often help the situation?

If it's determined that it would, I have a couple of suggestions. First, make it less of a hassle to vote from my iPhone. The buttons are nearly impossible to hit without zooming all the way in, and I often press the wrong one (with no way to undo it). Second, provide an alternative way to downvote. I tend not to downvote something unless it's downright terrible, because I don't want to bury comments that I simply disagree with. If there were a "bury" button for comments that I consider to be spam, offensive, or completely unfactual, I would use the regular downvote button more often for things that I simply disagree with.

catshirt 6 days ago 1 reply      
here are my unqualified 2 cents: have you considered showing a user's average karma next to the comment, instead of the comment points? alternatively, what if you hide the points, but let the user see them on demand (hover, show button, etc)?

i think both of those could help solve the commenting-for-points issue, while still helping users determine which comments are of higher quality.

SkyMarshal 5 days ago 0 replies      
It would be interesting see some stats on this vote. For example, are HNers with high karma more likely to vote for hidden scores? Are newer HNers more likely to vote for displayed scores? How about lurkers vs. frequent posters? Frequent submitters vs. non-submitters?

I'm guessing the poll is anonymous, though...

iamelgringo 5 days ago 0 replies      
I'd love to see comment scores for the first 3 levels of comments.

Arguments typically go on for 5,6,7+ levels, and nobody really enjoys reading those. If that's the behavior that we're trying to modify, why not eliminate scores for those types of comment trees. Or, better yet, how about fading the color of the text as the depth of a comment thread increases? I find deep comment threads are rarely interesting or insightful.

The reason that I miss scores, is because I often skim large stretches of comments, and only read the ones that are above a certain threshold.

Alternately, how about adding a bit of JavaScript that allows comment folding, and letting users define their own score threshold for having comments start collapsed.

Tiomaidh 6 days ago 1 reply      
For normal discussion, I think we're better off without visible scores. But there have been threads where two people go back and forth with two completely-opposite views about something that's not definitely Right or Wrong--and that I'm not very knowledgeable about. I found myself missing visible scores in that case, since I wanted to know what the HN consensus on the matter was.

So I guess you can put me down in the category of "Don't show numbers, but do discretize the points and show colors/symbols/something based on that." One thing I'd especially like is a designation for replies that outscore their parents.

sage_joch 5 days ago 0 replies      
"There was a nasty kind of argument that used to happen, where people would literally try to score points off one another, and users voting on the thread became like a mob egging on two people fighting."

This problem arises from two features: comment scores and arbitrarily deeply nested threads. Considering how valuable comment scores are, it seems worth trying to remove the other feature instead. (Perhaps by preventing replies until after a delay, which increases with the depth of the thread.)

JoshTriplett 5 days ago 0 replies      
I'd like to see something like scores return, simply because I find them a useful tool when reading; whether as an exact number or some kind of rough scale, it helps to know the difference between "meh" comments and exceptional comments, not just the sign of the score. Personally, I'd like to see some information about the distribution as well; a score of 3 could mean +3-0 or +10-7, but the latter indicates controversy while the former indicates "meh, not bad but nothing special" (or alternatively, "meh, not many people around at this hour" :) ).

See http://times.usefulinc.com/2007/03/06-sparklines for an interesting example of distinctions between numerically identical scores; in particular, see the "Middle ranking" section.

For one example of a non-numeric approach, what if comments had two horizontal bars next to them, one with about a pixel per upvote and another with about a pixel per downvote? You could easily draw useful conclusions both from the absolute lengths of the bars and their relative proportions. Consider the following, shrunk to fit within the height of one line of text:


hugh3 5 days ago 0 replies      
I preferred it when comment scores were diplayed.

While I can only comment (obviously) on the effect that the change has had on my own comment scores, it seems to me that it's changed the scoring for the worse. Nowadays it seems that my highest-rated comments seem to wind up being pithy one-liners (I try to resist writing 'em, but sometimes they just come out) rather than the more informative sorts of comments.

I think not having the comment scores there changes the way people vote. If you see a comment that provokes a positive emotional reaction you're more likely to say "heh, upvote" if you don't know its existing score than if you see it has already been adequately rewarded.

code_duck 5 days ago 0 replies      
I'm on the fence about it.

I liked being able to see the view scores when comments that were considered insightful or extra-correct about a technical practice were upvoted highly. I also think that there is a self-reinforcing aspect to high scores, though. Sometimes I downvote comments which I thought were 'overrated' (hearkening back to Slashdot!), which isn't an issue now.

I think there's less voting overall going on now. I cast fewer, myself.

I can't decide which way I prefer, but regardless I like that this site continues to evolve.

bhousel 5 days ago 0 replies      
Did you post this poll Saturday night of Memorial Day weekend on purpose?

Edit: I posted my thoughts on the comments/no comments debate here: http://news.ycombinator.com/item?id=2568995

Basically, it doesn't matter.

HN has become much more adversarial in recent months, and it has nothing to do with whether or not we see a number attached to the comment. You don't need that number to know whether the person who wrote the comment is someone that you probably wouldn't want to hang out with in real life.

My guess is that we're just all strangers now, and that it is human nature to implicitly treat strangers differently than we do friends. What's amazing (to me anyway) is that this change has affected even users whose names I recognize going back years. People whom I used to have a lot of respect for. It's not just the new users.

You either like debating strangers or you don't, and the people who relish these kinds of debates will pursue them even in the absence of a scorekeeper.

makecheck 5 days ago 0 replies      
Curious...polls still show point totals for each answer. It seems that any reason for hiding points before voting on a comment would also be valid for hiding points before voting on a poll.

So if you decide to keep comment scores hidden, I suggest also hiding poll answer scores until the reader's vote has been cast.

brlewis 6 days ago 0 replies      
If I had unlimited time, I would be happy to evaluate every HN comment on its own merits independent of any social proof. Not having unlimited time, I need heuristics.

Nasty argumentative comments should always be voted down, right or wrong. That people gained karma from them is a failure of the community, not the HN UI.

dexen 5 days ago 0 replies      
Please consider displaying both upvotes and downvotes separately. There's a huge difference between a `3 karma' comment with +62/-60, and a `3 karma' comment with +2/-0.

The first one is interesting, but raises serious opposition; the second one is probably uninteresting.

benologist 6 days ago 2 replies      
It felt weird at first but I'm pretty much indifferent now.
flipside 6 days ago 0 replies      
Personally I think a hybrid tier system is the best, simplest compromise.

1. Comments with 1 point stay as they are.
2. Comments with 2-9 points are marked as moderately popular.
3. Comments with 10+ points that are marked as very popular.
4. Comments with <1 points are treated the same as they are now.

Comment score tiers could be color coded or use any other marker to identify their popularity range. Hopefully this would keep things fuzzy enough that people don't fight over points while still making it easy enough to scan comments quickly.

sixtofour 5 days ago 0 replies      
"I prefer HN without comment scores, because those fights really disturbed me, and they've practically gone away since I hid comment scores."

I understand the feeling, but ... look away. :)

The more comments per post, the more valuable comment scores are to me, not to judge the worth of comments, but to help me decide whether to skip long sub-threads or read them. If I see a number of comments have high scores I'll slow down and read a few more posts in the sub-thread. High comment scores tell me that there's high interest in the sub-thread.

SoftwareMaven 5 days ago 0 replies      
I'm having a difficult time voting on this one. In general, I like it better, but there are a few specific cases I've seen where seeing points would have been very valuable, IMO.

I like the idea of removing any visible signs of total karma. If nobody knows how much karma anybody else has (including oneself!), there is no value in trying to game that number.

At that point, trying to make a name for yourself, like tptacek, grelis, or pg have done, is a result of people knowing who you are and the thoughts you've posted before instead of some silly, altogether meaningless number next to your login.

Sukotto 3 days ago 0 replies      
Personally, I get a lot of value from knowing the comment scores. Mostly as a first-pass filter for finding the gems in the discussion. I also really like them to help find the gems in a user's comment stream.

The value I get more than offsets any of the negative points mentioned elsewhere in this conversation (or the many others we've had since you started this experiment)

Please let me view comment scores. Make it a profile option or something.


This comment will likely be unread since the poll has already fallen off the front page. But I care enough about the issue to write it up anyway.

Why did you post your poll over the holiday long weekend? I took my family on a road trip beach vacation (almost entirely offline) so I missed the chance to participate when the conversation was fresh and active :(

vasi 5 days ago 0 replies      
I'm a mostly-lurker. I only comment when I something to say that surpasses my fairly high threshold of "worth spending other peoples' time on".

On this issue, I'll break my pattern and post my half-formed thoughts, because to me HN was much more valuable with scores shown. The two main reasons:

1. Being able to skim just the top few comments allows me to read the threads, even on posts I'm not initially that interested in. Without the scores, I have to make an immediate judgement of whether I want to wade through hundreds of comments to find the diamonds in the rough. I feel like I'm now limiting myself on HN to only those issues I already know about.

2. I'd like to think that people might occasionally notice the more informative of my comments, infrequent though they may be. Without scores, my ideas may be drowned out by the quantity of other posts, which discourages me from commenting.

I understand the need to avoid comment wars, though the fact they're happening at all is rather disappointing. (Is the HN community really that petty that out-scoring those you're talking with is felt necessary, and rewarded? I'd like to think not!)

But let's not throw the baby out with the bathwater: Leave some way to separate the wheat from the chaff. I don't really care whether it's low-precision indicators, or clamped scores, or a setting to filter the highest comments, just so long as there's something.

georgieporgie 5 days ago 0 replies      
Only seeing my own comment scores gives me feedback and doesn't allow me to jump on any bandwagons. Like it or not, I'm influenced by other people's opinions. A numeric score influences my opinion before my critical thinking begins.
mkr-hn 6 days ago 2 replies      
This is a "but at what cost" situation. While it did get rid of most of the worst stuff, sometimes the delicious cookie of a +1 was motivation to put more effort into a quality comment.
mirkules 5 days ago 0 replies      
Maybe we could concentrate less on picking out useful comments by themselves, but instead try to pick out useful conversations instead. So, for example, high-voting activity on one thread could move it up to the top of the comments page. This wouldn't necessarily mean that a comment with the highest score is automatically at the top, and would encourage discussions rather than trying to score points. Just a thought.
rms 5 days ago 0 replies      
I switched my preferences back again. It's nice that the fights have stopped, but fights or no fights, I haven't been reading comments over the last month because it's too hard to filter without scores.
jasonlotito 4 days ago 0 replies      
I'll say again what I've been saying: the current way of using points make them matter in all ways except for the one that matters to the community: as some level of indication of quality. Generally, comments with a high number of points in relation to the thread it's involved with are generally well regarded and considered to add value. Exceptions exist, but they only confirm the rule.

By using points to highlight quality comments, you'd solve numerous problems. You already use this method to hide comments. I don't see any reason not to do this for comments that are regarded more highly.

biot 6 days ago 1 reply      
If voting abuse was a problem, how about implementing a Heisenberg's Voting Principle... for any given submission you can either click a link to see the votes OR participate in voting, but not both.
marknutter 5 days ago 0 replies      
Without the comment scores, the voting is essentially worthless. It's just a cacophony of opinions I haven't got the time to sift through myself.

There's nothing better than reading a controversial article linked from HN, then going to the comments and reading the top voted comments that debunk it. I can never be sure where those comments are now that the votes are gone.

Bring them back.

peterwwillis 5 days ago 2 replies      
I'd prefer it if 'voting' went away entirely. I think you should be able to flag a comment as being hateful or abusive but that's it. The point of this site should be to read interesting articles, not focusing on the interaction of its commenting users.

In fact, i'd like it if you stopped displaying usernames too. Perhaps then people would actually consider all the comments instead of taking what one user says as gospel.

atlei 5 days ago 1 reply      
How about two options:

- Agree/Disagree buttons/links

- High/Low quality buttons/links

This way it is easier to separate if you upvote because you agree or because you think it is a high-quality contribution (even if you maybe disagree). Karma (if kept) should probably just be based on quality score, not on agree/disagree ?

In addition, showing a score in percent/color relative to the other scores in the same discussion ?

Edit: You should probably also separate between "article/submission" karma and comment karma, as some submissions of old articles get lots of karma, while the relative karma of quality comments often are much lower ?

Yes, it is very important that quality articles are suggested, but this is much easier than contributing quality comments (which is time-consuming, especially for non-native english speakers)

radu_floricica 5 days ago 0 replies      
I wonder how this poll looks weighted with by the voter's karma (even logarithmically).
pasbesoin 5 days ago 0 replies      
I'm not sure: While the points did ease scanning for "high value" comments (particularly as participation and the volume of comments has increased), the site feels more "peaceful" without them. Perhaps I even notice a bit more, without point scores to narrow my attention.

OT: I've continued to wonder whether eliminating karma credit for story postings would be a benefit. On the one hand, contributors of interesting topics should be valued. On the other, the "core" of HN has always valued information more than karma. And without a "karma carrot" for stories, perhaps the repeats and marginal/off-topic posts (many of latter nonetheless more recently managing to garner 3 or 5 votes) would decline. I don't care about the 3 or 5 points; I care about wading past the entries. Or finding comments divided between several postings/threads.

But I'd hate to suggest removing a significant aspect of the site's performance, if I'm misreading motivations or missing a significant secondary effect.

bambax 5 days ago 0 replies      
Arguments are possible with precise comments scores; why not display a range instead... or a percentile (calculated for each page and not the whole of HN of course)?

There could be any number of percentiles but the fewer the better: what we want to know is where the high scores are.

So maybe just signaling the top ten comments on each page would be enough...

thorax 5 days ago 0 replies      
This could just be a user option. Turn the scores off by default but allow people to turn them on if they want. You could then easily query how many people had them turned-on and it might give further data points to later query with regards to how it impacts the site.
tomlin 4 days ago 0 replies      
It's a bit of a quirky conundrum. If you retain the scores, you encourage the nerd's competitive nature; the desire to 1-up another in conversation, to be the most right with passionate detail. Strangely, this type of nerd tends to get it right. If your game is to be better at knowing then you'd be sure you always had the most logical, concise rebuttal or point of view that you can possibly have. You can take just about any sport and find an analogy to pair well here.

It's my guess that the nerds who prefer the scores hidden aren't as competitive as the others, in which case, see the scores as a distraction of truth rather than an indicator of it.

Personally, I think the scores are a distraction, but I can't overlook that they do seem to encourage (in a backwards way, it seems) a much more pointed and refined point of view.

magicseth 4 days ago 0 replies      
Perhaps not showing you your own comment scores would help. Only logged in people could see scores, and you don't see your own. Sure you could sign in as another account, but if you are willing to do that, then there's not much we can do from stopping you being a troll.
yakto 6 days ago 1 reply      
What is mixpanel telling you? I'd think you could make this call from your engagement metrics.
grandalf 5 days ago 0 replies      
I think the poll should frame the questions slightly differently:

a) When I see that someone has written a stupid comment, I enjoy writing a scathing retor that is likely to be upvoted by the mob.

b) When I see that someone has written a stupid comment, I occasionally choose to write a polite response, and sometimes I ignore it but hope that someone else will write a polite response.

erikb 4 days ago 0 replies      

over the years we all discussed a lot of times about the flaws we find using this kind of naive evaluation system. Now to get rid of some of the problems we obv. try to optimise the flawed model or the methods we apply to it. But didn't we learn here on HN that if your model is flawed in a way you can't accept, you should find another model?

Just an idea: Maybe a recommendation system would perform better then a voting system. Because we want to sort our posts and comments according to relevance to the reader. And of course there are articles, which are better then others in an object, global fashion. But most of the time 2 articles or comments of the same comment can have quite a big difference in relevance to 2 different people.

That is just my simple, unvalidated idea. It maybe totally wrong. I just use it as an argument to the idea that we should not discuss about things like "show comments yes/no" but "how do we better show a reader the stuff that is relevant to him?"

stock_toaster 5 days ago 1 reply      
I think a lack of displayed comment score reduces 'group vote' mentality, and makes the decision to upvote less social and more individual.
bambax 5 days ago 0 replies      
Would it not be possible to have comment points in the page, but not visible, and then have them shown as a personal setting (or maybe only with the help of an extension / greasemonkey script)?

This way only people who really want them would have them.

mgkimsal 5 days ago 0 replies      
Why not introduce more types of labelling of comments? up/down doesn't often give me enough of a spectrum, and usually there's not enough time to write something substantive in a reply. I may just be repeating someone else's comment (do I 'upvote' that one instead)?

Being able to tag/label a comments as well as vote - with the size/scope/quality of people here - would create a much richer opportunity to be able to sift/sort/filter the comments as desired.

gleb 5 days ago 0 replies      
I think improvement in civility that no-comments brought is a local maximum. I hope so anyway.

It's worth a try to experiment with other options. Like charging karma for commenting.

And maybe try these as A/B tests. Half the topics run with old rules, half with new ones.

Qz 5 days ago 0 replies      
Not sure if that has been suggested before, but how about showing the comment score only after the user has voted? This keeps most of the benefits of no-score mode (less voting because of bandwagon-y stuff), while making it easier to see which comments are good. Basically, you have to make up your own mind first, but then you can see what the general consensus is.

(And yes I realize it wouldn't be hard at all to make a mashup that just straight up shows the comment scores, but I doubt most people would feel the need to use that).

gommm 5 days ago 0 replies      
I think it would be good to keep hiding comment scores for new topics and then after a few hours show them...

I like sometimes revisiting old topics and in those cases, having the comment scores helps a lot for skimming and finding information. On the other hand, I agree with you that hiding them stops some arguments and makes for a better discussion.

So, why not have the best of both world by only showing the scores once the topic is a bit older and the discussion has tapered off?

reitzensteinm 5 days ago 0 replies      
What about this, for a 90% solution: comment scores are visible to everyone, except people that have commented on the page.

Skimmers can get useful information, but those involved in the discussion can't easily get obsessed with points, they'd have to ask a friend, or maintain multiple accounts (maybe have a karma threshold?).

Scores would have to be invisible to those logged out as well, and HN Scrapers would have to agree to delay publishing scores 24 hours.

Volscio 5 days ago 0 replies      
# of upvotes is just a proxy for what people really are looking for. Sometimes I just want to sort by Usefulness, Funniness, Truthiness, Creativeness. Those are the qualities I'm looking for in comments.
jamesbkel 5 days ago 0 replies      
There are already so many comments on this post that I hope this won't get lost in noise nor add to the noise. I did a quick search through the page and I don't think this idea has been addressed... apologies if it has.

I like not having having the score displayed, but I also like the option of seeing the score. My suggestion would be to keep scores off the main page but perhaps add an option to "click and see the score".

Not sure what the perfect solution would be, but to sum it up: scores should be accessible, but with traction.

latch 6 days ago 0 replies      
Can't we come up with a way to highlight "good" comments? Some indicator that shows that a comment (possibly nested, which is where sorting fails) has X% of the total votes?
nextparadigms 5 days ago 0 replies      
I'd like to be able to sort comments by Newest on a post. If I go through 30-40 comments at first, and then later come back to see another 10 new ones on that particular post, it's hard to remember which are the new ones and which aren't, and it just makes me go through all of them again.

I'm not really sure what "Comments" menu serves, because I see it's all the new comments from all posts mixed together. That doesn't make much sense to me. I want to see the new ones on each post.

bane 5 days ago 0 replies      
Perhaps don't show them when a story is moving up, but perhaps start to show them as a story descends/moves off the main page.

I personally haven't had any problem finding quality comments on the site. What is has made me do is be more selective about what topics I bother to read the comments in.

bajsejohannes 5 days ago 0 replies      
pg: Have you analyzed voting patters before and after the change? I would love to see a histogram of number of votes over percentage of comments in the two periods.
nhebb 5 days ago 0 replies      
Instead of the score I would prefer to see a rating. When a score passes a certain threshold, it gets a star. The low effort way to implement this would be to replace the digits with an 'X'. That way, people could see the relative merit of the comment without quibbling over numbers.
DanielRibeiro 3 days ago 0 replies      
To prevent mob egging, a Stack Overflow approach might be helpful: make voting down to cost karma.

Unfortunately this works better on Stack Overflow as upvotes give you either 5 (on submissions) or 10 (on comments) karma while voting down only costs you 1 karma.

jeremymcanally 5 days ago 0 replies      
Why not make it a user setting, defaulting to off?
thought_alarm 5 days ago 0 replies      
These polls aren't very scientific (much like comment scores) but at least we'll be able to gauge the opinion of the Saturday night HN crowd.
J3L2404 5 days ago 0 replies      
On technical issues it is vital to see which approach the best and brightest (HN users) are in majority. Not everyone has time to write up a definitive comment and as someone learning many techniques concurrently I need to separate the wheat from the chaffe as efficiently as possible. Paul is undeniably brilliant, as shown by the Distrupt video in which he probably made several lucky startups quite a bit better by shear acumen. But he says in this post that the voting on this was roughly even. Hmmm. Not quite. He knows what he wants to do and can (and has, I don't believe we are going back), but he would like the community to back him up on this but it's not happening. Last time was 57% in favor of submission comments and this time 60%. I plead, beg and grovel please let me see comment scores. I will be here regardless, and thank you for this tremendous resource I was lucky enough to stumble upon.
amanuel 4 days ago 0 replies      
As Edward Tufte would probably put it, I feel some fidelity to the comments has been lost without the scores. The score for me added a 'feel of the community' for each of the comments left and that only mattered IFF the comment actually was worth reading.

As to the disturbing situation you mention I'd look to finding a way to make such behavior more shameful.

scott_s 5 days ago 0 replies      
I think it's better without showing scores. But I'd scores to be visible after the threads have been retired - after we can no longer comment. It makes sense to me for comment scores to be a part of the archive for a thread.
adw 5 days ago 0 replies      
It's a social problem, not a technical one, so it needs a social solution. (That probably means assertive moderation, thread killing and user-banning.)

I don't see how you can align the game mechanics of a scoring system with the kind of tone and content HN's looking for. We're geeks, we'll work out how to game it.

sosuke 5 days ago 0 replies      
Don't forget the days and times you ask. This is a holiday weekend and while I may be lurking here over the weekend to vote for comment scores staying hidden you won't see the rest of the work visitors.
brudgers 5 days ago 0 replies      
To the extent that improving the quality of comments is the purpose of having comment scores, the current system performs pretty well since the scores of one's own comments are always visible.
ot 5 days ago 0 replies      
Kind of related, is it possible to restore the "by <username>" in the comment header? It made searching for comments by specific users much easier.
raghus 5 days ago 0 replies      
How about displaying comment scores only to users with a certain karma threshold and/or have been with HN a certain number of days or weeks?

Presumably the folks engaging in nasty arguments are relatively new folks who haven't added much value to HN?

clarkevans 5 days ago 1 reply      
I'm curious about the relationship to score and down-voting. He's a hypothesis: with actual scores shown, down-voting was less severe.

From my observations, this community tends to not like jokes, even if they are slightly clever. For new users, a getting solid -5 on a first comment is a pretty solid discouragement. In particular, I'm wondering if there was previously a community consensus that -1 or -2 was enough of a signal?

I think if you're going to signal anything, perhaps it's who new/infrequent posters are?

jefffoster 5 days ago 0 replies      
A very simple change would be to only display comment scores once you have voted on a comment yourself. I'd hope this would encourage more voting and dissuade group-think.
vecter 5 days ago 0 replies      
I like the idea that someone posted a while back: have a link to display scores. Once you do though, you can't upvote or downvote.
agraddy 5 days ago 0 replies      
Maybe one solution to the problem of comment wars would be to regularly display comment scores but every so often turn them off for a bit of time (a week or a month) to remind the community of the goals of the site. My guess is that if users are reminded every so often to keep things civil by removing the socres; it would probably go a long way in maintaining the community without losing the benefits that seeing the scores provides.
chopsueyar 5 days ago 0 replies      
Why can't I use my karma points to buy 'sneak peaks' of comment scores?

Maybe 5 points to display comment scores for a particular story?

J3L2404 5 days ago 0 replies      
How about letting users with > 5000 points see comment scores? Surely that segment can decide for themselves. Having devoted that much time to the site I think they deserve it.

EDIT: Yes, that number is biased to include me and not really the fairest way to go about it. I haven't witnessed the 'decline' maybe because I am only interested in a few topics, but the usefulness of HN has definitely declined. Could some moderation duties be traded for comment scores?

ThePinion 6 days ago 0 replies      
Not sure how reception to this idea will be, but with the whole "where people would literally try to score points off one another" thing you could just revoke someone's ability to see points if they're obviously/blatantly doing this. This doesn't solve all the problems obviously, but at least you're not taking the points away from the people that actually do find them useful. Some shouldn't spoil it for everyone!
wglb 5 days ago 0 replies      
I am preferring it with the comment scores hidden. I am more likely to judge based on content than if i see the scores.
wicknicks 6 days ago 0 replies      
Here are a couple of suggestions:

1. Use labels to indicate score comments. Rather than using precise numbers, labels like (low, medium, high). That way we can still see which are the interesting comments, but not see the exact score (so people stop nitpicking on 1-2 random points).

2. Remove the name of the person who posted the comment. Keeping it completely anonymous might help your mob egging concern. Also, that way the comment gets more attention than the person who made it.

jules 5 days ago 0 replies      
Here's an idea that has some of the benefits of both: don't display the comment score immediately; only display it after some time.
rms 4 days ago 0 replies      
How about $10/month to view comment scores?
beagle3 5 days ago 1 reply      
Reading through the comments, many people here claim "Without the scores, I can't tell whether the comments are worth reading". People, you are aware, I hope, that the comments are sorted by points in the display? The highest rated comment is shown first. And scores below 1 fade to gray (I think -3 is still readable without copy & paste, and -4 is the last displayed).

Could someone explain how the specific number (rather than just sort order) actually influences their reading of discussions? I'm trying, and failing, to see how that would actually be useful. Honest question here.

wging 5 days ago 1 reply      
>Lots of people have complained that without comment scores it's harder to pick out the good comments. Some say that's better, because now you have to judge a comment for itself.

I say it's worse, because now I have to judge a comment for itself.

On subjects I don't know, this is of course difficult.

mcn 5 days ago 0 replies      
I've noticed an increase in posts reiterating each other: it seems that the wrong kind of post is being used as a substitute good for the wrong kind of upvote. For that reason I would like to see voting made public again.
rodion_89 5 days ago 0 replies      
Can this just be an option in our personal settings so that everyone can either have the scores shown or not based on their preference?
watmough 5 days ago 0 replies      
I support keeping the scores hidden, but using the scores to hide inappropriate, or spammy comments.

Of course, what seem to be missing is a way to highlight comments that have been consistently voted up within the thread.

lelele 5 days ago 0 replies      
Instead of just scores, I'd like to see both how many upvotes and how many downvotes a comment has got. I second the suggestion to colorize the most upvoted or controversial comments.

However I think the issue with simple scores is that they cram together agreeability and usefulness of a comment, which should be kept separate. That way, users which just agree with an opinion would be able not to influence the usefulness score of a comment.

Yhippa 5 days ago 1 reply      
I don't know what it is about HN but for me this is now my "go-to" news site about computing. It used to be Slashdot, then Digg, then Reddit but now I find myself refreshing the HN an embarassingly frequent amount of times.

The thing that I'm quite honestly shocked at is the lack of trolling of flamewars on here. That is a huge turnoff for these types of websites for me. For example I used to love Fark until the comments went from funny to constant flamewars.

So based on that I don't think comment scores will make much of a difference for me. It's the kind of people this site draws and the comments they make. I will say that if showing comments tends to encourage flamewars by all means do not bring it back!

Daniel14 5 days ago 1 reply      
Since I'm new here, I'd like to ask a question that's been bothering me for some time: Do we upvote comments we agree with, or comments we think are worth reading? In my case, those aren't always the same, so I'd like to know how you guys handle it. Thx.
danssig 4 days ago 0 replies      
I said I liked it as it is now, but I would actually prefer a hybrid: Show the score but only -4 - +4. That is, make positive counts behave how negative used to.
Evgeny 5 days ago 0 replies      
Lots of people have complained that without comment scores it's harder to pick out the good comments.

That's a key thing for me. I believe I can distinguish between the cases when a popular person gets upvotes because she already has 999999 karma on HN and when the comment is genuinely useful because, well, it is. MYSELF. Please let ME make the desicion. Thank you.

(I stopped voting anything at all on HN because the scores are no longer displayed).

BJakopovic 6 days ago 0 replies      
if visual scores create problems, why not just obscure them, and use an algorithm to find a better relative popularity between the comments; and show a simple 3 level scale (spam/normal/popular).

If you still need more depth, add a fourth level to the scale.

YuriNiyazov 6 days ago 0 replies      
Why not give people the option to display comments, with the default hidden, and then redo this poll?
overgard 5 days ago 0 replies      
While I like being able to pick out things quickly, I find that displaying comment scores tends to encourage a herd mentality. It's subjective, but ever since the comment scores have been disabled it seems like there are a lot more diverse viewpoints being expressed.

Also I might just be weird in this regard, but I find being "scored" on the things you say to be sort of creepy. I like feedback, and that's difficult to convey well on the internet, but turning ideas into popularity contests seems perverse to me.

meatsock 5 days ago 0 replies      
i would like to see the average comment score for yes and no votes compared when this is all over.
antihero 5 days ago 0 replies      
I find comment scores instantly bias you towards or against a post due to crowd mentality. It happens on reddit even worse, so I'm not sure they're such a great idea.
Semiapies 5 days ago 0 replies      
A thought: Make comment scores the square root or log or whatever of the total number of upvotes. For good measure, cap it at +10 or some such.

That might not remove all competitiveness, but it might reduce the effect of the vote-up-everything-popular-guy-says crowd.

davetufts 5 days ago 0 replies      
As a casual visitor, I really preferred seeing scores. Yes, I can read through every comment and judge for myself what's best (note: this is still possible even when scores were shown). But I prefer scanning the comments and just reading what the crowd liked most.
bchjam 5 days ago 0 replies      
Showing scores made reading through comments feel more like a game, which was interesting but tended to take my attention away from the context of the original link. I was also less interested in interacting, largely because the context was deemphasized and it seemed too easy to devolve into a mentality of competing for upvotes.

That said, I do like the idea of using something like color coding (on username?) to highlight the top few comments.

blogytalky 5 days ago 0 replies      
Hi. I'd like to make a suggestion. May be you should try to enable scores for top level comment in the thread and disable them for other comments inside the thread. This would:

1. Highlight what you call 'good' comments.

2. Prevent fighting in threads for better scores.

3. Force commenters only to write replies in their replies and write a separate comment if they have more to say on the topic.

4. Finally, threads with highly scored top comment are probably 'better' in the same sense.

veyron 6 days ago 0 replies      
As a sort of middle-ground, does it make sense to give users the choice of showing comment scores?

You would be correct in arguing that it does not resolve the sort of argument as describe in the second paragraph.

However, with regards to your third paragraph, in this situation each person can decide whether or not to see the scores

squasher 4 days ago 0 replies      
AskMetafilter solves this problem by asking all users to pay $5 to make an account to vote or post. This low, easy bar is still high enough to dramatically reduce low-quality comments and questions.

But I wonder if the argumentative and negative postings on this site are coming from relative beginners on the site, or people who are more established?

stewbrew 5 days ago 0 replies      
It took me a few days to appreciate the change but I think hiding scores and consequently taming kharma whoring is an improvement.
profitbaron 5 days ago 0 replies      
I think comment scores should return. However, I also think they should be shown like the "down vote" option is when a user reaches a certain level of karma. I think this would be a way to combat the issue of the karma point fiasco from last time.
HedgeMage 5 days ago 0 replies      
Can we have an option for "I don't care as long as I stop seeing posts about the (un)desirability of visible comment scores" ?
bilban 5 days ago 0 replies      
How about just timing out threads, making them immutable after a predefined time limit? Once immutable display the final scores.

Though I realise this won't help in the here and now.

suking 5 days ago 0 replies      
Can you add it where you can close & open (condense) comments? That would make the site much more user friendly. Is there a greasemonkey script for this?
hackermom 5 days ago 1 reply      
Why do people bring a competitive mindset to a technology news website?
coffeedrinker 5 days ago 1 reply      
I don't know everything about everything.

I rely on the community to give insight on those other things.

Scoring helps me do that.

Without it I cannot tell if a comment is rated highly, or merely appears that way.

Scoring adds a ton of value to the comments.

richardw 5 days ago 1 reply      
I know it's democratic and all, but I'm not sure a simple poll is likely to produce the best result. What if you asked whether people would like inline images, emoticons and HTML messages, and 80% said yes?
sigzero 5 days ago 0 replies      
There is no "I don't care either way" option.
Devilboy 5 days ago 0 replies      
If you do change it back please consider capping the score at +5 or +10 or something.
oyving 5 days ago 0 replies      
After you made the change to the site, have you seen any changes in voting patterns and point distribution? It would be interesting to see that kind of data.
nazgulnarsil 5 days ago 0 replies      
I'd like to only see points given/subtracted by users with high karma. Elitism FTW.
ozataman 5 days ago 0 replies      
Why not offer both versions and let users pick the one they like in their profile?
pathik 5 days ago 0 replies      
I guess we have a winner. Now please bring the comment scores back.
walexander 5 days ago 0 replies      
I made an acount just to vote on this. The arguing you referred to is exactly the reason why i prefer hn to reddit. I would hate to see rhe same insular 'hivemind' effect here.
nabaraj 5 days ago 0 replies      
ignifero 5 days ago 0 replies      
Comments are data points and there's no reason to hide them.
ms4720 5 days ago 1 reply      
where is "I will complain about an change"?

It is the third choice

ManOwl 6 days ago 2 replies      
I think this poll is proof in itself that comment scores are valuable.
Skype protocol reverse engineered, source available skype-open-source.blogspot.com
432 points by jbk  1 day ago   130 comments top 20
jcr 1 day ago  replies      
I wouldn't venture to say this doesn't belong on HN since it really is
interesting (if it was actually done correctly), but the files available
for download are most likely illegal, were most likely created with
pirated tools (IDA Pro/Hex-Rays, and yes, as a customer of theirs for over a
dozen years I've reported it), and of course, the usual vilification of
reverse engineering.

If you're reading this on a desktop or laptop system (rather than a
phone), then you are most likely using an "IBM PC Compatible" even if
you're using an Intel based Apple, and hence, you're using the fruits of
completely legal reverse engineering.

The way to do reverse engineering legally is to have one team reverse
engineer the target and completely document how it works. Once it's
documented, another disconnected team writes a new implementation from
the documentation. This process is how you're using an IBM PC
Compatible today, so yes, reverse engineering for compatibility is
perfectly legal.

If there is a patented algorithm required, it's not a sure thing. There
are most likely compatible ways around the patent, but there's also the
fact that the patent is only valid in the US. With open source hosted in
some other country, who are you going to sue? The users in the US?
--Nope, users are the ones paying for skype.

You might say, "But we forbid reverse engineering in our license!!!"

Contract clauses forbidding reverse engineering are invalid in many
countries and jurisdictions, and of course, you also have to prove the
other party agreed to the contract/license. With this said, it's very
easy to create a international jurisdictional nightmare to render any
such contract clause tactically impossible to enforce.

The easiest way to think about this is security research. The folks
finding and reporting exploitable flaws in software are obviously
reverse engineering it. Occasionally companies have tried to legally go
after people who have published security research on their products, but
usually this ends very badly for the company. Additionally, doing
security research is protected use in some countries and jurisdictions.

In short, competition is good for markets, and competing by studying and
mimicking the competition is both normal and legal.

For the "rights" advocates out there, there are legal problems with the
three file downloads available:

1.) According to the first file name, the original binaries are being
redistributed which may be (and usually is) against the license terms
and default rights granted by copyrights.

2.) The IDA Pro database (most likely) contains the entire target
binary, so you do have (illegal) redistribution of a copyrighted work.
You can load only parts of a target binary into IDA, but that doesn't
matter since it is still a portion of the original work. As for whether
or not said portion could fall under fair use is debatable (i.e.
lawsuit). In general usage, the entire binary is loaded, since without
it, you're limited to static analysis (i.e. no debugging).

3.) Decompilation, and to a lesser degree disassembly, are equivalent to
"machine translation" in the sense of copyright. Creating a translation
is considered creating a "derivative work" and unless you have been
given rights to create derivative works, then you're in trouble. One of
the comments here on HN claims the "source code" file is the output of
the Hex-Rays Decompiler.

I've never used skype and I've never read their license so I don't know
if they specifically allow redistribution.

I have no love for skype or microsoft, but if this had been done
CORRECTLY by releasing written documentation so an entirely new
implementation could be written, then I'd have no problem with it.
There are right ways and wrong ways to legally create compatible (open
source) software through reverse engineering, and this is a perfect
example of the wrong way.

michael_dorfman 1 day ago 3 replies      
My aim is to make skype open source.

Isn't that decision up to the folks who own the rights to the original code?

Personally, I'd be hesitant to get into a project that has more need for lawyers than coders.

Animus7 1 day ago 7 replies      
If the goal is to open up Skype, this isn't the way.

Even if some insane insomniac de-twiddles the pages upon pages of optimized indirection in this code (which I seriously doubt), all Skype has to do is tweak the protocol or encryption and the researcher is back to square one. It's a losing battle. And that's not even getting into the legality of it it all.

How about instead of trying to fruitlessly crack Skype, we spend the time making something that's both open and better?

senko 1 day ago 0 replies      
Relatedly: there was a presentation about (reversing) Skype internals on BlackHat Europe 2006 (warning, PDF): http://www.blackhat.com/presentations/bh-europe-06/bh-eu-06-...

Skype probably updated their client/protocol since, but still an interesting read.

tttp 1 day ago 2 replies      
Two major issues:
|| Copyright (c) 2004-2009 by VEST Corporation.
| All rights reserved. Strictly Confidential!
The project is using existing code, and "All rights reserved" is not an approved OSI license I believe ;)

And things like compression algorithm are patented, and that's very likely skype is using some of it. Reminds me of a project by Intel of providing an implementation of g729 (a voice codec). The source was available, but it was "non commercial usage only" because of the patents mostly.

An interesting project, but doubt we'll see any usable implementation anytime soon IMO. And even if it does, skype will probably alter a bit the protocol to make it fail if it reaches a critical mass.

iwwr 1 day ago 1 reply      
Better start uploading this to torrents if its legit. Skype will issue takedowns any moment.
nikcub 17 hours ago 0 replies      
two things:

1. I would love it if MSFT, as the new owner, gave up on the cat-and-mouse of security through obscurity and obfuscation and settled on a published and peer-reviewed protocol. I am sick of the memory footprint and cpu spikes in having to run skype clients because 70% of its resources are dedicated to hiding what is really going on. I would love a nice, clean, light version

2. we can well assume that if this is happening in the public domain then it was probably done a few years ago behind closed doors at the NSA et al

dkalmnekfaiksmf 1 day ago 0 replies      
VMG 1 day ago 1 reply      
How easy is it for skype to change the protocol?
tibbon 1 day ago 0 replies      
Pastebin of the unpack-4142.c : http://pastebin.com/AY3abgEJ
Joeboy 1 day ago 2 replies      
From what I understand, the big unsolved problem open VOIP options have isn't the voice bit, it's negotiating a connection through NAT/firewalls/dodgy routers etc. I doubt this helps with those problems.
nextparadigms 1 day ago 0 replies      
I haven't tried Skype since 4.0, and I just tried it again now. When did it become such bloatware? I don't think I'll ever want to use it again if it stays this way. The new interface looks pretty confusing, too.
yread 1 day ago 1 reply      
with identifiers like unpack_41_715680 and the_thing it is not exactly easy to read
foxhill 1 day ago 1 reply      
aside from the awesome technical exersize in hacking, i don't see this as any net benefit for VOIP.

the time would have been much better spent working on the GNU VOIP client, not only would improvements have been usable without legal issues, they would be there in an (ostensibly, perhaps) understandable format - working code.

dennis714 1 day ago 1 reply      
It is looking like result of Hex-Rays decompiler.
evanwolf 1 day ago 0 replies      
Has anyone actually read the revealed code? Aside from the commented copyright text at the top, can you explain what the code does well enough to document Skype protocols? Is this in any way useful if you want to talk to Skype servers or clients?
braindead_in 1 day ago 2 replies      
It seems that the encryption algorithm has been reverse engineered. I guess you'll still need the keys to decrypt the voice data using this algorithm, assuming it works. It's a big deal if has been done, because a lot of people have been trying to crack it. Some governments are going to love this. The Skype client itself has a lot of obfuscation to prevent something like this.
gcb 1 day ago 1 reply      
So microsoft really bought skype and is already playing the bait-and-switch game?!
jduran 22 hours ago 1 reply      
this is just shady
ezioamf 1 day ago 0 replies      
Skype superiority will finish when IPV6 be fully adopted. When this will happen? Never?
Loren wants to work at Airbnb lorenburton.com
408 points by guynamedloren  4 days ago   104 comments top 33
pg 4 days ago 4 replies      
I just emailed the founders. I expect they'll contact you soon.
M1573RMU74710N 3 days ago 10 replies      
Hey, I hate to be "that guy"...but since you're applying for a front-end job, here goes:

- You're loading scripts in your head tag, and a crap-ton of them. At least some of these could be concat'd together. HTTP request carry a lot of overhead, so fewer is better. I bet at least some could be moved to the bottom as well, with some CSS/JS to avoid the dreaded Flash Of UnBehavioured Content.

- In a similar vein, you're loading a lot of images, more overhead again. At least some of these (like the clouds) could probably be combined into one file and then separated out with CSS.

- You're not setting long expire headers on your resources, bad for caching.

- You don't seem to be serving up gzipped resources.

- Some of the scripts aren't even minified.

Take clouds.js for example, you use a whole HTTP request on this one file, and it's small. It's also not very DRY, you have a bunch of repeats of basically the same code, it probably could have been re-written to something like:

/* ... */

I realize some of this stuff is not entirely necessarily on a site like this, but considering how easy it is and the fact that you're applying for a front-end engineering job, I'd think you'd want to follow best practices.

Overall it's a cool idea, very fun and nice looking. Good job.

(BTW, I'm available for work. If anyone from Airbnb is reading this, I can haz Job plz? )

tptacek 4 days ago 3 replies      
See, now, this is great stuff. How do you not hire this guy? A near-perfect example of the form:

* It's gorgeous

* It's simple

* It's precisely targeted and relevant

My only critique is that he didn't find a way to work Pocket into it; the story about his six-year-old sister using it is a winner.

Anyone want to put odds on how long Loren stays on the market? I give it a week. :)

guynamedloren 4 days ago 1 reply      
Inspired and motivated by some fantastic advice from the HN community, I've decided to pursue a frontend engineering position at Airbnb. I'm an entrepreneur at heart, but I love what they're doing and could see myself working as a part of the team. In fact, Airbnb is one of the only companies I've really felt the urge to work with.

In addition to submitting a resume and cover letter, I thought it would be fun to put together this little project. I'm not personally sending a link to Airbnb, so any chance they have at seeing this is entirely up to the community. I guess this is an experiment of sorts - we'll see how it goes. Thanks HN!

nsfmc 4 days ago 3 replies      
So, while this is novel and all, it seems to me that these "resume 2.0" applications indicate a real failure of systems like "jobvite" or whatever backend air bnb is using to simplify their hiring decisions.

On HN we've been seeing a bunch of these hyper-targeted reverse resumes in the past year because I suspect people have gotten tired of submitting waves of resumes and hearing nothing back within a reasonable timeframe from companies that advertise open positions on their websites but never, for whatever reason, respond.

What i'm seeing here, is that traditional recruiting practices are failing in some way because priority is achieved by a) making it to the hn frontpage which b) gets pg to vouch for you. Which is basically no better (or efficient) than the traditional method of hiring folks that you've networked with via friends/school/etc or what have you.

I understand that historically those online resume systems are a real losing proposition for applicants, but this suggests that they're more like a black hole rather than a mail slot.

Again, this is a critique of these automated systems which seem to be visibly failing because people are going out of their way to subvert them in order to achieve any success.

_harry 4 days ago 0 replies      
That's awesome Loren!

There are only four of us on the frontend team right now. We need more!

We like engineers with passion. People who get excited talking about the new frontend development frontier. Conversations like this happen all the time:

- Think Node.js would be cool for a realtime dashboard? Yes, do it.

- Maybe we could try using backbone.js and a fat client for this feature? Fork it and let's go.

- What's SASS? Install the gem. We're using it.

Our fourth frontend engineer just started last week. We sat him down, showed him how we were doing things and we asked him, "What do you think?" and he says, "Have you guys heard of Jammit? It's used for asset packaging. It's pretty cool" -- A couple hours later he was showing us the page load speed increase and how to use it.

If you're a frontend engineer wondering what it's like to work at Airbnb, feel free to email me: harry@airbnb.com or if you prefer character limits I'm on Twitter: @hshoff.

patio11 4 days ago 1 reply      
The ROI on these in terms of career growth absolutely ROFLstomps effort at improving a traditional resume and playing spray-and-pray. (The ROI of networking is probably better than both but, hey, what can you do.)
kmfrk 4 days ago 0 replies      
Nota Bene: Loren also made the Snowpocalypse T-shirt site, which garnered a lot of HN attention and praise: http://news.ycombinator.com/item?id=2173155.
rkalla 4 days ago 0 replies      
This is an excellent example of what I'll call "Resume 2.0" or "How to get a job in our new economy".

It's always boiled down to an employer knowing if you can handle yourself and get the job done. The only facilities we had available to try and assess that in the past were those stump-the-chump questions and interviews at a desk sitting across from a committee.

I love what technology has enabled us to do. It has enabled everyone to be a self-starter and enabled us to brandish our accomplishments in really damn creative ways (e.g. this Loren fellow rocking the AirBNB application).

HN is a bit of an echo-chamber for these types of cutting edge job acquisition moves, I don't expect most people to try and get a job this way, but it is the beginning of a trend that I think will make everyone happier as a result.

Loren, best of luck to you. You certainly got an upvote for me and it sure looks like ABNB would be lucky to have you.

blehn 4 days ago 1 reply      
This is one of the better "hire me" micro-sites I've seen... and it still made me cringe.
csomar 4 days ago 1 reply      
Loren, that's a very nice resume page. But I think it's missing something, which is quite important: Your portfolio.

Collect some of your best work, and create a slider in one of your clouds. If you have also worked in some companies in the past mention it. Stackoverflow, Github, Dribble... links are also useful.

hint: Change the white up arrow color, so that it contrast well with the background. You can detect this with scrolling and assign colors. This will show that you put lot of attention to small details ;)

kenjackson 4 days ago 3 replies      
Nice. And renders on a few browsers I tried. I'm always surprised with how many people consider themselves professional web devs, but can't bother to get IE to work properly.
pclark 4 days ago 2 replies      
I'll bite. Is this actually that good? Spending a lot of time on it, sure, but other than it being quirky (which is hugely valuable) is it doing it's job? No specific accomplishments, core skills, and no software in the wild for us to see.

You could have linked to your Airbnb profile ("hey i use it!") linked to your google map ("hey i travel!") and then listed all your software projects and links and things about each that you are especially proud of, and that'd have been awesome.

jmtame 4 days ago 1 reply      
I tried getting a meeting with him several times back when I was still a student at UIUC, I read how he launched a t-shirt web site and made some money from it, and it seems every response I got was: "sorry man, I didn't get any sleep last night, I was busy working on something. Can we try again next week?" Seems like a hustler.
cemregr 3 days ago 0 replies      
I was interested in working for Airbnb too. Last January I tweeted out about how much I love them, and through the tweet I ended up visiting their offices and chatting with Joe, one of their founders. They were very generous with their time.

Took me 10 seconds writing out the tweet, but the face time I got out of it was very valuable. This is another thing one might look into to get the attention of companies.



sixtypoundhound 4 days ago 0 replies      

Not going to critique the technical quality of the work (particularly since I'm not a front-end web guy), but as a hiring manager, I'd be frigging floored if someone spent an evening (maybe more? Although I suspect you're pretty quick by now, based on your comments) putting together a tight on-point microsite for my consumption.

The big job boards are a joke since they have reduced the entire process to buzzword bingo - and a fairly poor form of one at that. There's basically zero insight into how good someone actually is. Looking at someone's blog or contributions to open source projects is potentially an indicator, but there are a number of solid folks for whom that isn't a serious options (significant risk of their old economy management freaking out - a problem when you have a couple of dependents you need to keep fed, housed, and healthy).

So you get interviews like the one I conducted two weeks ago - "you have written free form SQL before?"...."yes, many times"... " Excellent, I have a little problem for us to solve..."

[walks to board, draws three data tables - one transaction file and two dimensions - and outlines a requirement for a basic report that involves joining, summing, and filtering data - I validated it with a recent hire, took him 3 min]

"uh"... twenty minutes later we bring the painful exercise to a close.... "I didn't think you would have me write code"

Huh? Huh? The job description explictly mentions that you need to know free form SQL COLD and asks that you have higher level programming experience (real languages). Do you seriously expect to get a role like that w/o coding?

Shoot - beats the cracker jack wanna-be pricing analyst we had apply off monster who explicitly mentioned his pricing experience...at Walmart... as a positive...

The one decent player in the online recruiting space that I've seen is linkedin - most of the contacts I've had off of that which show serious intent (eg. tailored message) are pretty promising. Especially if they are a connection of someone you trust.

Linkedin's biggest problem is that all participants get the same view of a candidate, including your corporate HR people (don't laugh - I was hauled in and yelled at about this a couple of years ago) - if there was a way to make more of this platform private, the value of the network would increase.

sucuri2 4 days ago 0 replies      
Awesome way to get their attention. Hope you get hired :)
Mz 4 days ago 0 replies      
Count me as jealous, for a variety of reasons. I mean that in the nicest way possible.

Best of luck!

brianbreslin 3 days ago 0 replies      
what i love about these types of sites is that the individuals applying realize they have to step out in front of the pack to distinguish themselves, and are showing true initiative. Far cry from the monster.com droves which send HR departments generic cover letters + resumes.
SeoxyS 3 days ago 0 replies      
I think the strongest resume / cover letter is when you're able to say "Google me" and have it be enough.
Omnipresent 3 days ago 0 replies      
this is what makes me visit HN every day. This will give me inspiration for the entire week. Hell of a resume loren, wish I had creativity like you and knew css better :)

How long did it take you to put this effort together and did you send your "formal" resume to airbnb before coming up with this idea?

shahedkhan30 4 days ago 0 replies      
Hey Loren, love your website! Great work you go there, and such an ambition!

I would definitely hire you for my start-up company (Viatask).

arctangent 4 days ago 0 replies      
Fantastic stuff. Best of luck to you!
mbrzuzy 3 days ago 0 replies      

I didn't realize the clouds at the top of the page were moving until after. Nice touch.

venturebros 4 days ago 2 replies      
Gah I hate it when people are good at design and can code!

Just not fair. /sighs

Good luck!

startupcto 4 days ago 0 replies      
Good luck and hope to hear from you if you did get your dream job at AirBnB.
mvs 4 days ago 0 replies      
Very Cool! hope you now believe 4 yrs weren't a total waste after all:-)
bvandusen 3 days ago 0 replies      
Lauren, what is it about airbnb that you love so much? There are a lot of amazing san francisco startups doing incredible things that would probably all love to have you on their team.
JacobIrwin 4 days ago 0 replies      
Good luck Loren, hell of a visual (i.e., resume)! :-)
arc_of_descent 4 days ago 0 replies      
Really nice looking page. Was going through the source and found this -
"I'm a self-starter and <strong>entrepreneur</strong>"
Omnipresent 3 days ago 0 replies      
you should do a detailed blogpost about the design and creative vision behind this!
teddytruong7 3 days ago 0 replies      
love this loren good luck!
ignifero 3 days ago 0 replies      
Talk about modesty...
You don't get shit you don't ask for humbledmba.com
379 points by jaf12duke  3 days ago   40 comments top 18
snprbob86 3 days ago 1 reply      
Great advice! Some people can be a bit shy about asking for things, so I suggest you practice.

For example, I was recently at a reception which included a fixed menu dinner. For the non-native speakers, a fixed menu means that you don't order your meal, they just bring it to you. A waiter was walking around to each of the many tables, quietly filling glasses of red wine. Red wine makes me sleepy, so I asked for a beer.

A few minutes later, someone at my table asked me "How did you get a beer?"

I said "Um, I ordered it..."

"Oh? We can do that?"

"We're at a restaurant, he's a waiter. I asked and he didn't say no. So, ugh, yeah. We can do that."

That tiny unfamiliarity of a fixed menu and an automatically filled wine glass undid a lifetime of training on how to order things at restaurants.

You've got to ask for things, even if it feels unnatural or uncomfortable!

akikuchi 3 days ago 1 reply      
The article's bullet points are great, but I would emphasize one thing:

Don't be afraid to ask, but only AFTER YOU'VE DONE YOUR HOMEWORK. People can be surprisingly helpful, but especially if it's clear you're working hard to make each request count. They're much less likely to think you're wasting their time if it's clear you've been hustling and see their help as a catalyst for you working EVEN HARDER.

pavel_lishin 3 days ago 1 reply      
This reminds me of the "Asker vs Guesser" culture comparison: http://www.guardian.co.uk/lifeandstyle/2010/may/08/change-li...
sosuke 3 days ago 0 replies      
The comments on the article directly quickly go into specifically not abusing VCs, they are very busy etc but I think they are missing out on the larger picture here that I am taking away from this story. No one is looking out for my best interests and if I need something or want something the best opportunity I have to get that is to ask.
rauljara 3 days ago 2 replies      
Excellent post, but I wouldn't have distilled quite the same catch phrase -- "You don't get shit you don't ask for" -- from that anecdote. It's more like, "When someone offers you help, take them up on it."
holdenc 3 days ago 2 replies      
Yes and no. There are some people who make a living asking 100 people a day, and banking on the five people who say yes. Then others are fortunate enough to have customers, investors and interested parties knocking on their door -- simply because they've made something really interesting. These people often get things without asking. And if you are bad at asking -- it's better to be one of those people.
MatthewPhillips 3 days ago 1 reply      
Good read. I had been sitting on an email for a few weeks to a possible business partner because I didn't want to look like a moron that doesn't know how things work (I am said moron). I just pressed send.
ChuckFrank 3 days ago 0 replies      
When leading my sales team, I always asked them to do two things clearly.

1. State clearly why they were there

2. State clearly what they needed.

Jason's note had two statements that did exactly that.

ONE >>Right now, we're preparing for the PhoCusRight conference next week, the travel industry's largest get together. We're trying to make some connections with people at the conference.

TWO >>Can you help us connect with leaders in the Business Travel divisions of Expedia or Orbitz? Or any other introduction that you feel would help us?

Remember without those two simple points, you are wasting yours and other peoples time. And then they wonder (very LA indeed) "And you are wasting my time because?" - which is the last thing anyone should be thinking.

gallerytungsten 3 days ago 2 replies      
Good article, but it's "root" and "rooting" not "rout" and "routing." Somewhat hilarious reversal of meaning via that typo.
warmfuzzykitten 2 days ago 0 replies      
So nobody else is offended by the title? I don't take advice from people who talk dumb shit.
hippich 3 days ago 1 reply      
"You should never ask anyone for anything. Never- and especially from those who are more powerful than yourself."

" Mikhail Bulgakov (The Master and Margarita)

niels 3 days ago 0 replies      
Reminds me of "All you have to do is ask" by James Altucher: http://www.jamesaltucher.com/2011/05/all-you-have-to-do-is-a...
hugh3 3 days ago 0 replies      
Inaccurate. I frequently get things that I don't ask for. Also, asking for things inappropriately is a great way to annoy people.
rglover 3 days ago 0 replies      
Wow. Perfect timing on this. I've been struggling to develop a mindset like this when looking for a job. It's difficult, but you must not underestimate the value of a connection, regardless of the initial interaction (i.e. it's better to receive a "no" and be recognized than not known at all).
hxf148 3 days ago 0 replies      
It's true. Ask for what you want in this world, it will usually not be offered.

I've had to get over my own personal shyness to help promote and market our startup (http://infostripe.com) and at times it's been difficult to just ask for help, opinion and testers. But when I do it's always quite rewarding.

Ask, speak your mind, be reasonable and the things you want are obtainable.

mannicken 3 days ago 0 replies      
Wrong. If you have something people want, they'll force money on you. Make something people want. People are so naive, like they've never seen "Soup Nazi" episode of Seinfeld.
tesseract 3 days ago 1 reply      
I don't mind the phrasing, per se. I do mind when people who aren't Dustin Curtis use it.
pizza_ 3 days ago 0 replies      
This is good advice. On a first date, after introductions, I present a brief presentation slideshow to the prospective mate where I lay out a dating schedule in Gantt format denoting dates, first kisses, sex and falling in love, including sexual position diagrams. For a personal touch, copy-and-paste both your Facebook profile photos onto the people in the diagrams.
Depression, Burn Out and Writing Code muddylemon.com
372 points by muddylemon  6 days ago   165 comments top 18
michaelochurch 6 days ago  replies      
He did a better job of addressing this issue than Noah Kagan, who was talking about mild fatigue rather than real depression (yes, there's a huge difference). Let me add:

6. Exercise, with a mix of activities: swimming, running (outdoor is better) and strength training. He mentioned that the brain interprets social isolation as a sign of low status. It also interprets physical activity (or the lack thereof) much in the same way. Using your body (running, swimming, hiking) tells the brain that you're an alpha and that it needs to step it up a notch. Mild depression is when the body and brain go into gamma mode (more severe depression is a biological dysfunction of that system) which would be a form of hibernation, as observed in medieval mountain communities during winter, but society isn't tolerant of that, so it leads to misery. (Some mild depressives would be fine, aside from the loss of time, if they could sleep 11-14 hours per day.)

7. Avoid cigarettes at all costs. Use alcohol sparingly: one drink is fine, but it takes a week to recover from a night of serious drinking... and if you're coding, you actually notice the loss. Avoid drugs like cocaine and heroin like the plague. Jury's still out on occasional use of LSD or shrooms (but people with any history of mental illness need to be very cautious) but they certainly shouldn't be a habit.

8. Relax. Sometimes it's necessary to spend time in the woods and just (sorry for sounding hippy-dippy) be mindful. 24/7 Internet connectivity is not always helpful.

9. Read novels and watch movies. There's something healing about narrative that isn't provided by (as much as they are enjoyable to read) books about formal semantics of programming languages.

10. Music. It helps. I have a lot of friends who say they never would have gotten through their teenage years, or a bad turn of business, without music.

11. Go somewhere new, do something different. It doesn't have to be an "exotic" or expensive vacation; if your goal is to heal, it's better that it not be. Just a trip to visit friends 50 km away can help.

12. 2 weeks of real vacation, as an absolute minimum, per year. Take unpaid leave if you need to. Four weeks is best. Total productivity is maximal at 3 weeks of vacation and per-day productivity is maximized at 7-9. Europeans have it figured out: you need two types of vacation: exploration vacations to new places (Alaska, Cambodia, Andes) where difficulty and stress are OK and relaxation vacations where stress levels are low and variation of activity is not required. Unfortunately, Americans get stuck having to choose one or the other, and generally do only the first kind of vacation when they're young and only the second kind when they're old.

For the record, if you only get 1-2 weeks but can take unpaid leave without it hurting your career, you probably should. A 2% pay drop for an extra week of vacation is worth it IMO.

13. Fruits and vegetables. The casual arrow is unclear and probably goes both ways, but depressed people tend to eat a lot of white carbs. Americans tend to get dieting wrong, as if it's some set of religious prohibitions that have to be followed to the letter, when in reality there are few foods one should never eat. Instead of dieting negatively ("I won't eat X") it's better to diet positively and replace unhealthy foods with better alternatives (e.g. 70+ percent of your desserts should be fruit).

14. "Don't fuck crazy". High-power programming/technology, as a career, is only 10-20% more time-intensive than the average job-- you haven't seen bad hours till you've worked on Wall Street-- but it's 200% more energy-intensive. You need a supportive and decent partner who will be your rock of stability, not someone who will drain you.

SoftwareMaven 6 days ago 5 replies      
I've always found it hard, if not impossible, to “join” groups. It feels like everyone already has the friends they need. I don't know how to fix that.

This has always been a problem for me. I used to think it was a sides-effect of the dysfunctional home that moved, on average, every three months that I grew up in. I knew I developed coping skills that put a wall between me and others. You can only say good-bye to friends so many times before you stop getting close enough to care that you are saying good-bye.

Then I had kids and saw much of my own personality reflected back at me, but in different ways and degrees. At that point, I realized I probably always would have had difficulty making friends and would probably only had a couple friends. My upbringing made that worse, but wasn't likely the entire cause.

Now, I find that I don't know how to get around that. Articles everywhere say, "meet with people", but I feel more lonely in those situations because I don't know how to break into the conversation. Depression makes it worse, because you are constantly asking yourself, "why would they even want to know about me?" There's only so many times you can go to the entrepreneur meet-up and not talk to people (even after presenting) before it starts doing more harm than good.

I agree with the article that connecting with other people is of number one importance to fighting depression. I would really like to find the guidebook for doing that. :/

nkurz 6 days ago  replies      
I'd love to hear more people's thoughts about his "Just Say Yes To Drugs". I'm at the point of considering drug treatments, but intellectually, I'm very biased against them. I avoid illegal drugs, and rarely take legal ones. I'm not absolutist, but I'd go out of my way to avoid coffee and ibuprofen. For his example, yes, I would feel the same way about blood pressure and cholesterol, and would work very hard to change my lifestyle before relying on the drugs. On the other hand, something needs to change.

I'm scared of both the dependency and the effects. I think my basic fear is that if I find something effective, it will be suicide-light, that I'll be killing off too many parts of my self that make me who I am. It's hard for me to trust the testimony of the treated --- I presume that there are many satisfied frontal lobotomy patients as well. So oddly, I guess I'm more interested in second-hand accounts. My personal experience with treated friends is mixed. A few seem to be more functional versions of their previous selves, others are much closer to zombie shells.

It's a hard subject to discuss, though, and I don't understand whether the differences are chance, personality, or drug regimen.

How do you feel about your newly drugged up friends? Specific treatment details appreciated.

georgieporgie 6 days ago 3 replies      
Has anyone found a working environment where they're able to balance social interaction with the isolation and focus needed to code?

I've only been happy at two jobs. At one, I shared an office with an awesome guy, and our personalities and senses of humor meshed well. We had a significant overlap in musical tastes, and enjoyed sharing music and witty banter throughout our days of work.

The other was a startup where five of us worked in an open room, regularly taking breaks for coffee and lunch together. We had a bat-shit insane CEO, which really gave us something to bond over. :-)

Looking at job postings is overwhelming. 50% claim a "work-life balance!" and a "work hard, play hard!" attitude, where I'm betting the work is harder than the play. The rest talk about "entrepreneurial spirit" and other codes long hours and isolation. It saddens me that listing hobbies and interests (more like passions) on a resume is considered unprofessional.

araneae 6 days ago 0 replies      
>He mentioned that the brain interprets social isolation as a sign of low status.

That might be the case, except most of my social interactions are failed social interactions, which make me even more depressed.

etherael 6 days ago 0 replies      
I have a strong urge to rebuke most of these "We're programmed to be like this, low status in our tribe etc etc" type lines of reasoning as outright ridiculous, however after considering it a little deeper it seems to me that's simply one interpretation of what actually happens.

My desire to socialise atrophies, I do not feel like leaving my house, I want to focus harder and harder on creating things, learning, etc and I have less and less time for "normal life". It's possible to focus on what I'm drawing away from and characterise it as a loss, but to my view it is simply a sacrifice I choose to make of my own free will for something that I find more valuable.

To my view removal from the tribe is not terrifying, it is not even undesirable. You lose some things, but you gain others, as with any narrow focus. There is no single right interpretation of what that actually means. Until you add in the subjective values of the interpreter, it's just a set of occurrences.

billjings 6 days ago 5 replies      
Stay active. Do something cardio intensive, and a lot of it.
Wickk 6 days ago 0 replies      
>Drugs don't have to be the only tool you use, but they can get you to a place where you're healthy enough to think clearly about alternatives.

This bears more emphasis. I was on medication for a very long while and whenever someone asks me how I got to where I am now, I'll jokingly say a very lengthy and close relationship with various different anti-depressants.

Of course this ruffles feather in people in all the wrong ways, but its the truth. Anti depressants and psychotropics aren't going to magically cure you, but they are going to help you get to that place where you can figure out what's wrong and work towards getting better.

raju 5 days ago 0 replies      
Great article, and some good insight. Like many others I have found that doing a little bit of exercise every day really helps.

I used to (and need to get back to - been slacking off for a month or so now) do burpees. They are relatively quick, with a mix of cardio and some strength training, and hardly take any time at all.


I usually do the "pyramid" that is start by doing 11, then a minute break, followed by 10 all the way down to 1. It takes me roughly 15 minutes, but I end up feeling really good after that.

Also, taking a break from arduous mental activity is a good thing - watch movies, read fiction, step outside with a point and shoot camera and take pictures. Anything that let's your mind relax for a bit.

Finally, I have found that meditation really helps. It helps to slow you down, improves focus, and the world really seems different once you get used to it. Just concentrate on your breath, for 15 minutes a day (or twice a day) and everything seems a tad fresher and clearer.

Of course, this article (and the comments) have got me thinking about a vacation. I don't take enough time off and I should.

gregfjohnson 6 days ago 0 replies      
There seems to be some deep intuition that anti-depressant drugs will adversely affect something essential about who a person is, how they will function, or how they will experience and perceive the world. I absolutely share those feelings and concerns. A therapist gave me great advice: let's be empirical. Instead of basing decisions on supposition, let's try a few things, gather some data in terms of how various options affect you, and base decisions on that accumulation of information. The only risks I could see with this strategy were (1) going on a drug and then stopping it would permanently damage me in some way or (2) I would become dependent on something.

The other thing that has really influenced my thinking is how it is for people I love to live with me. Doing something to treat depression might have a noticeably positive impact on them. So even if my preference was to not receive treatment, based on a sense that the joys I love about life might become muted, or that I might not be able to think as well, or whatever, there are other people to consider.

chmike 6 days ago 0 replies      
What strikes me is that you consider the need to fight against your depression. Consider the possibility that you are really exhausted as if you ran a marathon. Obviously, you'll be slower and slower and probably hardly running in the end. There may also be a reaction of your body to force you to slow down so that you don't exhaust you to death.

I also found out on myself that one may become tired for some type of activity, but I could switch to another type of intellectual activity giving a rest for the other. This is like switching to an activity using your arms when your legs are tired.

Every depression or burnout is different by its cause and they way we react to it. One should be carful with popular advices. I believe in the need to rest and get a break from what is exhausting. Then work out the small positive steps, etc. Or switch activity. Checkout vitamins, B especially.

Alex3917 6 days ago 3 replies      
"I always hear people beg off trying medication for mental health because they 'don't want to rely on a drug for the rest of their life.' I never hear that statement when someone is talking about blood pressure medication or cholesterol pills."

Then I take it you haven't been paying any attention to the news this past week.[1]

There are literally thousands of academic studies on depression. Is it really too much to ask that people learn something about the major findings in the field before giving others advice? Sorry, but it's extremely irresponsible to start taking drugs (or recommending them to others) without reading up on the academic research on what they do, whether we're talking about smoking crack or taking something your doctor prescribed.

[1] http://www.pbs.org/newshour/bb/health/jan-june11/cholesterol...

turoczy 6 days ago 1 reply      
Second that emotion on the exercise. And diet is equally important. While the Red Bull, pizza, and caffeine swilling stereotype is prevalent, a well-thought out and balanced diet can make your coding more productive. You'll be creating fewer bugs, thinking more clearly, and feeling more positive about your progress if your (exercised) body has the fuel it needs, too. And reducing the number of stimulants you're ingesting will help with that whole sleeping thing too.
gcb 6 days ago 1 reply      
have a hard time accepting the Yes to drugs.

have friends that take antidepressants and are ok. while others became totally disconnected with other peoples feelings, like they are small kids again. zero emotional intelligence, even tough they themselves seem happier.

lelele 5 days ago 0 replies      
This recipe worked for me:

- first thing, check for physical causes (thyroid, etc)

- listen to motivational/inspirational speeches by people who made it

- get out of your house! Be among trees as much as you can, and get as much (indirect) sunlight as you can.

- work out: both cardio and strenght training. Regarding strenght training, bodyweight exercises are fine. Regarding cardio: do not run! Unless you have proper technique, you can injure yourself. Fast-paced walking is OK. Other favourites: stair-climbing, biking.

- watch what you eat: avoid high glycemic-index foods

- avoid consuming too much alcohol

- don't be alone all the time

tritogeneia 6 days ago 1 reply      
I'm not sure which of these factors contributed most to getting out of my bad spot, but in no particular order:

1. Running. (Strength training is good for many other goals, but frankly handling heavy objects isn't safe when you're severely fucked up and can't see straight. I can run even when I'm pretty deep in the hole and it has an almost curative effect.)

2. Making friends.

3. Starting a project I was actually successful at.

4. Learning not to tie my self-worth quite so tightly to my performance. The notion of "human dignity" -- the idea that all people have worth, even the ones who have personal flaws or make mistakes -- is really important.

atlantic 5 days ago 0 replies      
Something that has helped me a lot recently is to hit the gym every day. This has knocked a couple of hours per day off my working hours, but surprisingly I am more productive than before, because my mind is clearer and my mood is more positive. What I actually do in the gym is not that important - at the moment my focus is on cardio and abs. A second important point is to stay away from IT work on weekends - that way I start the week fresh and eager to do the job.
xedarius 6 days ago 1 reply      
Christ, it's a job, you're writing computer code. If it's that bad do something else. Nobody will die (or even care) if you leave the project, I know it feels like the world is on your shoulders, but it really isn't.
What I Learned From Fixing my Laptop's Motherboard spinellis.gr
334 points by llimllib  3 days ago   80 comments top 26
noonespecial 3 days ago 4 replies      
I saved a hard drive with some very important stuff on it exactly the same way. I had ham fisted the power connector in backwards and the sacred smoke had escaped. Lucky for me, I had an o-scope and was able to trace to a capacitor that had shorted in the 5 volt path. I wired up a new one and all was well.

You know that awful sinking feeling you feel when you've just done something monumentally stupid and embarrassing that you're going to have to fess up to? It feels twice as good +5 when you rescue yourself from it with your geek super-powers.

mcantor 3 days ago 0 replies      
Hmm... I guess I'll be the one who posts the Fauxgeek's Lament on this thread.

Reading stuff like this always makes me feel like a second-rate geek! I've had my fair share of soldering iron burns (for my senior project in college, I cracked open an RC car's remote and connected it to a BASIC stamp, which I controlled via RS-232 and a network socket), but the parts of this article which the author considers "easy" stuff are still a mystery to me. For example, "According to the application note I could use a simple signal diode, so I just pulled a 1N4148 diode out of my component drawer." A simple what diode? A 1N4-what diode? My component drawer? I don't even have a... help! S.O.S.! Man overboard!

I'm sure that I could come to feel more comfortable with stuff like this if I went out of my way to learn more about it, but the applicable extent of my electronics hacking day-to-day has been unplugging the rumble packs from my Xbox360 controller. Most of the things I use on a day-to-day basis are not devices that I feel comfortable screwing around with!

timtadh 3 days ago 2 replies      
Reminds me of when I tried to fix my fried amplifier last Christmas. The amp was around 20 years old and had mysteriously stopped working in the night. Being cheap I enlisted an electrical engineering friend and together we stripped the amp down. However, the problem turned out to be 2 diodes failing hot. The short fried the micro-controller and associated eeprom thus making the fix more trouble than it was worth. Still, process of diagnosing the issue was a great learning experience for me. I almost exclusively work in software so it is always a pleasure to learn about the electrical engineering side of things.
ljf 3 days ago 5 replies      
top tip: use an ice cube tray to hold all the screws you remove. I remove the first set, put them together in the first cube hole and continue. then just reverse the order of cubes when you are putting it back together.
jamesbritt 3 days ago 2 replies      
When I was young I remember consumer electronics, like tape recorders, coming with their circuit diagram as part of their documentation.

I bought a Teac 4-track open-reel tape recorder many years ago. Still have it, as well as the docs. I swear you could build one from scratch with the info provided.

The option to open stuff up (phones, radios, TVs, tape players, etc.) and make them do things it was not intended for was magical. Maybe Arduinos allow for something like that feeling now; seems infeasible for most general consumer electronics nowadays.

juiceandjuice 3 days ago 0 replies      
Haha, I did the EXACT same thing on my (original 2006) macbook, 14 months owning it. My power supply died, so I clipped the cable, hooked it up to a 1A 18V variable power supply, but it wasn't charging fast enough, so I hooked it up to another that was apparently broke and sent 120V to my laptop (people at work don't throw broken things away apparently)

Anyways, I had experience soldering SMDs before so I wasn't too worried about the repair, but there were too many power mosfets and diodes and and a ton of devices at the same potential. Testing voltages and continuity with my multimeter was mostly futile (and I used datasheets the entire time). I'm glad you found the problem.

rb2k_ 3 days ago 1 reply      
I had a Samsung Laptop that I broke while trying to downgrade the Bios image without properly downgrading some other controller image first. The only problem was that the soldered EEPROM had the wrong image and thus the laptop decided not to start. Samsung wanted to charge 700+ Euros (CPU was soldered to the motherboard).

I was really confused that the world's biggest electronics concern wasn't able to replace an EEPROM.
Luckily I came across bios-fix.de and paid 65 Euros including shipping to get the old eeprom soldered out, flashed and soldered back in.

I wish bigger companies would actually try rather than going down the service path of least resistence :(

dhughes 3 days ago 4 replies      
Take pictures of each step don't use paper since flat surfaces and screws soon part company use paper cups instead .
samsonasu 3 days ago 2 replies      
Good story. Reminds of when I fixed a PlayStation 2 that wouldn't turn on by replacing a blown diode in its power supply. I wonder how many complex electronics are out there Sitting in landfills just a single diode away from working perfectly.
jdietrich 3 days ago 2 replies      
Genuine protip:

Replacing through-hole parts on motherboards is exceptionally difficult. There's so much copper in a modern multilayer board that it's very difficult to get enough heat in without cooking something delicate. Even with preheat, a reflow gun and a powerful iron it can be very hard going.

Replacing broken PSU connectors is one of the most common laptop repairs and should be an easy job, but has a dismally low success rate.

kabdib 3 days ago 0 replies      
Common sense will get you a long way. I was replacing failed memory chips on a Vax 11/780 and doing other hardware-level things in pretty much the same way; the downside of screwing up was a multi-thousand-dollar service call to DEC, since our machines were not under maintenance contract. Well, and bringing a machine that twenty people were using down, for days (which would have been the more expensive piece of the equation).
r0s 3 days ago 1 reply      
This is why I will always prefer a desktop PC.

Modular components and hardware standards with many third party manufactures. Laptops will always hold second computer status for me.

swishercutter 3 days ago 0 replies      
Always remember the Maker rule..."If it is broken it is fair game..". Open it up, do your best, if you don't fix it but you learn something then its time well spent.

If you are like me you also salvage as many parts from what cannot be (or is not worth) fixing there is always something else to use the parts on.

kylemaxwell 3 days ago 0 replies      
Really handy stuff and a reminder that "this product contains no user-serviceable components" is rarely true. I also liked the structure of the post. Isolating the lessons and documenting his work while still discussing the higher-level implications certainly made it more useful than just a walk-through of how he fixed the motherboard.
pacemkr 3 days ago 0 replies      
I fixed my t42 motherboard by soldering a tiny wire over a blown fuse.

I was in the process of replacing the failed fan on the CPU heatsink. (Had to transplant the fan from a smaller heatsink that came from a lesser t42 with a smaller heatsink that didn't fit... Long story.) Prior to the procedure, I wanted to carefully see if there was even voltage on the FAN pins. I wasn't careful enough and thought I shorted the pins. Indeed, when I connected the working fan, it didn't work. So I figured that a fuse blew somewhere. Did some searching on the web, found a partial schematic for the board, found the 1mm x 2mm FAN_FUSE (or something) and soldered a wire on top of it. To my surprise it worked. Still using that t42 when on the go.

This t42 is my Oliver, basically.

nopakos 3 days ago 1 reply      
Some trivia: Diomidis Spinellis is currently the General Secretary of Information Systems at the Greek Ministry of Finance. [ http://en.wikipedia.org/wiki/Diomidis_Spinellis ]
linker3000 3 days ago 0 replies      
Lesson 0 has to be 'have several years electronics repair skills under your belt' - and that's the kicker for most.

I agree though - a bit of knowledge can be a powerful thing; a 'resistor pack' for my car (to regulate the 4 speed aircon blower) is 80UKP + labour to replace when it fails but the most common fault is that the thermal fuse (approx 0.80UKP to replace) blows when the airflow is reduced by an air filter that just needs a clean-out.

samlittlewood 3 days ago 0 replies      
Addition to lesson 2: Use a camera or video recorder to record each step of the disassembly.
rcamera 3 days ago 0 replies      
I once had to replace my laptop's motherboard because the gpu part of it fried due to a malfunction that cause alot of heat. I had to buy another motherboard at Asus, that was actually pretty cheap, 150 dollars, but I didnt want to pay another extra 100 dollars for, the technician to change the boards.

Well, I couldn't find the service guide, all the ones I found were diagrams, but not a guide on how to dismantle the laptoo, so I decided to figure it out myself. After 4 hours I managed to take the whole motherboard apart and extract all components, then another hour to install the new one. It was pretty challenging, but I managed to close the case with all parts in place and no extra screws lying around.

chopsueyar 3 days ago 1 reply      
That is ghetto fabulous (heat-shrink tubing with diode). Nice juxtaposition with the SMD stuff.

Well, the diode did as it was supposed to, and prevented the reverse-polarity from damaging anything of vital importance.

Nice write-up.

dazzla 3 days ago 0 replies      
And I was proud of not destroying my laptop when I replaced the power socket on the motherboard.
Estragon 3 days ago 3 replies      
This is cool. How does one go about learning these kinds of skills?
Tarks 3 days ago 1 reply      
I love stuff like this because the time it took him to fix it was probably worth more than the cost of the motherboard BUT the fact that he pulled it off is what makes him so valuable (and cool ^_^ ) and he learned a lot while doing it. Simply awesome.
raheemm 3 days ago 0 replies      
After a zombie attack, these are the kinds of geeks who'll help to rebuild civilization. Seriously!
mattberg 3 days ago 0 replies      
Off topic: Anyone else read this title as "What I Learned From Fixing my Mom's Laptop"? I think I have been doing too much tech support for my mom and her friends lately.
cstrouse 3 days ago 0 replies      
Awesome post. I've always wondered what it would take to fix a mobo these days.
How Perceptual Hashes Work hackerfactor.com
314 points by brudgers  1 day ago   69 comments top 19
wickedchicken 22 hours ago 2 replies      
The two main problems with this approach is it is not rotation-invariant and it does not work well if the image is damaged or added to. A more robust system (that, admittedly, will take longer) is to use one of the affine-invariant feature detection algorithms pioneered by SIFT. SURF is a faster, open-sourced version of SIFT that has many implementations. Essentially it scans chunks of the image at different scales and identifies features that peak even as the chunk around it gets bigger. Once these are identified, they are described in a way forcing them to the same size and orientation for lookup. Since these features should presumably be scattered throughout the image, the image can be recognized even if certain features are obscured or modified. It's certainly not as straight-forward as a DCT metric on a downsampled image, but the nature of widespread image capture, creation and manipulation usually requires this robustness.
stevetjoa 21 hours ago 1 reply      
If you're interested in this article, then you may interested in locality sensitive hashing (LSH), a randomized hash that has been used seemingly everywhere. I recently used it to speed up music source separation (papers pending).

The idea is similar to the one mentioned in this article, but more general. Unlike a cryptographically secure hash where x != y implies that h(x) != h(y) (collisions aside), LSH says that if x and y are "near", then P(h(x) = h(y)) is "high". This quality is important when doing robust similarity search. For example, if your image is noisy or rotated or scaled, you hope that you can still find the clean version in a database.

LSH has been used in many application domains including images, video, music, text, bioinformatics, and more. LSH is not directly comparable to a feature extraction algorithm such as SIFT.

[Edited for clarity.]

piotrSikora 22 hours ago 2 replies      
This reminds me of OpenSSH's fingerprint visualization support ("VisualHostKey yes"):

    $ ssh A.B.C.D
Host key fingerprint is b0:c9:c9:96:fb:fd:ac:a4:ff:70:8f:1b:35:f4:f9:2e
+--[ECDSA 256]---+
| |
| |
| . . |
| o * . ..|
| O S o..|
| . . . ..|
| . o o .|
| . + + +E. |
| o.++*....|

me@A.B.C.D's password:

Original article introducing this feature: http://www.undeadly.org/cgi?action=article&sid=200806150... 2008/06/26).

kenjackson 23 hours ago 0 replies      
That was one of the best articles I've read in a while.
michaeldhopkins 1 day ago 1 reply      
Thank you, I learned something.

Since TinEye pre-computes the hashes, do they use something like Redis to retrieve information? Redis seems perfect for a such quick results using the hash as the key and a URL or object of some kind as the value.

seanalltogether 21 hours ago 5 replies      
I've always wondered how services like Shazam work. I'm amazed that they can do this kind of perceptual hash against ANY 10 second portion of a song. How do they search against something like that when they don't know the start or end time of the segment that is being input?
NickC_dev 21 hours ago 1 reply      
There's a publicly available implementation of a perceptual hashing algorithm called phash at http://phash.org. I use some of their c++ code to detect reposts on an image sharing site I run (http://lolstack.com).
CountHackulus 11 hours ago 0 replies      
This is really interesting, but I'm sad that they skip over the step "convert to grayscale." There's a ton of ways to convert an image to grayscale, each with their own pros and cons.

How do you weight each channel? Do you convert to HSL and just use L? Do you instead use Lab? HSV? Do you do a global or local algorithm? So many questions!

there 21 hours ago 0 replies      
libpuzzle is good for this type of work: http://libpuzzle.pureftpd.org/project/libpuzzle
noblethrasher 21 hours ago 4 replies      
"With pictures, high frequencies give you detail, while low frequencies show you structure."

I have a vague idea of what this means but can someone please explain it in a bit more detail?

krisw 5 hours ago 0 replies      
What a fascinating topic, it's too bad that Gazopa (mentioned at the end of the article) just announced that they're shutting down their CBIR service:


zackattack 14 hours ago 1 reply      
I don't understand why compressing an image is going to generate the same 8x8 image each time no matter what aspect ratio it was originally... whether it has been stretched before.. If you stretch and then recompress a bunch of times don't you eventually lose the information?

That math is for some reason totally counter-intuitive to me. Could someone do a proof?

andrewflnr 23 hours ago 2 replies      
Wow, I had no idea it was so easy. Could this in fact be used to combat copyright infringement for the likes of The Oatmeal on large scale, given someone (Google) with the requisite computing power?
iandanforth 21 hours ago 1 reply      
I have a feeling there is a deep connection between perceptual hashes and compressed sensing. Could someone more familiar with the latter weigh in?
buddydvd 21 hours ago 1 reply      
Isn't tineye's signature algorithm based on Fourier transform?


tobylane 16 hours ago 0 replies      
Tineye must take this so much further, one image I looked up was about half of one of the pictures returned, I think it was in a page of a book.

From the reddit link further down - "A Fourier transform takes a signal in the time domain and breaks it down into its frequency components. Simplified, it takes a CD and produces sheet music.", "To be clear, OP says this is a matching algorithm - it's not what tineye uses, because matching the signature from the searched-for image with the database of previous signatures which is probably a nightmare."

tsewlliw 22 hours ago 1 reply      
That is completely brilliant! I want to go out and write a image diff viewer that uses this on blobs in the image to detect pieces moving around!
joshu 23 hours ago 1 reply      
See also SIFT
mdonahoe 23 hours ago 0 replies      
"I'll use a picture of my next wife, Alyson Hannigan."

That had me doing a few google searches.

Try to digest the Groupon story shortlogic.tumblr.com
297 points by jagira  11 hours ago   141 comments top 24
kprobst 10 hours ago 1 reply      
Everyone remembers the Pets.com IPO for example, but no one remembers that it was Merrill Lynch that underwrote it and made most of the profit on the deal.

The problem is the market and VC environment that allows things like these to go down, not that there are unprofitable companies. Those have always existed. But there was a time when people didn't buy into them because they were unprofitable. The way things work now is just wrong and dangerous. But it's been 10+ years since the last dotcom bust, so I guess people will be people and start falling for it again. A fool and his money and all that.

starnix17 10 hours ago  replies      
In case anyone didn't notice, this blog post is by DHH.
ChuckFrank 8 hours ago 0 replies      
False valuation and poor accounting and financial transparency affects us all. We should do everything we can to discourage these type of shenanigans. A much better proposition would be to build companies that people want, and are willing to pay for, and then raise money based upon the future possibilities of those creations. Raising too much money is just as bad for us, and for Groupon, as raising too little money. Why? because any mis-allocation of capital will lead to market instability, which will lead to inefficiencies. Companies, Societies, Empires, rise and fall based upon these simple tenets.
ben1040 11 hours ago 4 replies      
Pardon my ignorance but how often do insiders get their stakes cashed out after funding rounds like this, as opposed to using the money for operations or growing the business?

My lack of knowledge about corporate finance and the part of that blog post that refers to cashouts is making me see Groupon as more of a DrKoop.com or Webvan than it may be. But nevertheless I can't see how struturing a funding deal like that benefits anyone other than the insiders who got in early enough?

nanoanderson 11 hours ago 0 replies      
It's always a big red flashing warning sign to me when investors who've stood by a company from the beginning take a major cash-out _before_ an IPO. Nobody can argue they didn't know an IPO was coming soon.

If you don't believe you'll make more money from an IPO than a private investment, then what does that say about your faith in the company's future profitability?

alain94040 10 hours ago 2 replies      
My main concern is: how do I tell my 401k and mutual funds to stay away from Groupon?

I'm a smart investor, I'm strongly deciding to not touch Groupon with a thousand feet pole. But I'm pretty sure bankers in NYC will do it for me. What can I do?

inkaudio 9 hours ago 0 replies      
Another major challenge Groupon face is:
They are competing with major players, companies with considerable more resources. Google, Facebook, Amazon all have competing products and a very large user base. Google has billions in profits and Amazon have billions in revenue and hundreds of millions in profits. Facebook has one of, if not the largest user base in the world. For Groupon to succeed it will have to be beat Amazon, Google, and Facebook. It's hard enough to compete against one giant, but having limited resources and competing against all three will make it too easy to fail. Now more than ever you have to wonder why they did not sell to Google. Perhaps they are hoping Microsoft will want to get in game after they IPO and buy them.
aresant 10 hours ago 6 replies      
I count 3 negative Groupon articles on the front page this AM.

A quick Google news search finds similarly uniform negative sentiment in the general web.

Brings to mind Warren Buffet's famous quote about investing - "The time to get interested is when no one else is. You can't buy what is popular and do well."

jolan 11 hours ago 3 replies      
They need to move on from low margin stuff like food and physical services.

* Copy AppSumo and sell software/webapp services.

* Try to work out deals on high value items like automobiles where there's lots of wiggle room in pricing.

* Talk to Apple. They have high demand and large margins.

Do anything to get rid of thousands of cold calling sales people.

nwjsmith 10 hours ago 2 replies      
I'd be interested in hearing Jason Fried's take on this. Isn't he an advisor for Groupon?
bproper 11 hours ago 0 replies      
And while Groupon argues it will be sustainable once it hits a certain threshold, its business is actually decaying in older markets like Boston -


evertonfuller 29 minutes ago 0 replies      
I've always seen Groupon as a joke. I don't know anyone who has bought anything from there. Maybe I just hang out with the wrong crowd, but meals at random sushi bars, spa treatments, karate lessons...? Random...
gaius 11 hours ago 2 replies      
People are only supposed to get rich after the IPO, and even then, only on paper, for the first few years.

Still, VCs are supposed to be grownups who know what they're doing, if they got screwed it's their own fault.

jrp 11 hours ago 2 replies      
If the article is correct (future really bad, tricked public going to invest on day 1), wouldn't it make sense to invest and sell very soon?

Is something special about an IPO where you have to hold it for some time?

miespanolesmalo 11 hours ago 3 replies      
I always wondered how Salesforce became so successful. Their saas is janky garbage. Their software is everywhere, but they excel at nothing.

Now I know their success is hype and creative accounting.

klbarry 9 hours ago 3 replies      
Does anyone else think Groupon has been a huge boost to the economy? They've created 7000 jobs in a few years, without displacing many old ones, and most of these jobs are low-skill.
sedachv 8 hours ago 1 reply      
I told you so:



With all the experience of the past 12-15 years, I'm convinced that "local" is a hopeless hole for tech startups.

dbaugh 11 hours ago 1 reply      
Whenever a business grows by hiring more and more people it won't scale well and be massively profitable.. People are expensive. They are always going to have razor thin margins.
raldi 9 hours ago 1 reply      
I don't know if I agree with this assessment. There's a land rush going on. They need to grab as much of the market's network effects as they can before Google Offers and the like manage to get going. This is definitely one of those extreme first-mover-advantage situations, like online auctions fifteen years ago.

So to answer the author's question: When will Groupon be profitable? When the enormous costs of their explosive growth are no longer counting against their bottom line.

danvoell 10 hours ago 0 replies      
Should they take customer acquisition out of the "fastest growing internet company ever" title? or just the financial analysis.

Great Write-Up.

I have to imagine the wall in their office with biggest internet failure magazines is a self fulfilling prophecy.

lightoverhead 9 hours ago 1 reply      
Completely agree with OP on this comment!
I really doubt Groupon's business model, hardly believe that coupon style business could be a main stream of profit.

Would like to short this IPO when it comes to the stock market.

kjames 10 hours ago 2 replies      
Why do they need to go public? Are they hoping that more ad buying power is going to provide them economies of scale? Graphic designers aren't that expensive to employ are they?
johnx123 11 hours ago 0 replies      
If the suspicion that it's a paid news http://rajeshanbiah.blogspot.com/2010/12/groupon-hype-or-pai... they'll have to shut it down when others start following that strategy.
rayiner 10 hours ago 3 replies      
This article is full of handwaving.

1) Lot's of businesses run in the red for awhile in order to build up their market share. Just pointing to losses in the first few years of major operations isn't particularly compelling.

2) The adjusted CSOI metric isn't as ridiculous as the author makes it seem. The claim is that their marketing expense is a function of how fast they want to grow. Presumably once they reach a size they want, they scale back that expense, and the revenues turn into profit. The Forbes article says: "The bottom line is that considering the profitability of Groupon without online marketing expenses is silly; Groupon without marketing expenses is not Groupon at all." How is Groupon without marketing not Groupon at all? It's an interesting-sounding statement without any meaning.

Zed Shaw: Github's Favorite Joke sheddingbikes.com
278 points by pufuwozu  3 days ago   136 comments top 26
rtomayko 3 days ago 1 reply      
I'm the GitHub employee that fixed the bug. I usually try not to get involved with stuff like this but I feel like I'm in a unique situation to correct the record here.

"They purposefully allow this glaringly obvious mechanism for insulting and annoying their members and are actually involved in the joke."

I've never heard of the joke. I've never heard of anyone at GitHub being involved in one of these jokes.

There has been one case that I'm aware of where someone mass added people to a repository in order to fill up activity feeds. That person was banned. It's an issue we'd like to address more generally.

"Until I broke their server they were all laughing at my 'testing' then they were pissed when they had to fix the bug I found."

I fixed the bug without being aware of any of this. I check our exception monitor every day. It was there. It was obvious. I fixed it.

It was a simple bug triggered by branch names that look like commit SHA1s. Here's the commit:


It affected only branch list pages with branch names matching [0-9a-f]{40}.

"If you don't believe me, look at the HackerNewsTips twitter account, which I know is astroturfed by a github employee."

That is not a GitHub employee. We don't hire anyone that witty as a rule.

nestlequ1k 3 days ago 1 reply      
Zed really has a victim complex. As Ryan validated above, I can guarantee you anyone who has any authority at all at github did not encourage or sanction this behavior. It may not have been high on the fix queue, but thats because they expected people to be rational, and they can just manually ban the people who abuse the system.

Zed.. seriously... the world is not out to get you. We love your code and your contributions. Don't let a few losers get you down.

marcamillion 3 days ago 2 replies      
Love Zed, Hate Zed or neither, it is disappointing if Github really allows this type of abuse to happen with no repercussions.

...and I have been a major Github fanboi from day 1, so it sucks for me to see such pettiness.

Quite frankly, I am not a big fan of Zed Shaw either...his hissy fits tend to get on my nerves....but come on man Github.

You are a damn company!

davidhollander 3 days ago 3 replies      
Interesting how this disappeared from HN's front page without being dead-ed.

I seem to recall the 'Programming, Motherfucker' link meeting a similar mysteriously early demise, despite its 800+ upvotes.

shaggyfrog 3 days ago 1 reply      
Is GitHub really aiding and abetting this kind of anti-social behaviour/feature? This kind of allegation demands an official response, and soon. Otherwise it's mega bad PR.

More than anything, I'm surprised that you can add a collaborator to a project unilaterally. A way to request an add seems like a natural choice considering you can request pulls.

peterbraden 3 days ago 3 replies      
What a bunch of children.

It's not cool to antagonize people

It's not cool to DDOS a service.

It's not cool to call someone out in a public forum.

Grow up people.

gooddelta 3 days ago 5 replies      
I'm a bit surprised that Github would allow one of their employees to behave that way without firing them. Love Zed or hate Zed, he's still a user, and you're still a company providing a service; show some professionalism.

Zed: I'm on your side here, but please stop rising to these assholes' attacks. You're not going to convince them to change their ways, and getting into a pissing match with them gives them what they're looking for. You won this one, but your time is worth more than this.

randrews 3 days ago 1 reply      
Livejournal had this problem a few years ago. You could add anyone to a community, and everyone's profile page had a list of all the communities they were in...

Zed should absolutely get off Github; he's enough of a troll magnet that people will figure out ways to mess with him even after this is fixed. It's easy enough to set up a bare repository on a random machine you have SSH access to.

Jd 3 days ago 3 replies      
Yet another reason why some of us fear getting too close to the Ruby hipsters.
eropple 3 days ago 1 reply      
The most entertaining thing about all this is that you can merge against nonexistent branches in git.

I think my eyes just crossed.

kenneth_reitz 3 days ago 2 replies      
It should be noted that GitHub does allow users to remove themselves from collab'd repos (the article stated otherwise.)

Being a collab on a repo has no affect (public or private), other than the fact that it will show up in your personal repo list when you're logged in.

GitHub has also banned users in the past for abusing the service. I'm sure they would have acted accordingly if a complaint was filed.

hardboiled 3 days ago 0 replies      
I think the larger community has to shun this kind of behavior from Github. It's unacceptable and unprofessional.

Although I don't like how Zed is repeatedly qualifying the homophobic Martini as a Ruby guy, as if Ruby made him this way, Martini is a douche from his behavior. This has nothing to do with languages.

I'm with Zed on this.

ryan-allen 3 days ago 0 replies      
This story is no longer on the front page for some reason.
tantalor 3 days ago 1 reply      
All the dick jokes are hilarious when you're hanging out with your buddies, but GitHub is a professional forum. Would you do the same thing on LinkedIn? How does that kind of misogyny reflect on your colleagues and employer?
thenduks 3 days ago 2 replies      
This whole thing is a waste of bytes.

Simply adding someone to a repo makes it show up in their private repositories list and nowhere else. Annoying? Probably. A big deal? Hardly. (To have it show up anywhere else you have to commit to it.)

The HackerNewsTips twitter account is not confirmed to be a GitHub employee and rtomayko, who is at least somewhat trustworthy, flat out says it isn't.

The only interesting thing here is the bug related to the `[0-9a-f]{40}` regex -- and that's been fixed.

It's over, and it would be great if we could move on quickly from this one.

grimen 3 days ago 1 reply      
Professionall trolling. GitHub made the world a better place - assuming open-source makes the world better that is which I do - and they've done this pixel-perfectly IMO downtimes aside. What this is about is pure hate from Zed, probably jelousy that some people from his past succeeded silently building one of the greatest web companies that exists with hardly no funding while Zed himself was writing rants. Mongrel might very well require a intelligent mind, but every intelligent person knows that the only way of owning trolls are by the dog treatment: ingore the trolls, and go on building great stuff. I know average people knowing this as common sense.
awakeasleep 3 days ago 0 replies      
Zed, you gave me a hearty laugh.

But please take a minute to think about starting a mud-slinging contest with a pig. Think about how you repeatedly stated that you didn't understand the motivation to show you the peni, where you direct your limited attention and time, etc.

If you are considering all that already, well rock on :)

ruethewhirled 3 days ago 2 replies      
Would be good to see github do something about it. Was looking to use github for some new projects but after reading this now I might have a look at alternatives
leif 3 days ago 1 reply      
While it would've been better to just ignore him, I must say, fighting back with working code lends Zed some amount of class.
kuahyeow 3 days ago 2 replies      
Time for Github to start banning people
Joakal 3 days ago 0 replies      
Reminds me of this troll 'Insulting source code': http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=477454
brown9-2 3 days ago 0 replies      
So does anyone know who runs the HAckerNewsTips twitter?
ryandvm 3 days ago 0 replies      
That was an obnoxious read.
prez 3 days ago 0 replies      
> Getting off github and onto bitbucket would another 20-30 minutes.

Actually, much less than that - 5 minutes at most, for creating the repo.


georgieporgie 3 days ago 0 replies      
Wow, so much drama.
fearsome 3 days ago 0 replies      
Zed: Here's how you fix it - Get the 4chan on it. They'll create projects with all sorts of unsavoury content then add all sorts of people. They'll fuck up GitHub once and for all and force it to shut down.

Live by the trolls, die by the trolls

The Revolutionary Birth Control Method for Men wired.com
272 points by spottiness  3 days ago   157 comments top 21
ComputerGuru 3 days ago 2 replies      
Wow. The entire premise of the article aside, for me the idea that medical treatments can be made with physics and not medicine was an incredible, incredible revelation.

This man (not a doctor!) has invented a method to neutralize sperm in a way that would never have occurred to me. He uses the most basic principles of physics (magnetic charge) to neutralize sperm!

Realizing that sperm is negatively charging, he simply coats a short section of the inside of the vas (tiny tube going from testes to the penis, located in the scrotum) with a positively-charged polymer. As sperm travels in this coated tube, the ionic attraction causes damage on a cellular level in the sperm, the pull effect effectively destroying the sperm "tail" and preventing it from fertilizing a female but without hormonal/medical methods!

For me, as an engineer, this was a true revelation.

haberman 3 days ago  replies      
In its report, the WHO team agreed that the concept of RISUG was intriguing. But they found fault with the homegrown production methods: Guha and his staff made the concoction themselves in his lab, and the WHO delegation found his facilities wanting by modern pharmaceutical manufacturing standards. Furthermore, they found that Guha's studies did not meet “international regulatory requirements” for new drug approval"certain data was missing. The final recommendation: WHO should pass on RISUG.

Is this an example of why health care is so incredibly expensive in this country?

Do these "modern pharmaceutical manufacturing standards" actually buy us extra safety?

noonespecial 3 days ago 1 reply      
Looks like the future of space, medicine, and technology belongs to the countries that have not yet hobbled themselves with lawyers and bureaucrats to the point that something must be perfect if it is to exist at all.
solarmist 3 days ago 2 replies      
The world has been promising male birth control for decades now and the techniques (There are several that all work at almost 100% and are fully reversible) are proven (numerous long term trials, I can think of two from a decade ago off the top of my head, one in Washington state and one in Australia, in their final stages, with fewer side-effects, complications and failures than female birth control) and much more effective than female birth control (100% in most cases). I've been reading about it actively since I was 15 and now I'm 30 and I'm still reading about how it's 5 years away. I want this so bad it hurts, but no one is bringing it to market.

The problem is there isn't (enough of?) an incentive to actually put out these treatments. I guess the money just doesn't compare to female birth control.

Aloisius 3 days ago 1 reply      
For those who aren't interested in seeing a picture of a scrotum being operated on, the technique described is a polymer that is injected into the vas deferens. As sperm pass the polymer, their membranes and tails are damaged. The technique can be reversed using a solvent.
jcromartie 3 days ago 2 replies      
I have nothing to add except that reading the description of the process made me feel physically uncomfortable.
thasmin 3 days ago 4 replies      
A 100% effective, low maintenance, reversible birth control method could have an incredible effect on the future of the species. I'm guessing the benefits of fewer unwanted children will be incredible. The possible downsides include a dangerously low birth rate and forced temporary sterilization, but I think the long term effects will be very positive.
rmc 3 days ago 2 replies      
Will this cause an increase in sexually transmitted diseases? If straight men don't have to wear condoms for pregnancy-avoidance, and straight men & women don't like to pretend/boast that they've had lots of sex (unlike gay couples), then the pressure to go bareback (without a condom) increases and there might be more HIV/AIDS/STIs
spidaman 3 days ago 2 replies      
Yet another illustration about how broken the health care system is in the US. Pharmaceutical and medical device companies are disincented to produce cost effective treatments, it's really an outrage. Medicine in the US is geared towards costly treatments that are needed on a recurring basis; we're inundated with marketing and profit driven medicine. IMO, the NIH and FDA should be promoting a research-rewarded system that promotes cost effective and potentially society-shifting treatments such as RISUG.
grantbachman 3 days ago 3 replies      
I would sign up for this procedure in a heartbeat, but it still doesn't get around the (albeit not perfect) STD protection condoms can provide. If this takes off, people probably won't wear condoms most of the time, and the number of people being diagnosed with STD's will skyrocket.
mkempe 3 days ago 3 replies      
I have a two-year old daughter. I want more kids, not fewer. If I had known how amazing raising a kid was, I would have started sooner. Smart people need to bring more smart people to the world.
rrrazdan 2 days ago 0 replies      
>“If it's no longer a crazy Indian idea and it's something that's working in India and in rabbits in Ohio and in the first 20 men in the US,” Lissner says, “then there's got to be a point where there's just no excuse for a Gates or a Buffett not to get on board.”

The words "crazy Indian idea" hit me as offensive at first. But I guess that's just the way the world perceives us and its up to us to change that.

BasDirks 2 days ago 0 replies      
newobj 3 days ago 0 replies      
hello NSFW.
jonsantana 3 days ago 0 replies      
seeing any kind of operation done in that region I come to think a condom is actually a pretty good idea
hippich 3 days ago 1 reply      
Now if only they could invent something revolutionary for treating infertility...
KeyBoardG 3 days ago 0 replies      
There should be a NSFW in the link title.
armored 3 days ago 0 replies      
The video is excruciating. Make sure you watch it.
saool 3 days ago 1 reply      
Bonus points for reading the whole article without flinching!
bakbak 3 days ago 2 replies      
This is going to change the world ... this is by far going to have deepest sociological impact (good & bad) on the society ... because many men will use this secretly without telling their partner to avoid having kids ... which eventually will bring enormous decrease in population ...

Now someone should come out with novel technique to avoid STD without using condoms ... combination of these 2 techniques will be VIOLA !!!!! :)

Pixar: Getting over embarrassment protoshare.com
267 points by 16BitTons  4 days ago   26 comments top 10
alexqgb 4 days ago 1 reply      
That whole talk is a gem of the first order. Don't skip over it thinking it's JUST about iteration. The segment on what it means to keep your crises small is especially good. It's also the single best explanation I've seen as to why Pixar has never released a dud.

Meanwhile, their less enlightened, less capable, and frankly less humble competition just assumes that most movies will end up losing money, and that what REALLY matters is getting a handful of hits that can offset endemic failure.

chrisaycock 4 days ago 3 replies      
Summary of Ed's talk:

- Success hides problems. Big organizations fall slowly.

- Daily reviews. Everybody gets reviewed everyday; this removes the embarrassment factor so that no one waits until "it's perfect" before presenting.

- Communication structure is not the same as organizational structure. A firm might require a strict hierarchy to ensure control, but that doesn't mean everyone should be prevented from talking to anyone else.

- The only measurement that matters is whether the team functions together.

- Don't copy successful products, even if they're your own product. Either invent something new or fix an unsuccessful product.

lotharbot 4 days ago 0 replies      
The tendency to only show things "when they're done" means we only produce things we're able to complete without any help.

By getting over the embarrassment of showing someone a product that sucks or isn't complete, we position ourselves to get feedback and assistance -- which means we can make things that we couldn't have made without help.

Brashman 4 days ago 2 replies      
My issue with being evaluated very often is that I find myself fixing small things so I have something to show rather than working on bigger issues.

However, the "just do it" idea of the post is great.

ams6110 4 days ago 2 replies      
This is interesting, because I've heard at Apple you don't demo anything that is not as polished as you can make it. E.g. layout perfect, no "Lorum ipsum" text, everything must be thought-through and a realistic example use case.
hoopadoop 4 days ago 2 replies      
I worked at a world famous design studio. I was horrified to discover that there could be no 'brain storming' - everything you said or every drawing you showed would be judged and held against you. It made it almost impossible for someone to come up with a truly creative solution. You had to be very careful and measured about everything you said.
Wickk 4 days ago 0 replies      
Very important thing to get over. In concerns to programming: The majority of what you code will look like shit to most people. Who cares. Do it anyway and learn.
prayag 4 days ago 0 replies      
This is actually analogous to that we teach in User Interface Design 101. There are distinct advantages of showing very early stage prototypes. When we show it to potential users they would be perfectly honest about what they don't like. Try it with a very finished product and the users would be polite and not mention the issues they faced.

Always, always start with putting a very bad replication of the product in front of your users. Paper prototypes work very well.

paufernandez 4 days ago 0 replies      
In my case this is great advice, thanks for sharing.

I've been meaning to polish it but... I think I'm going to
put all my code in github now!

hxf148 4 days ago 0 replies      
Love this, get out there, prototype, iterate, try things. The longer you wait to "release" the longer it will take you to be "done". If ever.
GitHub: Block the Bullies github.com
250 points by remi  2 days ago   141 comments top 19
mycroftiv 2 days ago  replies      
This is a very good feature, but I am also afraid it will make Zed Shaw and others think that his attempt to out-troll the trolls and multiply internet drama was a force for positive change. "If you throw a big enough temper tantrum and incite a large enough shitstorm, developers will address your concerns" is an unfortunate precedent. The fact that controversy and negativity attracts eyeballs and can trigger improvements is understandable, but it also creates perverse incentives.
jz 2 days ago 3 replies      
Background: http://sheddingbikes.com/posts/1306816425.html

In all fairness, the guy was being a dick (no pun intended) to Zed. However, Zed should have kept the insults to the troll and left Github, Powerset, Engine Yard, and the Ruby community out.

getsat 2 days ago  replies      
Is this due to the recent drama with Zed Shaw being invited to the "DongML" project? Very entertaining as an observer, but very annoying for Zed I'm sure.

Sad that they still haven't added a requirement that you accept an invite to become a collaborator, or at least a profile setting on your account that requires that confirmation.

pufuwozu 2 days ago 0 replies      
This is GitHub's response to the (killed) article I posted yesterday:


dwlathrop 2 days ago 2 replies      
I think zed had a reasonable complaint. was he a jerk about it? maybe, but it's a customer's prerogative to complain about a bad experience. github's prompt response to zed has made me a happier (paying) customer. way to go!
16s 2 days ago 0 replies      
I'm glad Zed made this an issue. No one should be harassed like that. Honestly, it made github look very unprofessional. I'm glad they stepped-in and fixed it.
chmike 2 days ago 0 replies      
The implemented solution is impressively elegant. Keep the barrier very low to join another project and provide a mean to block undesirable subsciptions. There is still a possibility for harrasement with multiple projects, but there is a report to staff option now to solve it.

These are very smart, efficient and pragmatic solutions. It's also a demonstration of what is meant by "good execution".

code_duck 2 days ago 0 replies      
It's a logical feature for a site with a social aspect to offer, and I'm surprised github hadn't done so previously. I suppose the audience for the site tends to be mature.

Good for Github for going ahead and doing it. I know of some companies who would have dug their heels in and ignored the issue, or stubbornly maintained that it wasn't needed.

willvarfar 2 days ago 1 reply      
The profile page needs tidy up - if only it split the projects into tbose you collaborate on,and those you are invited to collaborate on. Its just wording, but it makes it clear that being added to a repo you haven't contributed to will not be shown publicly.
frou_dh 2 days ago 1 reply      
Zed just seems to react strongly to anyone who tries to intimidate him. Same deal with all the people called out in the classic Rails post. I don't fault him for it.
drivingmenuts 2 days ago 0 replies      
I'm surprised that this feature wasn't in there to begin with. All social networking (which is primarily what Github is about - the source control really isn't anything new) should include bilateral confirmation.

What they've implemented is adequate, but I'd rather see something much stronger, like active confirmation from both parties before being added to a project.

senthilnayagam 2 days ago 0 replies      
convenience and privacy are the issues, github has tried to do a balancing act.
rgbrgb 2 days ago 0 replies      
Ah, the age old enmity between nerds and bullies.
mdg 2 days ago 0 replies      
I flagged this.

It wasn't submitted to let people know about the feature, it was submitted to stir the pot.

Fester 2 days ago 0 replies      
Now Github became a real social network.
sondh 2 days ago 0 replies      
That was fast!
joyce_ampah 2 days ago 0 replies      
endlessvoid94 2 days ago 3 replies      
I think this was a good response to an attention whore.

EDIT: I wasn't trolling. Not even a single response?

illumen 2 days ago 0 replies      
I wish I could block zedshaw from all of the internet. I wrote a web proxy once which cuts out adverts - I think I could extend it to block zedshaw too. But my proxy is old, and had bugs with some webservers.

Maybe a zedblock plugin for firefox would be cool instead?

Even better would be to just integrate it into firefox. But then you'd still have to put up with the character whilst using other browsers.

Which means a WC3 draft would be more appropriate, so that all of the browsers could implement it. I imagine WHATWG have already got something in the works though. They've been doing a lot of good work with the whole html5 thing.

Does it bother anyone else that github appears to be following a Concerned father approach here? I guess it's not that bad. Other internet forums have moderators and such, but github isn't really about the project - but the individual. So I'm not sure how letting other people moderate for you would work within the github garden.


How to handle 1000s of concurrent users on a 360MB VPS markmaunder.com
246 points by joelg87  6 days ago   75 comments top 18
jrockway 6 days ago 3 replies      
Why use nginx instead of Varnish as the frontend proxy? If you are using nginx as your http <-> fastcgi gateway, that's one thing, but if you're just using it as a reverse proxy, it seems like a lot of extra stuff to maintain.

I like living on the bleeding edge, so I do varnish -> mongrel2 <-> apps instead... but varnish -> nginx <-> apps or varnish -> apache <-> apps seem equally reasonable. The key is caching and being able to handle slow users without tying up your app. All three of these setups do that. nginx to apache only saves your app server; it doesn't do much caching.

mmaunder 6 days ago 1 reply      
This is surreal. I just wrote a draft blog entry about the changes in startup economics due to event-servers like node and nginx (publishing tomorrow), and I hit HN just before heading to bed and this blog entry is #1.

Thanks for all the love today guys. Have an awesome memorial day weekend if you're in the States.

geuis 6 days ago 1 reply      
I use nginx exclusively as my server and I run node instances and wordpress behind it. Why even use apache?
jacques_chester 6 days ago 3 replies      
Missing: caching. Caching. Caching. And more caching.

I speak of Wordpress and database-backed PHP applications of its ilk. Wordpress loves to spray MySQL in a thick stew of queries for every single page load.

Without sensible -- preferably aggressive -- caching, your Wordpress site will die in the arse.

My setup generates cached .gz files, which nginx can serve directly. It flies. It didn't used to.

otterley 6 days ago 1 reply      
This technique naïvely assumes the only thing that is slow is the client. If a backend data provider that the web server communicates with (e.g., database) is also slow, this arrangement merely adds complexity without much corresponding benefit. The origin server will still block waiting on the data provider, which can cause process starvation in the pathological case.

Moreover, issues with slow clients often can be solved by raising the TCP send buffer size. As long as the response size is less than the send buffer size, it really doesn't matter how slow the client is: write() will return immediately, leaving the webserver free to serve the next request. Getting the data to the client then becomes the kernel's responsibility.

rkalla 5 days ago 0 replies      
As a good number of folks have pointed out, this article (helpful for sure) is basically nginx + PHP-FPM.

nginx servers up static content (e.g. if you are using WP-SuperCache and it's generating those static cache files for you, JS, CSS, images, etc.) then you configure a pool of warmed up PHP VM instances via PHP-FPM (Check /etc/php5/php-fpm or some equiv dir on your server).

Then you setup a rule that directed PHP requests to the FPM service, that would probably look something like this in nginx:

  location ~ \.php$ {
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME /var/www/mysite.com/$fastcgi_script_name;
include fastcgi_params;


and instead of adding the layer of Apache servicing the PHP requests, you are having nginx pass through the PHP requests directly to PHP processes to run.

Going from an Apache2 + FCGID configuration (somewhat similar to this, but with Apache) to nginx, I saw a 75% drop in server load.

I'm almost certain this still isn't a totally tweaked out setup that someone more familiar with this process could do better, but for my needs it took my server crashing under load no matter what I did, to typically being idle most of its life (~900k/month pageviews)

So I'm a big fan of nginx. I'm not saying you couldn't configure Apache to do the same... just 5 years of attempts to do so never got me anywhere with it.

jschuur 6 days ago 1 reply      
Worth noting the the story is from Dec '09.
stock_toaster 6 days ago 1 reply      
At the end of the article, why is apache even being used anymore?

Just use nginx and php-fpm (or php-cgi/fcgi).

muppetman 6 days ago 2 replies      
You can do the same on a really high traffic Drupal site using the boost module. It's basically a module that will, for non-logged in users, generate a static html page and serve that instead. PHP isn't even called, the .htaccess just serves up the cached content.

And that's using Apache!

ericb 6 days ago 0 replies      
Dropping keepalives actually slightly hurts your best-case performance. As a config change, if you are using the prefork mpm (I wouldn't if scalability is your goal) it will generally help you though. With prefork the scaling math on worker memory * connections vs your box's memory is often the first place things will fall apart if you have keepalives on and real traffic.

I suggest the worker mpm if you don't want to switch to nginx.

gsharma 6 days ago 1 reply      
Repost: http://news.ycombinator.com/item?id=970682

Interesting comments on the original thread.

chopsueyar 6 days ago 1 reply      
Are the loopback requests limited by the TCP stack?

Is it possible to set this up where nginx and apache communicate via a socket?

james4k 6 days ago 1 reply      
I just find it odd that Apache hadn't employed the likes of epoll already. IRC servers have been using async sockets like this for yeeeears.
peterwwillis 5 days ago 0 replies      
Hint: you can just run Apache twice; once with a mpm_prefork config for your apps, once with a mpm_event config to handle client connections.
mise 6 days ago 1 reply      
I've been on Apache servers for years, so I come from that frame of mind. My managed VPSs are running Apache, and one of them is running nginx for static resources. I think my managing host (Servint) doesn't even support nginx. How, for argument's sake, would I move my PHP/MySQL sites to run on a managed nginx server?
iphoneedbot 6 days ago 1 reply      
This approach really works well... I would add that off-loading MySQL to a different VPS within the same network works well too; latency within the same network is very negligible. Doing simple optimization such as moving/disabling mail and stuff you probably dont need works too.
mp3geek 5 days ago 0 replies      
I use straight nginx on Fanboy Adblock (which gets hundreds of SSL connections per minute), no need for Apache.
nwmcsween 6 days ago 5 replies      
nginx is actually pretty bad at all benchmarks I've thrown at it, you can see some here: http://www.webhostingtalk.com/showthread.php?t=918355 this was on an 8 core intel Xeon 5520 server system it maxed 8 cores (18ghz) @ 7500 requests per second for static content.
Airbnb Has Arrived: Raising Mega-Round at a $1 Billion+ Valuation techcrunch.com
246 points by sammville  4 days ago   112 comments top 24
ChuckFrank 4 days ago  replies      
I've been waiting for some time to call out Airbnb, and I think this is the right time.

Firstly, the idea of short term residential rentals is not new. Long before Airbnb major cities, Paris, London, New York, and Toronto were dealing with the issues and the problems of short term / nightly residential rentals. In fact the latest condo boom in Toronto has been ascribed to international investors buying condos explicitly for this quasi-hotel rental market. Many cities have been passing local laws trying to prohibit this type of rental use for residences because they carry a whole host of problems. Some of those include -- emptying the downtown core of residences, artificially increasing housing costs for residents, removing the stability and security that comes from having permanent and community-invested residents, shorting cities on their hotel taxes, providing unlicensed locations for short term illicit activities, etc. So by increasing the market in these short term rentals, the communities will also feel the increase in the unfortunate side effects of these short term rental impacts. These are long term market problems that Airbnb will have to address.

Secondly, it's not a new idea because in the recent past, these rentals were managed by hotel brokers. So in that respect, Airbnb major idea is simply to replace the traditional broker that used to handle the short term rentals with a direct owner to renter model. Across the internet we see great success in online services removing brokerage inefficiencies to various markets, and for this they should be congratulated, but not overly so.

And finally, about their execution. Airbnb is not the market leader. In fact --http://www.vrbo.com/ is. This is what amazes me the most. Vrbo (I have no professional relationship or otherwise) regularly get better reviews in online discussion boards comparing the two services. Their price structure is much more competitive, their review structure is much more robust, in fact aside from Airbnb's more sophisticated visual design, Vrbo leads in many segment of this space.

So, to what do I think Airbnb's success comes from? I think it's Airbnb's marketing / hype / buzz that has been carefully choreographed to maximize their valuation. And to this, I congratulate Airbnb on their latest success, and to a dance well done.

olivercameron 4 days ago 4 replies      
I think a lot of startups put too much weight on what an investor thinks of their idea. Airbnb is a prime example of carrying on in the face of rejection, even if people are saying "it's a stupid idea". I know a lot of founders who, after being told by many of the Valley's finest angels that their idea is no good, would just give up.

It just goes to show that you can create a $1 billion company, even if no one really gets it in the beginning, and in my eyes at least, they are actually justifying it (making a lot of money in a lot of different places).

JacobAldridge 4 days ago 1 reply      
Fred Wilson was one of the VCs who passed on his opportunity to invest in Airbnb - his musings on that opportunity, and learning from it, are well worth a read - http://www.avc.com/a_vc/2011/03/airbnb.html
lionhearted 4 days ago 0 replies      
Congratulations to everyone involved. This is way cool. Super inspiring.
stevenj 4 days ago 0 replies      
"Don't worry about people stealing your ideas. If your ideas are any good, you'll have to ram them down people's throats."

-Howard Aiken


alex_c 4 days ago 2 replies      
Congrats to AirBnB!

The way I understand it, the biggest threat to Airbnb's model are local governments starting to clamp down on this kind of (usually unlicensed?) subletting. I'm curious how the high valuation factors this risk.

Of course, I suspect the high valuation means Airbnb is planning to expand to hotels and travel in general?

rst 4 days ago 2 replies      
Techcrunch says, "On any given night in New York there are more people staying in homes via Airbnb than there are rooms in the biggest hotel in Manhattan." For comparison: Starwood Hotels (whose brand names include Westin, W, Sheraton, and others) has more than 20 hotels in New York (some quite large), and a market cap of about $11.3 billion.

FWIW, Airbnb's commissions are 9-15% of the rate (3% from the host, 6-12% from the guest) according to their hosting FAQ[1]; no idea how that compares to typical hotel profits on a room listing...

[1] http://www.airbnb.com/help/topic/hosting

jhuckestein 4 days ago 0 replies      
Airbnb is in a prime position to expand into the market of web-based PMS (I'm not kidding -- Property Management Systems) that are super expensive and still run on old MS-DOS machines in many hotels. Give hotels a web-based solution that interfaces with hotels.com and other sites (there's a rather involved protocol that you need to license, but the name escapes me) and you're in business.
mrchess 4 days ago 2 replies      
This might be a stupid question but what do you do after you secure such a large series of funding? Go on mass expansion? Would anyone mind telling me how startups typically use this kind of money (closing Series with 100+mil)?
markbao 4 days ago 0 replies      
Huge congratulations to Airbnb. That's incredible.
tmsh 3 days ago 0 replies      
That's true. It's also true that there are quite a few marketplaces out there that serve this same market


The difference is execution. Any skilled designer (and I'm not a designer but I can respect design) would've looked at airbnb.com in the past couple of months and realized: this is different. The way Facebook was different. All those key design decisions add up and you get enough of an edge over everything else that you redefine the whole sector by making the sector suddenly interesting.

Anyway, I'm biased. I stayed at an Airbnb place over this weekend and it worked great. It's bigger than EBay though, for my money. It's next generation lodging that is OOM more efficient and the flood gates are probably going to start overflooding (economy seems just right with enough consumer spending to support some trips, but enough of a groupon-appreciating eye, for more and more people, to avoid hotels where possible, etc.).

I sent in my resume to airbnb after deciding to try the site (while booking a place for the night before Bay to Breakers). I didn't actually stay at a place (until this past weekend) -- but I knew that whoever created that site was way ahead of everyone. I found the answer to one of their programming challenges on the interwebs and e-mailed about it (so haven't heard back -- I expect they are pretty swarmed with interest now too). And honestly, I'm at the point where I'm not that into working for a successful startup (would just confuse things for my side projects, etc.). But I am a very happy customer. And Brian Chesky is a new kind of entrepreneurial genius (his Startup School presentation still reverberates). Cheers.

tpatke 4 days ago 5 replies      
I bet the guys behind CouchSurfing are kicking themselves about now.

Always keep a backup of your database! :-) In this case, I think it is pretty clear the execution is worth a heck of a lot more than the idea.

nl 3 days ago 0 replies      
Random financial speculation:

Around 5000 rooms a night get rented (via the video).

The make an average of ~10% per transaction (3% from the host, the rest from the guest: http://www.airbnb.com/help/topic/hosting)

Say an average of $60/night.

60 * 0.1 * 5000 = $30,000 revenue/day.

That's pretty decent money.

jpiasetz 4 days ago 1 reply      
I think this is the blog post from 2007 he references in the video http://www.core77.com/blog/events/airbed_breakfast_for_conne...
nostromo 4 days ago 1 reply      
I'm curious how well Airbnb is doing outside of New York. New York's real-estate market is so out of whack it strikes me as fertile ground for p2p hotel rooms.

One example: I wonder how many people are renting out rent-controlled apartments on Airbnb at a huge profit, while the owner maintains the unit at a loss?

derrida 3 days ago 0 replies      
For all the talk of bubbles, I think AirBnb is actually undervalued at $1billion. I want this noted. Before AirBnb space was being wasted. Now its being used. That is something of fundamental value. The dubious valuations are things like FourSquare, LinkedIn and GroupOn.
st3fan 4 days ago 1 reply      
What is their revenue? How much profit are these guys making? How big is their team?
pbreit 3 days ago 0 replies      
It will be interesting to see how AirBnB maintains the listing quality as it scales which I would argue is currently its main advantage over VRBO, HomeAway, Craigslist, et al. These types of sites tend to devolve into a homogeny of bland, "professional" listings, hence HomeAway's boring rollup strategy. What has made AirBnB somewhat unique and appealing has been the interesting listings and the care and personality of the property owners.
savrajsingh 4 days ago 0 replies      
Congrats guys! Great interview, Brian!
naeem 3 days ago 0 replies      
All things considered, this is definitely fantastic. I was watching the TechCrunch TV video of Ashton Kutcher earlier today about his surprise (at the time) investment of Air BNB. Looks like this guy really knows what he's doing when it comes to investing - skype, foursquare, not abnb. Pretty cool.

Anyway, congrats to the abnb team, much deserved!

hxf148 4 days ago 0 replies      
Wow, congratulations.
patrickgzill 4 days ago 1 reply      
Apparently $1B doesn't get you as much as it used to...
creativeone 3 days ago 0 replies      
What's wrong with craigslist's vacation rentals??
forensic 4 days ago 0 replies      
When tech companies raise billions, it's a bubble.

When hedge fund CEOs are paid $600 million in salary... it's a recession.

Congrats to AirBnb and YC on joining the top .001% :]

Visualisation of Machine Learning Algorithms epfl.ch
245 points by maurits  4 days ago   11 comments top 5
vedantk 4 days ago 3 replies      
The install.sh and install_script.sh scripts are Mac-dependent. How would you compile this on Linux? Qmake?

edit: Yep. Clone the git repo, then run qmake, followed by make.

edit #2: Here is a patch I wrote to make the project compile under OpenCV 2.2, and Linux: https://gist.github.com/999061

vinyl 4 days ago 0 replies      
Wow. This could prove an invaluable tool for teaching. Great UI (love the way you can build the dataset with a paintbrush), top-quality visualization, wide choice of algorithms... Much better than R scripts if you want to show basic algorithms at work to a bunch of students. Many thanks
jwr 3 days ago 0 replies      
Thank you! This is very, very useful. Not only to beginners, but also to those of us who want to roughly compare the behavior of several algorithms. You can estimate things like sensitivity to parameters and quickly visualize exactly how they behave.

It's a very useful tool.

Emore 4 days ago 0 replies      
Here's another one, a simple Applet that lets you apply various classification algorithms on 2D data: http://www.cs.technion.ac.il/~rani/LocBoost/
maeon3 4 days ago 0 replies      
Patch through a black and white video into this tool and watch how the algorithm tries to find patterns in a 2d reprsentation of a 3d space. Maybe it can predict where the table ends without full view of the table.

Neat stuff, it is the beginning of machine notion of object permanence. Filling in the gaps and making educated guesses about what kind of function controls where a sequence originated from.

Linux 3.0-rc1 is here kernel.org
242 points by angusgr  5 days ago   77 comments top 16
riobard 5 days ago 3 replies      
For those wondering what's new:

“So what are the big changes?

NOTHING. Absolutely nothing. Sure, we have the usual two thirds driver
changes, and a lot of random fixes, but the point is that 3.0 is
just about renumbering, we are very much not doing a KDE-4 or a
Gnome-3 here. No breakage, no special scary new features, nothing at
all like that. We've been doing time-based releases for many years
now, this is in no way about features. If you want an excuse for the
renumbering, you really should look at the time-based one ("20 years")

by Linus Torvalds

source: ttp://permalink.gmane.org/gmane.linux.kernel/1147415

timf 5 days ago 1 reply      
Linus' email is more interesting than the commit: http://permalink.gmane.org/gmane.linux.kernel/1147415
tzury 4 days ago 1 reply      
Despite all criticism and cynicism all over the Linux communities about the numbering and all that stuff, for me, every release of the Kernel (as well as any other major / dominant open source platform / project) is a reason for celebration.

It simply means, openness and freedom won the software/internet game.
The fact there are people out there, spend the best of their times, contributing code and manuals docs, debugging and filing bugs, etc. etc. Is a sign that RMS and alike were not _"a bunch of hippies who likes to code for free"_ or even worst, a bunch of communists as some used to say at the beginning of Linux breakthrough.

From my own personal experience, it also means, the more open you will be, the more open software you will rely on, the more money you will make by the end of the day.

Thank you Linus, and all kernel contributors, for the great tools and platform you provided us for FREE!

jordan0day 4 days ago 5 replies      
One the one hand, the "benevolent dictator" model has shown itself to work pretty well for Linux, compared to some major failings of the "committee" model on other projects. On the other hand, does anyone think the fact that "major" changes can be done by one person who is "just going all alpha-male" will potentially spook people? I mean, I doubt too many people here on hn have a problem with Linus' personality. I imagine most of us appreciate his sense of humor and respect him as a penultimate hacker... I'm just thinking about starched-shirt types who call the shots in big businesses -- can you imagine being an IT manager and trying to explain to your boss that after 15 years there's a new major version of the Linux kernel because "well the head guy got tired of calling it version 2."

Edit: Penultimate apparently doesn't mean what I thought it meant, so, what I meant was "...respect him as a very good hacker...".

lpgauth 5 days ago 0 replies      
icco 4 days ago 0 replies      
My favorite part is the commit message on the tag: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6...

"Version numbers? We can increment them!" -- Linus Torvalds

tybris 4 days ago 1 reply      

  -VERSION = 2
-NAME = Flesh-Eating Bats with Fangs
+NAME = Sneaky Weasel

jsaxton86 5 days ago 0 replies      
A summary of the changes can be found here: https://lkml.org/lkml/2011/5/29/204
BruceForth 4 days ago 0 replies      
So, when will Linus be released from mental institution? [0]


autalpha 4 days ago 0 replies      
I don't know why, but I always get a good chuckle or two from reading or watching Linus' materials.
FlowerPower 4 days ago 0 replies      
Its name was Sneaky Weasel.
uabbasi 5 days ago 0 replies      
kbd 5 days ago 1 reply      
Wow, he said he might go to 2.8. This is madness.
aclark 4 days ago 3 replies      
I know they invented git, but that website sure is fugly. Time to switch to github, perhaps :-).
haruka 4 days ago 3 replies      
Seems a bit strange to me. A major version increment is supposed to announce "this is significantly different and possibly incompatible".

But now it doesn't mean that. It means 20 years.

There are extremely good reasons not to increment the major version. For instance tools which take the conservative approach of checking for 2.6 will all be broken. Those tools correctly elected not to risk wrong behavior resulting from a major version change.

When you add together the time it takes to compensate for this change for all maintainers everywhere (tools developers, distro developers, everyone), surely it will run into the hundreds of hours, perhaps thousands. All for, functionally, nothing. And sorry for calling you Shirley.

Seems to be an objectively bad move. Oh well. This needless-on-purpose change will cause resentment, but after time it will dissipate.

The Quora post that killed Bitcoins. Please discuss if his arguments are valid. qr.ae
228 points by andreasklinger  1 day ago   301 comments top 40
mycroftiv 1 day ago  replies      
The current design of Bitcoin is obviously not correct for the long term, and I am sure it will be supplanted by more evolved crypto-currency systems. There is absolutely nothing preventing people from taking the open-source code and creating an alternative "Bittoken" currency which changes problematic design elements such as the minting/issuance model.

Given the rewards that have already accrued to Bitcoin early adopters, I am surprised other competing cryptocurrency networks have not yet already been created. Lots of people who wish they had started mining coins back in the days before GPU mining would be eager to get in on the ground floor of an alternative.

Most problems with the critiques of Bitcoin like those in the linked Quora post is that they are debunking a straw man - the idea that Bitcoin is intended as a universal replacement for all other currency systems. By those standards of course it will fail. Looked at in another way, think of Bitcoin as something more like MMORPG loot. If people are willing to spend vast amounts of "real money" on items for their WoW character, spending that money on Bitcoins instead makes it seem like a sane investment rather than just flushing it away.

I've mentioned before that I think governmental action to criminalize Bitcoin and analogous technology is an even more serious threat to long-term value. If there is one thing that really might make the world's governments team up to completely ban strong encryption, it is a potential threat to governmental control over taxation and currencies.

acslater00 1 day ago 9 replies      
Hi, I'm Adam, I wrote this answer on Quora. Against my better judgement, I'm going to wade in here and make a couple of clarifying points.

1 - Scam. I regret using the word "scam" in my answer, because I think it overshadows the broader point I was making about Bitcoin's severe structural design flaws. After watching the community react to this debate, I'm pretty confident that the average bitcoin supporter is not intentionally defrauding people. Consider my inflammatory rhetoric redacted and apologized for.

2 - Volatility. Since April 1, the average change (up or down) of the bitcoin/dollar exchange rate has been just under 8%. That's unprecedented. It's a clean order of magnitude more volatile than the stock market, which is considered a "volatile", "risky" class of asset. It's two orders of magnitude more volatile than a legitimate currency pair like the Dollar/Euro. Now, it's all fun and games while the exchange rate is basically trending up, but if that trend reverses look out below.

3 - Liquidity. All of that volatility comes on an average bitcoin volume of something like 20,000 per day. If I put in an order for $15,000 worth of Bitcoin right now, I'm pretty sure I could move the market between 5% and 10% to the upside. It's probably even worse to the downside. Call me a skeptic, but with that kind of easy price manipulation, I'm not quite ready to denominate my paycheck in Bitcoins quite yet.

4 - What It Means. A number of people criticized me for not understanding that Bitcoin is designed to be complementary to traditional currencies, and that analyzing it as an alternative is invalid. A number of different people criticized me for not understanding that Bitcoin is a different kind of currency, which will inevitably replace traditional fiat money, and that analyzing it as a component of the fiat currency system is invalid.

The Bitcoin community is diverse, and no one really knows what this is supposed to be yet. That's totally fine with me. I can assure you I don't have any skin in this game. My money is in LinkedIn stock.* But I've studied world currencies extensively (and trust me, a normal currency debate is not this lively). So, I don't know what Bitcoin will end up as, but I know two things with certainty:

a) that it's unstable as a currency

b) that it's currently behaving as a speculative vehicle in an aggressive bubble

* That was a joke

cturner 1 day ago  replies      
Trollish headline.

One fair non-structural criticism, a weak, temporary criticism and two things billed as criticisms that aren't.

1) Early adopters. I couldn't find a point here. I think he's saying that it's unfair that some people get in early. But in the current system, it's unfair that government (through tax law and control of force) mandates that you use a form of money that doesn't function well as a store of value. Bitcoin wins here - it has a bootstrap disadvantage, and I'd expect that to diffuse.

2) Deflation. There's a creed in mainstream economics that deflation is bad, repeated here. "if your money is getting predictably more valuable, why would you want to spend it?" It's just false. Would you buy a computer today if you can buy a more powerful one by waiting until tomorrow? People do. Mainstreamers like the leaves-on-the-fire effect of inflation and also like to overlook the fact that stimlus impairs cyclic corrections and detracts from the role of money as a store of value.

3) Convertability. It's valid to criticise the use of bitcoins on this basis, but it's not a systemic problem. If we had a period of high inflation, it's reasonable to expect people would trust finite-supply bitcoins over an unending supply of newly-minted, inflating notes.

4) "When Something Goes Wrong, It Will Die." I found the logic here non-sequiter. Bitcoin will be tested but from what we know about it at the moment, it looks solid.

nl 1 day ago 2 replies      
His arguments are valid to some extent. They may not be fatal though. Taking each one in turn:

1) Seeding Initial Wealth

This isn't really a problem. It's not very fair, but I
don't think he argues that is actually a problem.

2) Built in Deflation

This is a big, big problem. It seems the creators of Bitcoin have a philosophical disagreement with conventional economic theories that state that increases in money supply (and some limited inflation) are outcomes of a healthy economy. Unconventional thinking is fine, but the outcomes he predicts here are real problems.

3) Lack of Convertibility

Yes, there is a problem here. It could actually be worse than the author suggests, because if state actors move against Bitcoin they could easily introduce high penalties for Bitcoin conversion. Even without that state actors, the dependability and predictability of convertibility is currently unacceptable.

At the same time, it's possible this could be overcome. It is basically a matter of trust, and modern fiat currencies all rely on trust. Usually that is trust in a government (or system of governments in the case of the euro), but it isn't inconceivable that "the internet" could be as trustworthy as a government. People trust billions of dollars or transactions to operating systems and databases that were built by "the internet", so it is possible this could be the same.

4) When Something Goes Wrong, It Will Die

This is the trust issue again.

However, there is one additional point - if someone wanted to deliberately attack the currency it would be fairly easy to buy a large number of Bitcoins, then deliberately destroy them. Once that occurs, they disappear from circulation, which causes instant deflation.

dogas 1 day ago 8 replies      
I think the other article (http://apenwarr.ca/log/?m=201105) written about bitcoin failing is more detailed and contains potentially more plausible ways of how bitcoin can fail. The openwarr article refers to potential failures with the SHA256 backbone of bitcoins.

"With bitcoin, a single failure of the cryptosystem could result in an utter collapse of the entire financial network. Unlimited inflation. Fake transactions. People not getting paid when they thought they were getting paid. And the perpetrators of the attack would make so much money, so fast, that they could apply their fraud at Internet Scale on Internet Time.)"

IANA cryptography expert. Is this feasible in the way the author predicts? Could SHA256 be cracked quickly, a la MD5? My gut is that it won't be able to, but I can't back up my argument.

vladd 1 day ago 3 replies      
Bitcoin is a marketing-based currency in the sense that it has no real value besides the fact that it was the first currency with cryptographic properties that gained widespread adoption: if someone would take the code and fork it into an alternative ("Webcoins"), the intrinsec value of a Webcoin would be zero (unless it also found a way to gain market share or early adopters).

That's very similar with a Ponzi scheme, where you need early adopters to pour money in, so you can have a positive cashflow on which to iterate with other customers. But otherwise you don't have an underlying asset to sustain the value.

Vivtek 1 day ago 3 replies      
It's a pyramid scheme, just dressed up in high-tech, vaguely libertarian clothing. That's why sensible governments (not the "bank lobby") are starting to ban it - it's inherently a bad idea (unless you're an early adopter - just like any pyramid scheme).

Bitcoins are no different from tulip bulbs.

TerraHertz 1 day ago 2 replies      
I've only recently begun paying attention to Bitcoin, and am still evaluating. Have yet to read the technical details, for instance. So I have nothing to say on the validity of the technology.
However I can offer some comments from a political perspective.

I find the timing of Bitcoin's appearance rather interesting. Here we are, entering into a whole mess of interconnected monetary crises -
* Crisis of confidence in the ubiquitous fiat currency systems, with ALL the world's currencies in an inflationary race to the bottom. A race that will very likely blow up into hyperinflation on the way to demonetization, followed by some as-yet unknown replacement monetary system.
* Crisis of legitimacy of the US Federal Reserve Bank. Which is neither federal, or a reserve, or really even a bank. It's actually a cartel of private bankers; many would say a criminal cartel.
* An ongoing raging battle between fiat currency in general (unbacked by anything other than 'faith') vs traditional stores of wealth - gold and silver. The forward trenches of this battle lie in the bullion vaults of COMEX and the LBMA, and the bullion holdings of the SLV and GLD funds. Right now, the Silver Liberation Army fights to expose the current economic order as the paper illusion it certainly is. When COMEX inevitably defaults (because they have less than one hundredth the actual metal they should have to back all the paper silver and gold they have sold over the years to suppress precious metals prices), then silver and gold will suddenly resume their rightfull place as the only reliable, inviolable store of value. And thus actual currency, or backing for currency.

Now, there's a method for manipulation of mass psychology called 'well poisoning'. It goes like this. If you are in a position of power and control, and you become aware of a concept or information that if it became widely accepted could threaten your position, you 'poison the well' from which that concept could rise into public awareness. You do this by introducing very similar concepts, but all with fatal attachments or flaws. This conditions people to automatically reject anything in that whole class of concepts. For example, if you know someone has gone to Kenya and obtained a copy of a genuine birth certificate of interest, you whip up a series of obviously faked 'Kenyan birth certificates' and saturate the opposition media with them. Result: no one wants to look at any more damned fake birth certificates, especially not one that's being offered on ebay. Heck, a while later you can even officially release your own grossly fake certificate, and still no one pays attention!

So, supposing you are the rulers of the global fiat monetary/banking system, and being able to print as much of that virtual funny money as you like is working very nicely for you. Then a small problem arises, as the fundamental systemic instability of 'never enough money to pay it all back plus interest' has unprotected sex with assorted hideous creatures from the banking black lagoon (derivatives, credit default swaps, collateralised debt obligations, mortgage/title/MERS nightmare.) All of a sudden you are up to your ears in Hellspawn like massive naked short positions on precious metals, PIIGS running wild and Vikings seeking banker blood, mark-to-market guillotines, the head of the august International Maid Fxckers organization casting aspersions about the gold in Fort Knox (or not), and your global reserve currency about to go pooof!

What to do, what to do?

Worst of all, there's this damned idea of a non-fiat currency raising its head again. Curses, you thought you had that one permanently dead, thanks to that fast deal you did back in 1913, after a private chat among friends on Jekyll Island. And why _shouldn't_ your banks create the national currency and lend it at interest to governments and the peasants? Stupid idea that Constitutional rubbish, about the government issuing silver money, interest free. Ugh! Where's the profit in that? Not to mention the inability to steal virtually everything over time via inflation and unpayable compounding interest on the entire monetary base.

However... you and your banking mates haven't quite finished God's Work of stealing absolutely all assets from everyone else. Needs a bit more time. But here's this bloody idea of a specie-based currency popping up again all over. States declaring silver coins valid as money, what next FFS! Time for some well poisoning! And fast!

What you (as an Elite banker) want to do, is set up something that is going to badly burn all the early adopters of alternate currency. Something that will make them wish they'd never seen or heard or even dreamed of anything but nice safe paper dollars. Distract them asap from any thoughts of buying (shudder) actual physical silver and gold. Definitely you want to minimize the numbers of these ... financial terrorists ... who end up holding an ounce of physical silver or gold in their hand for the first time and having that no-return moment of 'AH HA!' where they suddenly, at gut level understand what _real_ money is.

Also preferably something that will brand them on the forehead as currency outlaws, all the better to round them up and send to the camps if it comes to that.

Enter, stage left... Bitcoin. Riding in to save the day.

I don't think so.

There are a few other concerns I have, besides the awfully suspicious coincidence of timing.
Firstly, there's this thing about digital patterns and copyright, intellectual property rights and so on. I'm solidly in the camp that says you can't own information, and efforts to legally enforce ownership of information are fundamentally incompatible with deep principles of the Universe. Data is like Time, Matter, Energy and Space - it's a component of reality, with its own unalterable properties. One of those properties is that it can be duplicated indefinitely, without data loss. It can also be very easily destroyed, leaving nothing.
It's a huge topic of course, with sub-issues like secrets, lies vs truth, cryptography, the difference between data and knowledge, the philosophy of data/knowledge sharing and its social benefits, etc.

But as it relates to Bitcoin, I find myself very, very uncomfortable with the concept of founding a medium of exchange on pure data - both infinitely copyable, and utterly ephemeral.
Regardless of the soundness of the cryptographic methods, it sounds to me like asking for trouble. And that trouble might possibly have been intended from the start. It might hurt, a lot. Particularly if it's the fiat-banker ancien régime holding the other end of the cane, and it's you with your Patriot Act pants around your ankles.

Then there's a few more practical matters, that one shouldn't overlook these days.

* How do you bury Bitcoins in your backyard or out in the bush somewhere? (Silver & gold - no problem.)

* More to the point, even if you can somehow bury them, how will they be any use when someone digs them up in 10, 50 or 200 years? (Silver and gold - no problem!)

* When the police, BATF and FBI break down your door without a warrant, trash your place and take your computer, do your Bitcoins go with it? Ditto for burglars without badges and guns. (Silver and gold - at least you have some chance of hiding them.)

* TSA, airports, laptops and latex gloves. Where do Bitcoins go in this scene?

* Is it possible to melt down Bitcoins and cast bullets from them, and does Bitcoin ammo work against zombies, vampires and other flesh eating ravenous menaces?

* That '21 million Bitcoins max' figure. What?! Regardless of how that limit is set, and how dollar-Bitcoin exchange rates are determined, that figure tells me this was never intended to be a real currency, not even for one small nation. Less than one Bitcoin per person? Huh?
Bitcoin was clearly designed as a 'crippleware demo', as opposed to a workable system. But a demo of what? Of pain-bringing, I suspect.
(Silver and gold - there's enough to serve as a global currency. That there are two kinds, with about a 17:1 natural abundance ratio helps a lot too.)

* Quantum computers. Are now commercially available. How will this affect Bitcoin?

* Wishful thinking. Yes, we the people of the world do desperately need some medium of exchange that isn't owned by the banker and government flesh eating monsters. It would be great if it had all the good features of Bitcoin - electronic, anonymous, untraceable, unstoppable, untaxable transactions.

This doesn't mean we should leap joyfully at Bitcoin. Take a very close look, for strings, hooks, bear traps, punji-pits, etc. It's a mean world, and there are powerful, wiley forces who'll do anything, ANYTHING to preserve the existing fiat banking system a while longer.

richcollins 1 day ago 0 replies      
Question: if your money is getting predictably more valuable, why would you want to spend it? Answer: marginally speaking, you wouldn't.

This argument never made any sense to me. You can't eat money, watch it for entertainment ... etc. You might be more inclined to save money that you would otherwise have spent carelessly to avoid its loss of value, but how is this a bad thing? Wouldn't our economy be stronger if the effort we spent was on making things people actually want / need instead of making things people kinda want because their money is burning a hole in their pocket?

ap22213 1 day ago 3 replies      
I would love to see a currency like this work, and I would be interested in using one, but the major issue for bitcoin, for me, is that deflation is built into its core. Inflation is a good thing.

Inflation encourages holders of capital to use it to invest in other things. That is, inflation encourages productivity. The inflation has to be relatively low, of course. Probably low enough to allow someone to retire without worry.

lorax 1 day ago 3 replies      
Talk about an overly dramatic headline, the "post that killed Bitcoins"? but the submitter isn't even sure if the arguments are valid? (And has hacker news turned into bitcoin news? it seems like there are several bitcoin links a day
njharman 1 day ago 0 replies      
Not a fan of bit coin. Convertibility is the only issue in author's diatribe. Author has not understood several concepts re bit coin.

Bc is not a normal currency, it's not reserve currency. It's much more like a stock that's tradable without exchange. Limited quantity, market set price,

Mining is not free. When something succeeds early adopters always mKe out big. It's their payoff for risk.

wintersFright 1 day ago 3 replies      
As an zeroHedge reading, armchair economist i think there are a lot of flaws in that guys argument

"Bitcoin does not have a central bank capable of printing and lending bitcoins"

>>I rather think thats the point. Lending is inflationary.

"Built in Deflation"

>> God knows how the computer industry makes money when if i hold my $$$ just a little longer, i can buy an even better machine.

"For Bitcoin to work as a currency, it would have to act as a predictable store of value"

>>No, currency does not have to store value. The USD is a terrible store of value. Gold is a store of value. Govt paper is not.

" That's called a bank run, and for obvious reasons we want to avoid them."

>> um - I'm not an expert on bitcoin but there are no banks i thought. your money is in your e-wallet. you dont need a bank to keep your money away from theives. there are no banks with fractional reserve lending that are fragile to a run on deposits

This guy is dunning kruger at its best.

mchusma 1 day ago 0 replies      
I think each of these arguments do not hold up. Most people have already discussed but I will summarize.

Severe Problem Number 1: Seeding Initial Wealth
This problem is similar to Gold Miners or anyone initially to an asset class. You already retracted your scam comment, so thanks. Remember, governments are currently seeding banks and itself with wealth through inflation. Its a different set of winners based upon market forces.

Severe Problem Number 2: Built in Deflation

The money supply is actually inflating (slowly). People will make educated, market based choices about how to act with their money once the currency settles down. That will likely be assigning higher value over time to Bitcoin. The Euro or Dollar have the opposite, and many would argue much worse problem: you don't want to hold them because the issuing government slowly saps value through effectively built in inflation.

Severe Problem Number 3: Lack of Convertibility
Try trading government currency in a collapse situation, you will encounter problems. Right now, Bitcoins are roughly as easy to convert as any other currency. Bitcoins are inheritly no more or less convertable except for the lack of current market makers (a good point you brought up).

Severe Problem Number 4: When Something Goes Wrong, It Will Die
Look at Zimbabwe to see an example of governmental currency collapse. Any currency can die. People used seashells at one point, they died. I'm not sure Lyra or Franks are worth anything today.

Bottom line: Yes, Bitcoin can fail (so can any other currency). It is earlier than other currencies in its lifecycle, so has a different set of risks. Is there speculation going on right now, almost surely. Bitcoins appeal to people because it is not government based, electronic, and low cost. Many of the problems you note are just different problems than existing currency, for better or worse.

jcr 1 day ago 2 replies      
> When the federal reserve "prints money", it doesn't just mail million-dollar checks to random Americans. It does one of two things. It either (a) purchases some other asset [generally US treasury bonds] on the free market, thereby injecting more cash into the system than there had been before, or (b), loans money to a bank, who will then loan it to other people who will then spend it.

The part (a) about "purchases" is not specifically incorrect, but it is misleading and shows the author may not completely understand how things work. Typically the federal reserve SELLS treasury bonds. They are sold for less than face value with a promise to pay the full value on the redemption date. The difference, of course, is the yield, or more commonly the "interest" earned by holding the bond to maturity.

rglover 1 day ago 2 replies      
The problem that Bitcoins will have in gaining traction is inherent in this post. Experienced economists and financial experts will stomp it into the ground. Granted, this post made a lot of strong points, however, it was a bit of a witch hunt at the same time. To call Bitcoins a scam outright is the sign of a person who may not have tried putting money into the system to see how it actually works. As with anything new that might present problems, I guess you'll just need to practice caution. Bitcoin isn't dead because of this, though.
richcollins 1 day ago 0 replies      
But Bitcoin is not designed to be a functioning currency, it's designed to enrich early adopters. Again, that is why it is a scam. Period.

How is this different from our current currencies, which first go to banks that then use fractional reserve to lend out money that they don't have. At least with Bitcoin anyone can mine them.

jongraehl 1 day ago 0 replies      
All that matters it the ratio of nominal risk-free return to inflation. I guess nobody offers depositors of bitcoins a risk-free return in bitcoins.

If there's a positive risk-free interest rate available to depositors, and inflation is less than X, then you effectively have deflation already. TIPS sometimes offer this for $USD (depending on demand and whether the measure of inflation they're indexed to is fair). Nominal price stability is a psychological benefit only (albeit important).

peteretep 1 day ago 1 reply      
These points are weak - there are other weaknesses of BitCoin, but these aren't really them. Almost every issue he raises attacks BitCoin as a medium of exchange (which it doesn't need to be) rather than as a medium for storing value. Taking them one at a time:

> Severe Problem Number 1: Seeding Initial Wealth

He doesn't really explain why this 'problem' is a bad thing. There is a limited supply, and there needs to be a distribution system. While this one may not be perfect, it doesn't cause any systemic weakness in the system. Early adopters will do well, yes.

But the degree to which people mine or buy bitcoins now is the degree to which they think they'll be valuable in the future - bitcoins already have a price. People who have mined them will sell them to you. Nothing stops you purchasing bitcoins now if you believe the price will go up. This is how markets work. Hearing someone complain about it is like hearing someone complain that people who buy assets cheaply on the stockmarket will make a significant return.

> Severe Problem Number 2: Built in Deflation

A bad thing if you're trying to replace a country's currency system with it, and excellent thing if you see them as a store of value. This is the same as people complaining that any other asset deflates over time.

> Severe Problem Number 3: Lack of Convertibility

This is the closest point he has to being right. Lack of convertibility is a current problem, but economics should solve that. All you need in order to convert your bitcoins is a person who is willing to trade. To have an efficient market, you need an exchange. Here's the crux: bitcoins have a value as long as someone is willing to buy them off you.

His key point here seems to be that 'no one is completely invested in the long-term success of the system' - yes. This is also true of your MSFT shares. No-one is guaranteeing that they will be worth anything in the long run.

> Severe Problem Number 4: When Something Goes Wrong, It Will Die

Perhaps. I went for dinner with some Fx quants last night. One of them brought up this point, and then they all laughed and said "And that's when you invest heavily in it". All assets change their prices based on bad news - most recover again and people who doubled-down at that point make money.

hippich 1 day ago 0 replies      
Not sure how it killed bitcoins, but while most of authors points are valid, conclusion - is not.

People get in to this "Ponzi schema" exactly for reasons pointed by author - deflation currency, unable to track down, limited supply, free/crazy market value. They look for such currency. Why it could not co-exists with rest fiat, gold, game currencies?

If bitcoins will pass test of time, eventually its exchange rate to major currencies will settle down. But for now when accepting payment in bitcoins in my store I immediately convert these to dollars on mtgox.com, since I do not know will tomorrow exchange rate be +50% or -50% =) and I need to pay my supplier in dollars, not bitcoins.

motters 1 day ago 0 replies      
I agree that deflation is probably the biggest criticism of Bitcoin. The finite supply means that there's more of a temptation to sit on your heap of gold rather than to participate in transactions. This could lead to the currency becoming increasingly static and difficult to obtain over time. A better system would be to have a modest but constant rate of inflation, such that there is an incentive not to stash your cash for too long.
pnathan 1 day ago 0 replies      
I think his arguments sum up to, "It's not a government currency", and represent a fear of the unknown.

Further, after I spent some time reading economics books, economics sounds nothing so much like some people with suits on giving opinions, and then having their opinions proved wrong, then claiming that people should still listen to them.

ignifero 1 day ago 1 reply      
Treating bitcoin as a currency is kind of wrong. It's a commodity with limited and predictable supply, like gold for example. You wouldn't call it a pyramid scheme, just like you wouldn't the gold market a pyramid scheme. His points against its use as currency are sound.
seles 1 day ago 0 replies      
Everyone keeps debating whether or not the built in deflation is good or bad. But... how does bitcoin have built in deflation? It only has a decreasing rate of inflation, which can't go negative (since bitcoins aren't destroyed). So assuming the demand for bitcoins becomes constant, their value will be constant.
Dove 1 day ago 0 replies      
That's an excessively linkbaitish title. Please use original titles.
jasonkolb 1 day ago 1 reply      
Some of the primary arguments he uses against Bitcoin are actually its strengths. Fundamentally, Bitcoins will be in higher demand than dollars because their value will continue to rise instead of fall. This means that goods and services as priced in bitcoins will be discounted compared to dollars, driving their adoption. You can see this same dynamic at work in black markets around the world where the USD is the de facto currency vs the highly unstable local currency.

It's all about the "reserve currency", which is the currency that sellers around the world prefer to receive. If bitcoins are more attractive to sellers because of built-in deflation then they'll not only accept them but offer their products at a discount to get them. All of this creates an ecosystem and drives adoption, it does not make the currency a scam as he implies.

pathjumper 1 day ago 0 replies      
None of his arguments are invalid save the "early adopter wins" scenario.

On every other count he either misunderstands current banking systems, bitcoin, presents a pseudo-argument, or some combination of all three. Honestly, except for the early adopter winning part, the whole thing reads like a shill post designed to do nothing but defame bitcoin. He even calls it a scam. By that measure The Fed is the biggest scam that ever invented if you look at who receives the "printed money" first. They are the "early adopters" in his "bitcoin is a scam" analogy. And they are still in power. So does that really make it any worse?

socrates1024 1 day ago 0 replies      
The most important lesson to learn from the Bitcoin experiment is that 'money == speech' on a fundamental, cryptographic level.

Anonymous speech (i.e. using public keys) enables anonymous trades with other people.

Transactions are signed and can't be forged, just like publishing a blog post signed with your public key.

And a distributed network can be strong, just like bittorrent or Tor.

Regardless of whether the current implementation and economy surrounding Bitcoin succeeds (I think it will), this is without a doubt how our future information-enlightened society will develop

Vlasta 1 day ago 0 replies      
Deflation is the big big big problem.

The only bigger problem is that many people in the bitcoin community refuse to consider deflation a problem - some even deny it is happening.

The consequence is that bitcoins are good for investors and speculators, but bad for actual sellers and buyers. People just hoard bitcoins, and sellers are constantly lowering their prices every day while no one buys from them, because who would buy in a deflation?

Bitcoin bubble will burst unless they manage to give incentive to actual sellers and buyers.

I am a fan of digital currency and currently accept bitcoins, but I am considering to stop doing that, unless bitcoin changes from investing commodity to a real currency.

mrerrormessage 1 day ago 0 replies      
As I write this, the value of bitcoin is at an all-time high. I dispute that this article has actually "killed" bitcoins.
tariqk 1 day ago 0 replies      

Jury's still out, but that's okay. All we need to do now is wait.

marckremers 1 day ago 0 replies      
All that article did for me was make me want to become an early adopter and download bitcoin mining apps.
adrianwaj 1 day ago 2 replies      
I found that post painful to read (I disagreed on all headline points,) but I see bitcoin as the start of a new financial reality. I hope bitcoin becomes a new core currency: to me it is like ogg, rather than a bank produced equivalent, which would be like mp3.

Money is fully moving into the digital age, as has media with mp3 and mpeg. It is the medium that will shape its container.

- early adopters can help create or adapt financial institutions

- exchanges are and will go open source

- when something goes wrong the network will adapt. Bitcoin has financially supported developers

- I think all it takes is 1 large company and 1 country to support the existence of bitcoin on a global scale as a test case whilst the rest live in fear

Check out slide 4 of http://www.slideshare.net/15Mb/ages-of-money that is the best way to think about bitcoin in my view: money 5.0.

dynosaur 1 day ago 0 replies      
To be honest, Cohen sounds envious of the early adopters.
weinerk 1 day ago 1 reply      
Since this post - the value of Bitcoin went from aprox $2 to over $10 :-)))))))))


jparicka 1 day ago 0 replies      
Qoura clearly inviding ycombinator....... sick!
pbhjpbhj 1 day ago 1 reply      
I'm wondering if this is an attempt to force down the bitcoin trading price - not necessarily by the author but possibly by the publicists.

Force down price, buy up bitcoins, profit. It's like playing at being a city trading firm.

foxhill 1 day ago 0 replies      
having just read the article, it's apparent that the author has misunderstood, or entirely does not understand the principles behind what's going on.

the number of new bitcoins, for instance, that are allocated for a time period, is fixed, and from what i understand, is allocated to users by the proportion of time spent "mining". ie, 10x more mining on everyone's part does not mean 10x more coins mined.

secondly, the idea that a currency with a finite supply is a bad thing, is just rubbish. gold backed and silver backed currencies have all existed at one time or an other. some would argue that they have more benefits than fiat currency.

his final point is almost entirely non-sensical, and is related to the part before. during the bank bailouts in the financial crisis (which again, some would argue we are still in the midsts of), the bailouts created money out of air - devaluing the currency for everyone, increasing inflation, and THEN passed the debt to tax payers. how this is construed as a benefit for citizens of respective governments is beyond me. during recessions, the poor get poorer and the richer get richer. money doesn't disappear, it gravitates to those that already have it.

his 3rd point is the only one that comes close to making sense, alas, it's not really an issue. people are their own converters. how much someone will pay for something is entirely up to them, and whilst i don't see there ever being an extra column in XE.com, that's really neither here nor there.

doctororange 1 day ago 0 replies      
Wow. Have to disagree with every one of his points. Total link bait here - I really doubt the success or failure of bitcoin will depend much on what one guy on the internet reckons.

I love that his final argument is just "Something will go wrong - I dunno what, but something will." Mm. Top notch.

spot 1 day ago 3 replies      
When he showed the graph of the number of tweets vs the number of bitcoins he lost me. This is an apples and oranges comparison. http://bitcoinreport.blogspot.com/2011/01/bitcoin-transactio...

It's just FUD.

Groupon files for IPO wsj.com
225 points by 44Aman  1 day ago   153 comments top 28
aresant 1 day ago 7 replies      
Andrew Mason's desire is for Groupon to be where you go when you think "I'm hungry or I'm bored"

If you're calling bubble and confused about how they're running at a loss don't miss this line:

"Participating Merchants: 56,781 in the first quarter of 2011, up from 212 in the second quarter of 2009"

That database of 56,781 merchants is GOLD.

The way their sale staff works is to create direct relationships, phone contact etc - that is not a cheap proposition.

In terms of growth potential there are 10x as many restaurants in the USA as Groupon's entire universe of merchants today.

If they continue to capture the consumer mind that they're the best in the world to answer those two questions, their valuation and growth potential is insane.




hncommenter13 1 day ago 1 reply      
One important item that hasn't received a lot of attention is Groupon's "Accrued Merchant Payable." If you'll indulge me, a longish thought experiment (yes, it relates to Groupon).

Imagine a sandwich shop that allowed customers to purchase future sandwiches--buy one today at a 50% discount, eat it sometime in the future. The sandwich shop would receive $3 for a sandwich for which it normally charges $6, and it would owe me a sandwich at a future date. Also assume the sandwich costs the shop $1.50 in direct costs (50% margins at a $3 price).

This proves to be a popular promotion with the shop's customers. The shop sells lots of $3 "sandwich rights," bringing in $3 in cash up front. It spends a good deal of that $3 in cash to pay ongoing expenses and to get the word out about its 50% off sandwich deal.

But then the growth of its "sandwich rights" business slows. Other sandwich shops offer a better deal--$2 for a $6 sandwich--and it begins to saturate the market of local lunch eaters, causing a slowdown in the sales of sandwich rights and the cash they've been paying the shop in advance.

Now the sandwich shop owes sandwiches to all of its rights holders, each of which costs $1.50 in cash expenses (to pay suppliers, employees, etc). However, instead of holding the cash it previously received for the sandwich futures, the shop has already spent it on marketing to other potential purchasers of sandwich futures. Clearly, if the shop doesn't have the money to pay $1.50 x # outstanding rights or can't get financing, it will go out of business. Because the shop was dependent on sales of sandwich rights to finance its growth, when the growth rate slowed, the money dried up. In essence, the shop borrowed from the future by sucking in cash today for discounts on tomorrow's sandwiches.

This is exactly what Groupon has done. Its operating cash flow includes "Accrued Merchant Payable" of nearly $291M (3/31/11). But its cash balance is about $208M (3/31/11). Because it collects cash up front from individuals and pays merchants over time (or, in its non-US operations, only when coupons are redeemed), Groupon is showered with customer cash before it must pay merchants. Roughly half of this cash eventually belongs to Groupon, while the other half is eventually owed to merchants (true, there is breakage, but if nobody redeems the coupon, that adds little value for the merchant, so significant breakage/non-redemption isn't necessarily in Groupon's long term interest).

In other words--and Groupon spells this out--if the growth rate in coupons sold to customers dives, Groupon could face a cash flow problem. It's not a ponzi/pyramid scheme exactly, but it is a highly risky financial practice to spend cash you will owe tomorrow on expenses you incur today. As long as the company grows and/or can sell shares to the public and increasing prices, it will do fine. Once the growth slows or access to capital dries up, it's vulnerable. Groupon may well outrun the cash demands it has piled up by going public. But it can't maintain these growth rates forever--remember those other sandwich shops selling similar products?--and will ultimately face the music.

Don't believe me? Here's a quote from their S1:
"Our accrued merchant payable, which primarily consists of payment obligations to our merchants, has grown, both nominally and as a percentage of revenue, as our revenue has increased, particularly the revenue from our international segment....We use the operating cash flow provided by our merchant payment terms and revenue growth to fund our working capital needs. If we offer our merchants more favorable or accelerated payment terms or our revenue does not continue to grow in the future, our operating cash flow and results of operations could be adversely impacted and we may have to seek alternative financing to fund our working capital needs."

nicpottier 1 day ago  replies      
Can we call it a bubble yet?

Those of us who were around in the early 2000s recognize this game. Brand new ventures filing for IPOs based on 'amazing potential' and even more amazing valuations.

This is all going to come crashing down soon. The question is whether it happens before or after Bitcoins. :)

bryanh 1 day ago 3 replies      
Absolutely wild. Revenue was almost as much in Q1 2011 as it was in all year in 2010 ($645m vs. $713m) with a 20,000% revenue growth since June 200.

They are hoping to raise "close to $1 billion at a valuation of about $20 billion." [1]

And they still aren't turning a profit. (~15% loss Q1 2011 and ~54% loss in 2010) [2]

[1] http://online.wsj.com/article/SB1000142405270230374530457636...
[2] http://www.sec.gov/Archives/edgar/data/1490281/0001047469110...

zmmmmm 1 day ago 1 reply      
If nothing else the timing of this seems suspect to me. The Linked In IPO has investors who missed out champing at the bit for something else. I think Groupon knows their current business model is unsustainable and have picked this moment precisely because the market is particularly irrational right now. (Note: I'm not saying there is a bubble in general, just that the Linked In IPO has created a unique opportunity for them to go public with less scrutiny than would otherwise be the case.)
alain94040 1 day ago 1 reply      
Am I wrong, or any idiot can spend 1 billion dollars to make 700M?

It's very scary to me that they scaled and failed to prove they can profit, before going IPO. I have no issue with that tactic as a private company, but you shouldn't go public until you can prove profitability. Otherwise, the bubble word is truly deserved.

vessenes 1 day ago 2 replies      
My read on this, especially given the extreme flexibility series G investors showed in cashing out founders, is that when Groupon is up and running in a locale, it makes a whole shitload of money.

I also would anticipate from reading the expansion numbers and having a little bit of business experience myself that Groupon grew literally as fast as it possibly could in the last few years; there was no way for them to successfully move any faster, no matter how much cash they were given.

I'm guessing you'll see some gyrations as they continue to try and solidify their global lead, then slow move to profitability, then one day, (if margins hold up) BAM. Major Net Income.

Right now the market clearly is going to reward a company who can get this done successfully in as broad a portion of the world as possible; if they can demonstrate that existing locales are profitable after a certain period of time, they will have happy shareholders as well. It's a landgrab, and Amazon is a good comparison.

jdp23 1 day ago 1 reply      
From the article:

'Don't expect profits anytime soon: Groupon hasn't turned a net profit in any of its first three years of operations, including a net loss of $389.6 million in 2010. The company said it expects its “operating expenses will increase substantially in the foreseeable future ...“'

Sounds very 1990s dot-com bubble to me ...

jcampbell1 1 day ago 0 replies      
I know there are people scraping Groupon deals and sales figures and estimating revenues. Does anyone have the name of these sites? It seems like a hell of a opportunity to sell $10k research reports based on public information to wall street traders.
dsplittgerber 1 day ago 1 reply      
Cue for dozens of Fortune, Newsweek et al articles proclaiming the new bubble. If everyone's so confident in their assertion, go short that stock.

Groupon will grow like hell until there are finally no new deals to lure with left and the 'extreme couponing' lifestyle has been grinded to death (regular couponing will have a place like it always had).

The problem with extreme valuations/bubbles is not so much to recognize them, it's to pinpoint when they will burst.

paraschopra 1 day ago 1 reply      
When and how can they eventually make some profit? If they aren't making a profit at expected ~2.5 B revenue, how would an additional infusion of 1 B make them profitable or let them grow more in the long run? Isn't that what companies get listed for?
megamark16 1 day ago 5 replies      
I just don't understand why anyone would invest in a hot IPO for a company that hasn't even turned a profit yet and had a $389.6 million net loss in 2010! Can someone explain to me the allure of an investment like this?
dr_ 1 day ago 0 replies      
"Don't expect profits anytime soon:"

Based on their first quarter 2011 results, they are on a revenue run rate of 2.4 billion a year.
It's a little surprising that they are not going to generate profits on 2.4 billion a year in revenue, despite the fact they employ around 7000 people and have other operational expenses.

thomasgerbe 1 day ago 3 replies      
Does the fact that they aren't turning a profit concern people that much? Didn't the Xbox division take years before profitting? I'm not a business guy so I'm genuinely asking this out of curiosity.
danielharan 1 day ago 0 replies      
No net profit, but: "gross profit (revenue minus expenses, which was $280 million in 2010)"

So it sounds like they're just spending madly on acquisition and growth. Or am I missing the gross/net distinction here?

iqster 1 day ago 1 reply      
I'm surprised by the loss they are running. Think how big their cut is. My understanding is that on a 50% off deal, they get half of the merchant's take! Granted their effective cut is lower since they provide some costly/valuable services .. e.g. they pay you right away while redemptions might happen over months. But still ...

I wonder if merchants aren't just curious about the groupon model as opposed to groupon having "cracked the local nut". I guess we'll know in 2-3 years. If their revenue flattens or goes down vs if it keeps going up up and up.

On a side note, I hope this makes Facebook file for their IPO already. For some reason, I feel the Facebook IPO will be a turning point of sorts in the current tech boom. Not sure why though.

bhangi 1 day ago 1 reply      
From what I could make out of the prospectus, the revenue number is the sum of the face value of all coupons sold. Since Groupon has to pay the merchants a predetermined percentage of the face value, I'm having a tough time understanding why the entire face value should be considered as revenue. To take an analogy, this would be like Visa claiming the total value of transactions as revenue instead of the fees it charges the merchants for said transactions. Am I missing something?
kinofcain 1 day ago 0 replies      
$200million in ad spend + dozens of competitors spending just as much or more on ads = buy GOOG.
Apocryphon 1 day ago 1 reply      
Looks like they beat Zynga to it, there's only a handful to high-profile tech companies who are likely to go IPO soon, perhaps we could bet on who goes next + what their valuation will be at.

Maybe the next one isn't one of the big ones (Facebook, Zynga) but is instead something like Yelp or Pandora.

Edit: Didn't realize that Pandora just filed. Overshadowed, indeed.

naeem 22 hours ago 0 replies      
I worked as a sales manager at a group buying site which is a competitor of Groupon, and it amazed me how well off they were despite playing a domain you would expect to be monopolized by a goliath like Groupon. Just goes to show how much room for expansion there is in the niche.
suking 1 day ago 2 replies      
Wow, Eric Lefkofsky owns a shit ton of that company, and already cashed out a decent chunk of change personally and through his various LLCs just recently. Never even heard of him before.
tocomment 14 hours ago 0 replies      
When is the actual ipo though? These articles never seem to mention that.
hxf148 1 day ago 0 replies      
It does feel like 1998-2001 but it also feel more solid this time. Business with actual revenues rather than the hypothetical revenues of so many early dot com's.

I kind of hope we aren't' in a bubble but a rise in the economy. Either way head down and back to work. I missed the first bubble and related opportunities being distracted by school and the fun of school.

Not this time. I doubt that http://infostripe.com will IPO anytime ever but if there is enthusiasm and growth in the industry then I want to be in there somewhere in the wings fighting over the scraps.

matt_s 1 day ago 0 replies      
Wouldn't it be typical of a growing company like Groupon to run negative because they keep investing all the profit into people and technology? Paying over 7000 people and their benefits probably takes a majority (e.g. 51% or more) of the revenue.
MatthewPhillips 1 day ago 0 replies      
Coincidental timing that they happens right after Google announces it's baking its clone into all future Android phones?
oceanician 13 hours ago 0 replies      
It's not a boom. It's not a boom. It's not a boom
veyron 1 day ago 0 replies      
ticker symbol grpn
Google open sourcing voice and video engine for the web google.com
221 points by gaika  3 days ago   43 comments top 12
zmmmmm 3 days ago 2 replies      
Is this Google's response to MS purchasing skype? Seems like the classic strategy of commoditizing your competitor's business (and they get a 2-for-1 deal since Apple is also wielding FaceTime).
r00fus 3 days ago 4 replies      
Apple, where is the Facetime open spec that you promised?
Google may just beat them to the punch.
angusgr 2 days ago 1 reply      
Does anyone have any ideas what the timeline for browser support is likely to be?

The website currently says they've been working closely with Mozilla and we expect to see WebRTC support in Firefox and Chrome soon!.

jbk 2 days ago 0 replies      
How do iLBC and iSAC compare to the (also open source) Celt, Silk and Opus (Silk+Celt) codecs?
car 2 days ago 1 reply      
Very nice, thank you Google.

Glancing over the API docs, I'm not clear what will be used for signaling. It appears to be based on XMPP/Jingle, am I right?

What about SIP?

zokier 1 day ago 0 replies      
What does this mean to existing SIP/Jingle based systems? Are they now deprecated, or is there some kind of interoperability planned?
yblokhin 3 days ago 1 reply      
Finally we know what's going with iLBC codec. For more than a year it's been in a limbo.
dblock 2 days ago 1 reply      
One way to decimate adversaries is to open-source your entire competitive advantage. Unfortunately that only works when you're the market leader.

So nobody cares about open-sourcing WebRTC. Something that would be actually noticeable in this field right now would be Microsoft that open-sources Skype and gives everything away under the BSD license.

tobylane 2 days ago 0 replies      
If Pidgin (the cross platform IM framework that Adium/others use) doesn't get this, I'll be sad/grumpy.
reustle 3 days ago 2 replies      
Not long until I'll be able to video chat on my Android via Google Talk, woo!
odiroot 2 days ago 0 replies      
So is this another segment of HTML5 block or a completely novel idea?

It looks like a <device> element with streaming but they don't mention it at all.

chopsueyar 2 days ago 0 replies      
Good work, Microsoft!

Embrace and extend.

The Architecture of Open Source Applications: LLVM aosabook.org
220 points by ryannielsen  4 days ago   29 comments top 7
yuvadam 4 days ago 0 replies      
I've been hearing about LLVM for quite some time, but never fully grasped what it is. This article tackles everything you should know about LLVM. I'm finally starting to get it.


forgottenpaswrd 4 days ago 0 replies      

Very good article, I started liking LLVM some time ago. Now I will use it on my programs, it seems really easy to do, I agree so much with the gcc design flaws(that forced me to create my own parsers and code generators).

obiterdictum 4 days ago 4 replies      
I've never built a compiler, but how well does having common optimizer for all platforms work in practice? (11.5.1) Does it cause a "lowest common denominator" effect thus making the code take less advantage of the target architecture than a architecture/vendor-specific compiler would?

I'd love someone with the experience of building compilers to chime in.

henning 4 days ago 1 reply      
As this essay discusses, GCC has serious technical flaws. LLVM is one of the best examples I've ever seen of "the best way to complain is to make things." That and the iPhone.
nhaehnle 4 days ago 0 replies      
To anybody curious about how LLVM looks on the inside, I recommend reading their excellent tutorial on building a front-end, which you can find here: http://llvm.org/docs/tutorial/
akkartik 4 days ago 1 reply      
Looking at the big picture raised one question for me: how does LLVM deal with dependencies between optimization passes?

It is well-known, for example, that the whether you do register allocation before or after instruction scheduling can have huge implications for performance. Some programs prefer scheduling done earlier, and some prefer it later. gcc's optimization flags switch phase orderings among other things.

Does LLVM have any abstractions for managing phase orderings and intelligently picking between them? Or must the optimization classes be manually instantiated in the right order?

kenjackson 4 days ago 1 reply      
Does anyone understand their purchasing table? It seems like some money goes unaccounted for: http://www.aosabook.org/en/index.html#purchase
Introducing schema.org: Search engines come together for a richer web googleblog.blogspot.com
213 points by Uncle_Sam  1 day ago   73 comments top 20
rauljara 1 day ago 8 replies      
All I could think while reading through the getting started was: that is an awful lot of added text. After a little more thought: that is an awful lot of added work. And while it won't be hard to have tools that make the process easier, the sort of work that goes into adding that data can never be completely automated (otherwise, we would have no need for it). Given that all the search engines will be using it, all major sites basically have to implement this or they risk falling in their rankings.

So, at the end of the day, Google, Microsoft, and Yahoo just made web development more expensive. They probably also just made the web a better place, too.

Loic 1 day ago 1 reply      
For most of my "quick searches", I already have the answer from within the search results listing. It looks like it will go one step further, we are not going to have to leave the results page to have complex answers.

I am not sure if I want to provide all my hard work in a format which will maybe help the search engines a bit, but mainly the spammers a lot as they will be able to automate the creation of content farms even more.

Mixed feelings... all the world data in a well structured format is a wonder but at the same time, what will be the incentives to create such an easy to digest content if the world at the end do not even know you are the one how produced it?

Kind of the old media against new media dilemma but applied to the new media. Interesting.

ryanisinallofus 1 day ago 3 replies      
This is going to sound curmudgeonry but it is seems like one more way search engines want to use your data without giving you the page view. It makes allot of technical sense and I can imagine some really great ways to use this data but in the end I guess I would just need to ask "what's in it for us- the content providers?"
PaulHoule 1 day ago 0 replies      
if they actually wanted people to use this they'd write better documentation.

if I picked 10 web devs off sitepoint and instructed them to add 10 assertions to an HTML page and didn't give them a validator, I'd be amazed if more than 3 got 80% of them right.

i like the taxonomy though, but honestly i think instances are much more interesting than types... rather than saying "George Washington" a :US_President, can't we say "George Washington" is :George_Washington where :George_Washington is his identifier in Freebase?

izendejas 1 day ago 0 replies      
Assuming that authoritative sites, at least, don't abuse these schemas, this will help all search engines and data mining/nlp researchers build better models. The biggest gain isn't quick view of search results, it's that now search results will be better in general because Google et al will now understand if a page is really about a person and in many cases, who specifically it is about to point out a specific example.

Information extraction, just got that much easier. Hello, baby semantic web.

ivan_ah 1 day ago 0 replies      
Very cool, but why didn't they use the existing RDFa format/keywords but invent their own?
[ see http://en.wikipedia.org/wiki/RDFa ]
[ itemprop vs property ??? ]

I guess when you are big G, you can do anything you want.

ecaron 1 day ago 0 replies      
They neglected a format for job listings, which is unfortunate. We (LinkUp) put one together at http://wp.me/pJYG0-1H, it will be interesting to see if it gets any attraction and if they're actually seeking external input.
callmeed 19 hours ago 0 replies      
What's going to stop people from gaming this by doing things like adding fake 5-star reviews to their website? (especially brick and mortar stores that show up in google maps/places)
andresmh 1 day ago 4 replies      
I am wondering why they didn't go for something that allows for richer semantics like XHTML+RDFa http://www.w3.org/TR/rdfa-syntax/
fbnt 15 hours ago 0 replies      
I thought that snatching the type of content of my pages should be their job. I'm so naive.

No problem, I'll add a few kbytes to every single page of my sites, so I can replicate the information I've already stated in a number of sitemaps, video sitemaps, headers and XML files.

P.S: I'm not agaisnt standardization at all, I'm just saying, this comes a bit late.

Kilimanjaro 1 day ago 1 reply      
Nothing an xml data island could not have solved. Even an external xml data island with internal references for better performance and less clutter. Even an external JSON data island would have been better for web consumption.

Microformats? I'll pass.

tantalor 1 day ago 2 replies      
Did Google just deprecate microformat?
takinola 1 day ago 0 replies      
This sounds rather similar to the Facebooks OpenGraph protocol (http://developers.facebook.com/docs/opengraph). I wonder if this is related to Google's planned entry into social. It would help if they knew the context of search terms so they could match it up to ads in Gmail for example. or use your gmail conversations to help reorder search results...
ericmoritz 1 day ago 0 replies      
Whatever happen to HTML5's data-* attributes?
stbullard 1 day ago 0 replies      
Am I alone in thinking schema.org is a direct answer to Facebook's Open Graph Protocol?
ciex 1 day ago 2 replies      
The question remains: Who needs this?

I am still not really convinced that it is possible to integrate handcoded schemas for a wide range of use cases into search results in a meaningful way.

The solution Google proposes here will also restrain the content of websites in a lot of ways if it becomes widely adopted. Look at the recipes-example, it defines markup for including nutrition information for recipes:

"Can contain the following child elements: servingSize, calories, fat, saturatedFat, unsaturatedFat, carbohydrates, sugar, fiber, protein, cholesterol"

Every company that serves recipes on the web and decides not to offer this information because it deems other properties of recipes more important is now at a disadvantage. Google will show more information about the recipes of their competitors and presumably also rank them higher because they have included 'valuable' markup information in their recipe.

This approach favors shallow information ressources over complex ones as the former can be more easily parsed by metadata-crawlers.

netgineer 1 day ago 2 replies      
For a site put together by search engines, the URL structure for the site search is atrocious. "#q=Product" and not "?q=Product"? Who thought that was a good idea?

Site also looks a bit like spam. Needs more Firefox-esk awesome graphics, imo.

MatthewPhillips 1 day ago 2 replies      
So Google both ranks on page speed and encourages you to double your bits by adding a lot of cruft to your html. Wonderful.
baconner 1 day ago 1 reply      
Putting aside the technology and schema decisions they've made IMO its great to see these three throwing their weight behind some common metadata even if it does step on some toes.

Now if they'd only add some schema targeted towards downloadable public data sets. I'm dying for a good global public dataset search beyond competing data markets and data.gov.* sites.

mgkimsal 1 day ago 3 replies      
Seems that if you use this, your documents won't be able to be considered 'valid' by validators (tested on the w3 validator). Unless, perhaps, you just mark your doctype as html and be done with it?
How Facebook pushes updates to the site facebook.com
214 points by creativityhurts  6 days ago   40 comments top 13
akent 6 days ago 0 replies      
I'm pretty impressed by the "push karma" system for gauging how risky individual engineers' commits are on average.

Here's another (much shorter!) video where Chuck Rossi talks about push karma very briefly: https://www.facebook.com/video/video.php?v=778890205865&...

abi 5 days ago 0 replies      
Anyone know of a more automated system to handle forward/backward compatibility? Obviously, there's a lot of manual coding work that has to be done but is there a system that categorizes these various changes (schema or new URL for a page or change in backend service interface), automatically tracks and gets rid of these dependencies after a certain period of time? To give a concrete example, let's say I switched the Facebook messages URL to "/mail" from "/messages". I would mark the old handler as deprecated and eventually, after the new changes have been pushed to everyone, the system will prompt the developer to get rid of essentially the dead code. This is a very simplified example but I believe such deprecation tracking would be useful for more complex changes too.
shin_lao 6 days ago 0 replies      
What he says at the beginning is to me the most important. Having great tool is great (!), but the most important is to have the right culture about QA and releases.
rasmus4200 6 days ago 1 reply      
I loved this video.
Gatekeeper blew me away.

Summarized some of the highlights here if you don't have time to watch:


avar 6 days ago 4 replies      
It's interesting that their entire release architecture seems to be focused on never pushing bad things out to production, whereas given their traffic they could probably push things out much sooner (minutes after they're committed) to small parts of their overall traffic, and slowly increase the traffic on those pieces of code as they prove themselves to be stable, or quickly revert them if they're not.

That would mean having a lot of versions of Facebook live at any one point, but as those parts prove themselves stable they'd gradually be rolled out to all of their traffic.

One point that also wasn't covered is that as they're pushing things out they'll only cherry-pick parts of their codebase depending on which engineers they have around. I wonder if they have a lot of hairy merge conflicts around release time due to that, and bugs in production resulting purely from those merge conflicts. Or worse, subtle bugs resulting in change A going out, but being programmed against a function that was changed in change B, which is not going out because the author of change B isn't in today.

baby 6 days ago 3 replies      
So facebook is programed in PHP but everything on the server is in C++ thanks to "hiphop"? mind=blown
madamepsychosis 6 days ago 1 reply      
I like how their entire development cycle revolves around people getting drunk on weekends.
chuckr 5 days ago 0 replies      
I just wanted to mention one thing about this video. It's missing the first 3 minutes or so where I introduce myself and my team. It's also missing the part where I gave credit for some of these slides to John Allspaw and Paul Hammond from flickr who gave an awesome talk at the 2009 O'Reilly Velocity Conference. Their talk inspired me to put together this presentation.
brown9-2 6 days ago 2 replies      
Does anyone know if a non-video summary or set of slides for this video exists?
swah 6 days ago 1 reply      
This is how I push updates for now:

    git pull
lein uberjar
sudo restart myprj

cpg 5 days ago 1 reply      
Anyone know the URL for the code review tool they said they use and open sourced? He said "fabrication", however I cannot seem to find it. It's also not listed in http://developers.facebook.com/opensource/
maeon3 6 days ago 1 reply      
Was able to download this video with savevideo.me
ignifero 6 days ago 2 replies      
So they don't really test the code, they just push it out, and fix the bugs on Thursdays. Genius strategy. No wonder their platform breaks so often.
       cached 4 June 2011 02:11:01 GMT