hacker news with inline top comments    .. more ..    3 Sep 2012 News
home   ask   best   7 years ago   
Left Alone by Its Owner, Reddit Soars nytimes.com
91 points by donohoe  4 hours ago   12 comments top 6
thesethings 2 hours ago 3 replies      
"Steve Newhouse, the chairman of Advance.net, decided very early on that his company would not be the blob that ate Reddit, and for the most part, left well enough alone. “We had some ideas about what would be good, but it might not have worked,” Mr. Newhouse said. “We paid attention to the community instead.”"

I may be remembering stuff wrong, but wasn't there more tension + conflict during the period right after the acquisition? Didn't Reddit struggle to pay for basic infrastructure stuff due to their new parent company's skepticism about its viability and perception of its value?


And it was a long while until it got spun out. (http://venturebeat.com/2011/09/06/reddit-break-conde-nast/)

Not that there's anything crazy in any of this. Just that the NYT story paints it like right away they 1) knew Reddit's worth and 2) knew to have a hands-off approach.

I'm far from a Reddit power user, happy to hear other perspectives.

tokenadult 1 hour ago 0 replies      
"Strategically, what they have done should be a model of how to create and support a virtual start-up within a larger corporation,' said Anil Dash, a writer and entrepreneur in digital realms."

Well, strategically, they might eventually want to think about making some return on investment for the acquiring investors with actual cash profit. Otherwise, they will eventually be shed at a loss (is anyone else here old enough to remember AOL and Time Warner?).

Strategically, they may also be serving as a cautionary example of why start-ups without a secure revenue model shouldn't be acquired at all, unless there is financial upside for the investors. After all, the article further reports, "Reddit is not an exception to every rule in the digital world. Like many digital media companies, it has a big audience and minuscule revenue."

brackin 2 hours ago 2 replies      
I think it's a shame that Reddit sold so early but at the same time they can think of the users (Until Conde Nast gets involved). There's no question about their valuation and how they'll monetize. They as a company can work on that and build a great product but there aren't daily blogposts on TechCrunch about it.
carleverett 2 hours ago 0 replies      
I'm sorry but the first sentence made me laugh:

"There are many ways to measure the traction of a social media platform: [3 ways, to be exact]"

noirman 49 minutes ago 0 replies      
Keep in mind that when Reddit got acquired, Reddit gold doesn't exist yet. So they were losing (more) money every month.
taybenlor 57 minutes ago 0 replies      
"If the leader of the free world stops by" - sorry, he's what?
Reddit's database has only two tables inburke.com
139 points by kevinburke  6 hours ago   40 comments top 13
tzs 5 hours ago 5 replies      
> Adding a column to 10 million rows takes locks and doesn't work.

It does not take locks, other than for very briefly.

1. Make a new empty table that has the same structure as the table you wish to add a column to. Add your new column to the empty table.

2. Put triggers on the old table that, whenever a row is added or updated, makes a copy of the row in the new table or updates the copy already there.

3. Run a background process that goes through the old table doing dummy updates:

    UPDATE table SET some_col = some_col WHERE ...

where the WHERE clause picks a small number of rows (e.g., just go through the primary key sequentially). Since you aren't actually modifying the table, all this does is trigger the trigger on the specified rows.

4. When you've hit everything with a dummy update, rename the current table to a temp name, and rename the new table to the current table. This is the only step that needs a lock.

There are tools for MySQL to automate much of this. There was a post either here or on Reddit a while back about this which linked to them. I'm sorry but I didn't save a link to it so you'll have to search if you want it.

prodigal_erik 5 hours ago 1 reply      
This is called a http://en.wikipedia.org/wiki/Entity-attribute-value_model or http://en.wikipedia.org/wiki/Triplestore. I think the author is understating the price, though. There's a lot of existing software you could reuse if your data was stored in more conventional relations, and "manually enforce consistency" is a pipe dream. Your code has expectations about your data, so in the abstract you do still have a schema, and not writing it down merely prevents any tools from helping you keep your data sane over time. I've seen Notes databases decay to the point that not even the dev team could explain how a document got into its contradictory state nor how the apps should (much less currently would) handle it. The few people diligent enough to do completely correct work without a checkable schema, aren't the people who would be tempted to try.
zzzeek 4 hours ago 6 replies      
Pretty sure Reddit has thousands of tables - last time I looked, it was really hard to see this but this is how it seemed like it was working. It has "thing"/"data" tables for every subreddit - created on the fly (a crime for which any DBA would have you put to death, normally). While I'm honored they use my library (SQLAlchemy) for relational database access, their actual usage of the relational DB couldn't be more...well... let's just say please don't imitate their style. If you want to build a Reddit, use Mongo/Cassandra or something like that. They'd very likely have done so themselves if NoSQL products were mature when they first developed their platform (I am vaguely recalling/guessing here on that one).

Edit: if any reddit devs want to correct me here, feel free, as I found the reddit source extremely difficult to follow back when I looked.

kjhughes 5 hours ago 0 replies      
This is a thin wrapper around a good, but two year old, High Scalability post:


which was discussed on HN recently:


and on HN long ago:


EDIT: Fixed "long ago" vs "recently". Thanks, sync.

ocharles 4 hours ago 0 replies      
That quote is just painful to read, littered with FUD and not a single bit of evidence to back it up.

You should worry about the database because it's probably your canonical storage of data, which for most of us is the most important part of our product/service/whatever. A good schema enforces consist data, invariants, and all sorts of other stuff that you don't want to be dealing with a manual (and buggy) basis.

Schema updates do not need to be slow. They might not always be as elegant as you hope but the big databases are improving on that front, and as tzs mentions - there are tricks that can be employed. With the latest and greatest PG, I believe we're even starting to get event triggers, so it may well be possible to do schema updates with replication. I also have a feeling the binary replication in PG 9 and up can even do it out of the box, with hot standby to still allow responses. I'm not entirely convinced replication is a backup solution, so maybe that was an operations antipattern. That's some baseless assertion from me though :)

If deployments are a pain, work to alleviate pain. They are pretty mechanical, even if involved, which lead very nicely to being automated.

Seriously, we're smart people, let's not throw at least 30 years of research out the window in favour of glorified entity-attribute-value schemas.

ionforce 1 hour ago 0 replies      
My fear with headlines like this is that people with no business working at the scale that Reddit does will suddenly eschew years of best practice SQL and by like "my dad's pizza shop CRM only needs two tables, JUST LIKE REDDIT".

Everyone please, use your brain before repeating such specific configurations. Reddit is quite exceptional. Your burgeoning to-do list app is not.

rfurmani 5 hours ago 0 replies      
Umm, assuming the github repo is what they actually use (i assume so given how often it is committed to) there are two tables per object. Reddit_thing_link, reddit_data_link, reddit_rel_thing_savehide, etc etc
cageface 1 hour ago 0 replies      
I used to work at a company with a huge LDAP database that had been growing in uncontrolled and organic ways for years. If you've never worked with LDAP before, it's basically a big key-value store, with some optional schema enforcement. We spent an inordinate amount of time manually cleaning it up and trying to enforce some kind of data integrity and we were always bemoaning the lack of a real structural schema.

I think the kind of ad-hoc generic data typing described in this article is sometimes the right solution but it comes at a cost.

lazyjones 5 hours ago 0 replies      
Seriously? Sounds like a half-baked reimplementation of a https://en.wikipedia.org/wiki/Triplestore to me ...

The post he refers to is more than 2 years old, things may have changed.

gioele 2 hours ago 0 replies      
Basically they rediscovered RDF.

They should switch to an RDF triplestore, at least they would be able to exploit some of the RDF-only optimizations and their validation tools.

SoftwareMaven 3 hours ago 0 replies      
This is roughly the path I'm taking on my current project. There are some things that have a consistent, "traditional" schema. However, many of the feature-important things are being stored as JSON blobs. This means that I don't have to worry about schema migrations when I add a new feature; instead, I just define its JSON schema. This works particularly well because I don't need to join on these items. The downside is that I do have to do occasional manual processing on them (sum up all values of type FOO) that would be trivial if everything were perfectly normalized.
sjs 1 hour ago 0 replies      
No, they have <thing> and <thing data> tables for each entity. This is a grave misunderstanding of what was actually said.
pud 4 hours ago 1 reply      
I wish I read this article a year ago before I started developing my latest app. I have that exact problem, where adding a column to a 10M+ row table takes over an hour. So instead I end up with "things" tables all over the place.


SlabText.js frequency-decoder.com
56 points by electic  4 hours ago   7 comments top 4
wamatt 2 hours ago 1 reply      
Nice :)

Just some feedback, there's about a 1 second delay before the final text is rendered on initial load. This creates a slight jarring experience due to resizing, kinda analogous to the infamous @font-face FOUC/FOUT issue

Tested on latest versions of Chrome, Firefox and Safari.

vhf 4 hours ago 0 replies      
Great work, the effect is stunning !

Really nice that it's automatically disabled when displayed on a small screen than viewport defined.

pud 3 hours ago 0 replies      
Beautiful. Does it work in IE? (i'm too lazy to fire it up)
trimetric 3 hours ago 1 reply      
This is a really impressive effect, but I'd hesitate to use it as part of a generic cms backed site since the auto-sizing could inadvertently editorialize the meaning of a headline at different page widths.

"Roddick SURVIVES, Staving Off Retirement" vs "Roddick Survives, Staving Off RETIREMENT"

Apple Never Invented Anything mondaynote.com
154 points by chmars  7 hours ago   118 comments top 19
cromwellian 5 hours ago  replies      
One interesting point is he brought up Einstein's invention of relativity. Notice that in the scientific community, humanity has been building upon previous work for centuries with no copyright or patent protection, and nothing more than honor, citation, and shame against plagiarists.

The fundamental defense made by many is that without patent protection for software, people would not put much effort into innovating.

1) If you look at science, open source, fashion, food, and other areas where humans continually build on culture, you see plenty of continued innovation without insane legal protection.

2) The amount of effort pales in comparison to the monopoly granted. You could make the argument for say, pharmaceuticals, that if it takes 10 years from lab through human trials and hundreds of millions of dollars, that a 2-decade long protection period might be needed. But there is no FDA for software, and Apple actually spends far less on R&D than other companies, and with $100 billion in the bank, you can't claim that haven't gotten an incredibly good return in their investment.

Therefore, it is insane to grant 20 year protection to Apple for stuff like pinch gestures. 2 years maybe. But 20? It's absurd.

drats 6 hours ago  replies      
I made this case last week.

"Apple products are like a good classy restaurant or hotel chain. They take ingredients everyone has and put a lot of work into fit and finish, they make the customer feel special for a slightly higher price."[1]

And I stand by it. I'll go even further actually, I think Apple is one of the least innovative big companies. Look at all the big research labs at Microsoft, Yahoo, IBM or Google. Anyone who seriously follows this stuff knows a) Apple doesn't have a profile in the academic world and b) knows enough computer history to know Apple is claiming things invented decades ago.


VengefulCynic 6 hours ago 3 replies      
Whenever I'm using "enterprise" software or some equally-awful tool that's sold using a feature checklist, I can't help but think about Steve Jobs's line about how innovation is saying no to 1000 things.

edit: Well, a feature checklist and $50,000 worth of sales dinners, games of golf and "gifts that do not violate the professional ethics rules of the company buying said tool."

AnthonyMouse 6 hours ago 2 replies      
I think the article is right in what it says but is wrong because of what it doesn't say. Apple succeeded where others had failed, and that is certainly commendable, but now we have a problem: Apple doesn't have a patent on "an iPad" meaning a device with all the individual characteristics that make an iPad an iPad and make it successful, instead they have individual patents on all the individual features.

But the individual features are the things that Apple didn't do. Yet that's what they sue over because that's how patent law is set up.

So now we see Samsung lose big in court and popular reaction is split, and here's why: People looking at the actual facts of the case are outraged that Apple could win that way because the actual grounds of the win had nothing to do with copying or Samsung's actions and everything to do with the fact that anyone with a million over-broad patents on obvious "inventions" and laws of nature and mathematics can win in court against anyone who produces a computing device, arguing that any actual copying on the part of Samsung is irrelevant. On the other hand, we have the people who look at the result and the fact that Samsung's devices do actually look entirely too much like Apple's and think Samsung got what was coming to them, ignoring that in order to do it Apple had to adopt a long list of bully tactics that they've now demonstrated that they or anyone else with a sufficient patent arsenal can successfully use against their competitors (including those whose devices aren't intentionally copied, because there are too many patents to possibly even attempt to avoid them all).

Nobody seems willing to say that Apple should potentially have some remedy against Samsung for actual copying but that what they got is the wrong remedy in the wrong way, not least which because the same tactics can be used against anyone whether they've done anything wrong or not.

drblast 3 hours ago 1 reply      
This article is unintentionally an excellent argument against patent protection for Apple's products.

If Apple is a company that uniquely has the talent and taste of a good chef, the patent protection is unnecessary. They will be able to continually outdo other companies that don't have the same talent.

Arguing that a company has so much talent and is so successful that it needs legal protection seems absurd to me.

belorn 6 hours ago 1 reply      
Its an old idea to remove everything until just the essence is left. Apple succeed in identifying markets that was making over complex products.

The silly part is that complexity goes in cycles. New products need to differential themselves with old ones, so they add new features. After a while, you end up with a car stereo with 20 buttons, and suddenly a "new" competitor comes out with a clean design with just 3 buttons and the 20 button design looks ill-designed in comparison.

Stereos is one of the earlier examples, but you can see the same phenomena in web-design with today's White and clean design vs the old dark and complex design.

cageface 2 hours ago 0 replies      
Right, because once that chef figured out just the right twist on those common ingredients no other chefs in the world were allowed to make mayonnaise any more and we would have no fine cuisine without extensive IP protection enforced by law.

Oh, wait...

pippy 6 hours ago 1 reply      
>Software is all zeroes and ones, after all. The quantity and order may vary, but that's about it. Hardware is just protons, neutrons, electrons and photons buzzing around, nothing original. Apple didn't “invent” anything, the iPad is simply their variation, their interpretation of the well-known tablet recipe.

You can make cookies and cakes from the same raw ingredients, doesn't mean that the person who invented cookies also invented cake.

Also the 'Software is all zeroes and ones' argument irks me. It's the bridge between an expensive paperweight and a practical device.

smartkids 4 hours ago 1 reply      
The iPaq was/is a fantastic device. You could attach peripherals. Try doing that with an iPad. I could put an entire OS on a CF Card and assuming I could boot from the card, the expanded functionality is limited only by the hardware specs. They are durable. I've seen consumer electronics businesses still using them to track inventory.

I wish HP would revive the iPaq.

My only imagined use for an iPad is as a portable display. I want the Retina quality, but I have a more powerful hardware to attach and I need a real keyboard. There's nothing the iPad can do that my open, unlocked hardware cannot do.

I believe it was even possible to attach a real keyboard to an iPaq. That's the kind of flexibility I want. I can get data into and out of the device in any number of ways, without hassle.

pge 6 hours ago 0 replies      
I think this article makes some good points but overlooks the crucial question - what inventions are patentable? Like a great chef that combines known ingredients, apple has made excellent products, but apple has not produced new innovation that is worthy of patent protection. Apple's path to success should be winning customers, just like the chefs he describes, not through patent litigation.

The examples given by the OP actually support this point. It's true that Einstein didnt discover relativity first - both lorentz and poincare had worked out the mathematics, but Einstein articulated the concepts best. I often use this case study as an example of how innovations often arise independently from different inventors simultaneously because the conditions are right. The other famous and illustrative example is newton and leibniz inventing calculus independently.

As a VC, I see this all the time - the market conditions are right for a new idea, and suddenly 4 or 5 companies appear doing variations on the same thing, none aware of the others. Let good execution and the market decide which one is best, not the date on a patent filing for something each came up with on their own.

wallflower 4 hours ago 0 replies      
A better article on the Counternotions site, "Why Apple doesn't do 'Concept Products'"

"Concept products are like essays, musings in 3D. They are incomplete promises. Shipping products, by contrast, are brutally honest deliveries. You get what's delivered. They live and die by their own design constraints. To the extent they are successful, they do advance the art and science of design and manufacturing by exposing the balance between fantasy and capability."


ThomPete 6 hours ago 1 reply      
Apple invented the ability to take geeky stuff and turn it into cool stuff for normal people.

That's a pretty big feat in itself.

endlessvoid94 6 hours ago 0 replies      
A breath of fresh air.
noirman 5 hours ago 0 replies      
Of course Apple did not invent anything. Steve Jobs did.
jpincheira 4 hours ago 0 replies      
Come on. Shut up. I use Windows on a Dell laptop and I can tell you that's not true. (Pun intented)
hastur 5 hours ago 0 replies      

so what's new

they did other stuff: made things better, connected in new ways, etc.

i still don't like their stuff - for aesthetical reasons

so what

get a life

al_biglan 5 hours ago 0 replies      
This is the same argument aimed at Microsoft in the 90's. next big tech company will get the same thing in 20 years. Meh.
Tichy 6 hours ago 1 reply      
Patents apply to the recipe, though, not the actual mixture.
zerostar07 6 hours ago 2 replies      
Do we to start another debate with an inflaming article?
The Next Step in Apple's Thermonuclear War Against Android groklaw.net
62 points by esolyt  5 hours ago   50 comments top 8
spaghetti 1 hour ago 1 reply      
I'd love to see control of the "mobile space" be in the hands of individuals instead of large, for-profit and often evil companies. Clearly mobile computing is here to stay. If it wasn't then companies like apple and google wouldn't be fighting so hard for control.

Imagine a decentralized app store where people can make a living without having to deal with apple's ridiculous review process or ideas of what's appropriate. I suppose the major challenge is getting all the non-geeks into this ecosystem and away from iTunes and other privately owned app stores.

This is obviously wishful thinking. However given the importance of mobile computing going forward I think it's worth imagining the best mobile ecosystem possible.

othermaciej 3 hours ago 5 replies      
Does anyone see groklaw as a credible source any more? They used to have good reporting, but all I've seen since the Apple v Samsung verdict is spittle-flecked ranting.
verganileonardo 4 hours ago 2 replies      
> These Four Horsemen of the Android Apocalypse are patents for what Apple claims are “key” product features -- “Slide to Unlock,” “Text Correction,” “Unified Search,” and “Special Text Detection.

Wasn't "Slide to Unlock" invalidated in the process against HTC?

DanBC 3 hours ago 3 replies      
Groklaw links to a BBC interview with the jury foreman.

> we as jurors were sworn to abide by the rules and the stipulations in law as they exist today, at the time we made the decision.

Can't jurors just ignore laws that they feel are unjust?

adsr 3 hours ago 1 reply      
Why would you need bounce back at all? It's one of those things that gives iOS it's personallity, you could definately make a touch based user interface without bounce back.
dvhh 3 hours ago 1 reply      
somehow Apple reminded me of monsanto. Maintain your dominant position on the market by agressively suing everybody.
marginalboy 3 hours ago 1 reply      
One can't help but suspect these players are acting in concert, to manufacture the most ridiculous exercises of technology patent abuse, so as to motivate the public and Congress to advocate for a reconsideration of the whole mess...
Steko 3 hours ago 4 replies      
Some companies will have to pay a slightly higher per device royalty than they had wanted to and/or remove/modify some functionality. It's the Apocalypse! It's Thermonuclear War!
Samsung flew bloggers to Berlin, then threatened to leave them there thenextweb.com
216 points by rounak  7 hours ago   47 comments top 14
eddanger 6 hours ago 1 reply      
The only winner out of this is Nokia who paid for the hotels and flights home of those bloggers. http://twitter.com/clintonjeff/status/242358009249026049
KaoruAoiShiho 6 hours ago 3 replies      
I don't get this story. They were asked to wear a samsung t-shirt. That's not how shills typically work, isn't it typically, pretend you're independent, give good review, rather than, pretend you're part of our PR team?

Why would it benefit Samsung to have random Indians bloggers demoing devices in Berlin?

Something about this story doesn't make any sense.

indrax 5 hours ago 0 replies      
This is a human trafficking story.

Clearly this is not remotely as horrific as the usual connotations, but the structure of bait-and-switch coercion is very similar.

Roritharr 5 hours ago 1 reply      
I'm part of the Samsung Mobilers program which is very similar to what they did.

I've never experienced as much pressure as described here, but i guess the Indian Samsung subsidiary is managed by different minded people.

It was clearly just a communication problem. Samsung expects you to do things at these events and you get your trip and stay for free in return, sometimes a little cash on top.

Noone wants to use the word work, for all the red tape this would create...

spartango 5 hours ago 1 reply      
From the start, the offers that Samsung made seemed ethically questionable. Yes, they may be common, especially where review device access is limited, but they seem very much like bribes.

I like The Verge's ethics statement, which they post publicly, for this reason.


"We do not allow trips or any portions of trips (including but not limited to airfare, hotel, or car rentals) to be paid for by third parties (these are known in the industry as 'junkets')."

They make expectations for readers and device-makers crystal clear.

philhippus 5 hours ago 0 replies      
Pretty sure that had they gone to the German authorities and explained how Samsung "trafficked" them into the country with an expectation of being provided a ticket home, Samsung Berlin would get a call from German immigration - and promptly pay for the tickets.
truxs 6 hours ago 1 reply      
This ain't new for Samsung, they did the same during the olympics.

Officially they were invited to live the games from the inside but in the end they worked as Samsung publicist for free.


grannyg00se 6 hours ago 1 reply      
How could this scenario have gone well for Samsung? Did they think the bloggers would suddenly change their moral stance and do a complete 180 on what they had been insisting for weeks?

It seems it would be much easier to find people who are willing to be brand ambassadors and be up front about it if that's what you are looking for.

fruchtose 5 hours ago 0 replies      
This is the ugly side of "new media"--the big companies are able to push around the little people who lack the support of their own big companies.
azakai 6 hours ago 1 reply      
"Use your left/right keys to browse stories" - seriously? A single click of the left or right arrow moves to a completely different page?
Empro 7 hours ago 1 reply      
Wow. Samsung really needs to clean up its act.
denzil_correa 7 hours ago 1 reply      
Raises quite a few issues about conflict of interests. Should there be a rule for such declarations by writers who are paid to promote brands ?
sbierwagen 6 hours ago 0 replies      
The extended quotes, and TNW's reputation, makes me think that this may be blogspam, but I can't find an original post.
revelation 6 hours ago 3 replies      
Being stranded in Berlin is certainly not the worst that could happen to you.

That said, this is exactly what happens when, as a journalist, you start to blur the lines. As a reader it's hard to feel any sympathy when reading paragraphs that try just a bit too hard to rationalize the behavior:

Again, a reminder " Behavior such as Samsung's is not uncommon in the world of tech coverage. It's perhaps considered more normal in some parts of the world

Why Nonprofits Are More Like Businesses Than You Realize seliger.com
7 points by jseliger  2 hours ago   discuss
Typicons - free-to-use vector icons embedded in a webfont kit typicons.com
63 points by vacipr  9 hours ago   8 comments top 8
DigitalSea 10 minutes ago 0 replies      
My opinion is that the fonts are definitely not correctly optimised. I think they require some serious hinting, the anti-aliasing on a lot of the vertical lines is really really bad. Great idea and I applaud the effort, but Typicons need some serious work before I'd consider using them.

As already pointed out, reminds me of Font Awesome: http://fortawesome.github.com/Font-Awesome/ - which did it correctly in my opinion and perhaps Typicons can learn a thing or two from that.

pbhjpbhj 5 hours ago 0 replies      
What's the license?

Note that Font Awesome has the license and details of the attribution requirements plainly displayed on the front page - "free-to-use" is not nearly clear enough.

chevas 7 hours ago 0 replies      
ck2 4 hours ago 0 replies      
Not sure if it's just a problem with this particular demo or font icons in general but they are not anti-aliased for me in Firefox 16 on Windows XP - the rss icon is the best one to examine for the problem.

I guess we are a year away from these being ready for mainstream?

johanbrook 6 hours ago 0 replies      
Ironic that the site creator has deactivated zooming on iOS devices, so the font icons' vector capability is in vain.
Groxx 7 hours ago 0 replies      
I get pretty bad AA on many vertical lines, once it zooms in on mouseover. Seems like it needs hinting?
cristianpascu 8 hours ago 0 replies      
This is a good looking set. Maybe the creator should consider contributing some of them to the FontAwesome collection. Just a thought!
pedelman 7 hours ago 0 replies      
For people interested in treating fonts as vectors, check this out: http://keyamoon.com/icomoon/

Also allows you to upload SVG files to create your own fonts. I have had mixed results with the upload, but I plan on using this for most of my upcoming projects.

US flew Global Hawk spy drone missions from Australia abc.net.au
5 points by bootload  1 hour ago   discuss
Attention Arbitrage nateberkopec.me
26 points by nateberkopec  6 hours ago   4 comments top 4
jval 5 minutes ago 0 replies      
This is a great piece.

Building and scaling a business model is just as difficult and important as building and scaling your technical platform. I think that people often talk in general terms about how they plan to monetise, but miss the details. It's one thing to have an overall plan (e.g. advertising) but as we start to get more detailed these plans tend to fall over.

Advertising --> Types of advertising (Banners? Sponsored Results/Posts? etc) --> Precise details (Where are you going to put the banner? How big is it going to be? How are you going to source your inventory?)

Bubble 2.0 was built by having big web 'properties' value themselves the same way they did 10 years ago. But as the article validly points out, the competition for attention isn't what it used to be, so the valuations don't correlate.

New hot startups are no longer 'properties' in the traditional sense and actually build businesses with technology as an enabling factor rather than an all-consuming strategy. I'm looking forward to seeing how these turn out!

jacques_chester 2 hours ago 0 replies      
Basically, supply and demand. Let's look at advertising, since it's still the major way for publication-oriented websites and consumer webapps to make money.

1. Demand as in demand for advertising.

This is actually a proxy for the ultimate consumer. Advertising demand is driven by the amount of money that can be made from the viewers of the advertising.

But this changes relatively slowly. With 2 billion internet users we've either reached or are fast approaching the inflection in whatever sigmoid function describes the growth in users. After that taps out growth will rely entirely population growth -- ie, internet businesses will be constrained to the same dynamics as everyone else.

2. Supply as in the supply of "inventory".

Inventory is just ... a box where you put your add. Inventory does not require 9 months to create followed by >10 years to raise. It is being created fantastically quickly.

So if supply is expanding exponentially, but demand only sigmoidally ... well it's not pretty for anyone outside the top few percentiles.

001sky 5 hours ago 0 replies      
“Outsourcing monetization” has got to be the worst oxymoron I've ever heard. The purpose of a business is to make money, and it doesn't take a business school degree to know that you outsource to reduce costs, not to raise revenue.

--While this is true to some extent, a few counterexamples are worth noting. Namely, transactions services. These are of a real business that scale. Think: paypal, visa, etc. There is also a class of services that are pre-transaction: like amazon referral fees. These services all monetize through external sales, but represent real-value-added services. They are orthogonal to attention, in that the make life easier, simpler, less complex.

dllthomas 4 hours ago 0 replies      
> The problem with a “just build a great product” strategy is that it assumes that all created value can be captured.

Well, really it assumes that enough created value can be captured.

Lessons Learned While Building Reddit to 270 Million Page Views a Month highscalability.com
103 points by rjim86  13 hours ago   49 comments top 8
arkitaip 11 hours ago 3 replies      
I love reading about how companies scale their BigHuge data but it bothers me that we still haven't reached the point where scalability is a commodity instead of a patchwork of technology that everyone actor solves in their own way.
chaz 12 hours ago 1 reply      
Their most recent infrastructure blog post was from January 2012, and shows that they're using Postgres 9, Cassandra 0.8, and local disk only (no more EBS). I'm curious if the recently-announced provisioned IOPS would enable them to go back to EBS.


ecaron 11 hours ago 3 replies      
Only disagreement (although I feel like I'm arguing w/ Linus about git) is don't memcache session data (lesson 5.) Memcache's 1mb max-block (exceeding that removes too many performance perks to be considered viable) introduces a "I need to constantly worry about my sessions getting too big" mental overhead that isn't worth it.

Go with Redis for storing session data.

citricsquid 12 hours ago 0 replies      
Article is from 2010, if I remember correctly their architecture has changed substantially since this article.
ndemoor 12 hours ago 2 replies      
"Instead, they keep a Thing Table and a Data Table. Everything in Reddit is a Thing: users, links, comments, subreddits, awards, etc. Things keep common attribute like up/down votes, a type, and creation date. The Data table has three columns: thing id, key, value."

I hope they introduced some NoSQL sweetness by now.

petercooper 8 hours ago 0 replies      
It's pretty cool how nowadays Redis can cover several of those concerns quickly and simply (open schema, caching, replication..)
dotborg 7 hours ago 3 replies      
How are PHP, RoR, node.js, Perl/CGI etc. engineers supposed to build offline processing? Crontab?
surferbayarea 11 hours ago 1 reply      
that's just ~12 queries/sec..not that huge a deal..
Live air traffic of the world flightradar24.com
337 points by vibrunazo  1 day ago   71 comments top 32
state_machine 22 hours ago 4 replies      
Pretty, though, just to nit-pick since I happen to be sitting on a plane, seems to be not quite live: http://cl.ly/image/002u2g152Y0T
zrail 9 hours ago 0 replies      
This is very neat! I love it. Uh, I don't know if this is an error or if they're doing flight testing, but this is a very weird track: http://fr24.com/KAL32
pitchups 22 hours ago 3 replies      
Very impressive indeed. One of the coolest features of the site is the Cockpit view - a very creative use of Google Maps and graphics to give you a Flight Simulator like view from inside the cockpit of any of the thousands of aircraft in the air.
Zaheer 23 hours ago 0 replies      
Wow. That looks beautiful.

Note that just because there aren't that many planes over Africa or other places doesn't mean there aren't planes there. From their site: "Today about 60% (about 30% in USA and about 70% in Europe) of the passenger aircraft and only a small amount of military and private aircraft have an ADS-B transponder."

joshzayin 22 hours ago 1 reply      
Cockpit view is generating an error: "The Google Maps API key used on this web site was registered for a different web site. The developer of this web site can generate a new key here.

(here links to https://developers.google.com/maps/)

69_years_and 22 hours ago 2 replies      
Boats more your thing?
colinhowe 16 hours ago 1 reply      
I never truly appreciated how many planes are in the air until now..
mikeash 1 day ago 0 replies      
Really cool, but the constant nagging popups telling me to download the app instead were unbelievably annoying.
001sky 4 hours ago 0 replies      
Thanks for posting this is amazing.

The cockpit view is very cool. Reminds me: You can climb Mt everest also now on Google Earth.

South Col etc. At 8000m, somewhat crazy but similar views =D

smcl 18 hours ago 1 reply      
Just remembered that I found Donald Trump on this a while back - he was flying in to my home city of Aberdeen:


chmars 16 hours ago 0 replies      
A similar project with an academic background is 'AirTraffic LIVE':


It was created by students of a Swiss college of applied science in 2007. The test site is focus on Zurich International Airport but they have completed other projects based an their research. A spectacular example is a globe showing air traffic world wide created for a science museum:


There is also a Google Earth extension for private use:


piffey 20 hours ago 1 reply      
Anyone else surprised by the number of flights in the air at any given time? I never even imagined that there was this much activity even though I'm a frequent flyer.
progrock 7 hours ago 0 replies      
Love it! There's alot of activity! How many planes, how many people up in the air in an average moment? All this talk of a third runway / increased flight support in the UK, but really - how much oil is left - how sustainable is this industry?
mirsadm 9 hours ago 0 replies      
Cockpit view is brilliant. Just watching a plane land back home right now :)
jvandenbroeck 18 hours ago 0 replies      
Cool!:p I'm living next to an airport and it looks like it doesn't pickup everything (or some too late), but just a few minutes ago I heard a plane coming by & it was also on flightradar =')

Idea: people give their location & the app says when to expect noise from airplanes and when it will be away=)

Andrew_Quentin 13 hours ago 2 replies      
I thought there would be a corridor of never ending planes between north America and Europe. According to the map, there are no transatlantic flights.
kgarten 20 hours ago 0 replies      
nice ... are there any APIs for getting up-to-date flight traffic? Might be fun to play around with it.
josscrowcroft 17 hours ago 0 replies      
That is absolutely incredible!
nja 22 hours ago 1 reply      
Interesting...does JetBlue use two types of radar? I'm often seeing two icons for one flight: http://imgur.com/a/kENc0
bencoder 18 hours ago 0 replies      
Enjoying it combined with ATC feeds from http://www.liveatc.net
dhughes 12 hours ago 0 replies      
planefinder.net is similar.

I live in south-eastern Canada and I can confirm it's very accurate, big jets to and from Europe constantly rumble overhead.

jasonzemos 22 hours ago 0 replies      
Looks like the Virgin Atlantic and British Airways 747's hit 600+ knots on the redeye across the pond.
avaku 18 hours ago 0 replies      
Why cockpit view is not working for me? It says: The Google Maps API key used on this website was registered for a different website. The developer of this website can generate a new key here.
cefarix 23 hours ago 4 replies      
The flight map over South Asia, China, and most of South America looks to be very incomplete.
nodesocket 1 day ago 1 reply      
Amazing, didn't realize at any given time, how many planes are in the air around the world.
nja 22 hours ago 0 replies      
This is crazy cool. I love the cockpit view!
curiousDog 22 hours ago 0 replies      
This is beyond fantastic. Just wow!
mukaiji 19 hours ago 0 replies      
bravo! Way to kill my productivity.
khet 17 hours ago 0 replies      
Amazing how many will not find this amazing.
rapidstuff 23 hours ago 0 replies      
Absolutely amazing!
ricksta 22 hours ago 0 replies      
no plane around beijing?
arunoda 22 hours ago 0 replies      
Wow, this is amazing.
Heading Out on Your Own: 31 Life Skills in 31 Days artofmanliness.com
130 points by ginozola  13 hours ago   13 comments top 10
datr 11 hours ago 0 replies      
I liked the section titled "Good manners are not stiff, formal, or awkward." It reminded me of a story I was told of a dinner party at which one of the guests suffered from cerebral palsy. The host seeing that the guest was having trouble with her knife and fork immediately put down her own cutlery and started eating with her hands. Once the guest noticed this she felt she had leave to do the same. The story's a good example of "people over process".
natep 5 hours ago 0 replies      
For anyone that wants to read these posts in an RSS reader one day at a time, I suggest Feed Playback [1]. Just enter the feed URL [2] and a start date of 2012-08-01. Unfortunately, the feed is mixed in with every other update that gets put on the site, so it will take longer than a month to complete.

[1] http://www.streamspigot.com/feed-playback/

[2] http://feeds2.feedburner.com/TheArtOfManliness

tgrass 6 hours ago 2 replies      
It's a sad state of men when day 15 is How to Change a Tire, when it ought to be how to fix one: http://www.alpharubicon.com/bovstuff/tirepluguzi.htm
taroth 36 minutes ago 0 replies      
In #13 Networking:
>"If you're in San Francisco looking for a tech job, you'll want to be using the latest connection apps"

I've never seen this in the LA community. Is this the norm up north?

SonicSoul 11 hours ago 0 replies      
this is great. i esp wish someone encouraged me to make an effort and "Day 22 - Learn how to small talk".

growing up i hated small talk and this translated very well into hobbies involving staying home and hacking things, and later doing that at work.

it wasn't until much later that i realized how important it is to make real human connections and be able to interact with strangers. This seems to be harder to get right the older one gets.

ohashi 11 hours ago 0 replies      
The most relevant one I saw and read was the Networking one. I think a lot of people I meet (or see at networking events) need to read that.
mmanfrin 9 hours ago 0 replies      
AoM is one of the few blogs that I feel genuinely compelled to read. Very good writing, and about skills/knowledge that we all need in our life (regardless of profession).
peter_l_downs 8 hours ago 0 replies      
This is awesome " I just recently moved out of my parents' home for the first time and have been doing my best to adjust to life "on my own". Thanks for posting this.
vishaldpatel 8 hours ago 0 replies      
The article on stress management should be listed first.
mukaiji 10 hours ago 0 replies      
oh... how i wish i would have read "Living With Roommates Tips" beforehand.
Self-Driving Cars Approved by California Legislature cnbc.com
32 points by spdy  8 hours ago   1 comment top
savramescu 4 hours ago 0 replies      
I'm really curios when it's going to be approved somewhere in Europe. That's going to be a big signal of moving towards a finished product.
Memristors' one-year delay will hit IT in the wallet zdnet.com
61 points by ck2  11 hours ago   46 comments top 13
ck2 10 hours ago 5 replies      
Why on earth was my title replaced? Sigh.

Revolutionary memristors purposely delayed to prevent cannibalization of flash

It's 100% accurate and more relevant.

pbharrin 9 hours ago 2 replies      
This is a stupid article. The author makes the assumption that HP and Hynix COULD ramp to high volume production, however it is clear from HP's statement that this is not true:

"As with many other ground-breaking technologies being developed at HP Labs, HP has not yet committed to a specific product roadmap for memristor-based products,"

Let me also say that I worked in the non-volatile memory business for seven years. Any organization that can scale better than its competitors will do so, unless they don't like making money. End of story. If HP and Hynix are not ramping to high volume, it's because they are not ready, not some stupid conspiracy theory that they want to stagnate the development of cloud-buzzword "technology".

mtgx 11 hours ago 1 reply      
What short-sighted thinking. This is why you don't see much innovation from the big companies usually. This is like Apple thinking they shouldn't release the iPad because it would cannibalize their Mac sales. Instead of being an also-ran in flash, Hynix/HP could be the leaders in memristors. But it seems you won't see them thinking like that.

Oh well, I guess Samsung will take the lead in that market, too, then, since I think they were the other main company working on memristors.

ahuibers 8 hours ago 0 replies      
I haven't seen any lab demonstrations of multi Mbit devices. Until we see that this is years away from production in my experience as a device physics researcher. All storage technologies so far have taken 10 years plus to have large market share and I don't see this as an exception.
ryanmolden 10 hours ago 1 reply      
Does HP have patents here that would prevent others from introducing this tech? Or are they/Hynix simply the only ones with experience with the tech to pull it off? This is such a clear example of one of the fears spelled out in The Innovators Dilemma it isn't even funny.

Edit: whoops, mtgx says Samsung is also in this space, so it sounds like patents can't be an obstacle, or at least until the suing starts.

batgaijin 10 hours ago 0 replies      

It seems like HP still hasn't been issued patents for it (at least by my googling) so I guess they are happy to just sit on their tech.

Also, Memristors are already being actively used for military technology, most notable the SyNAPSE project.

Klinky 8 hours ago 0 replies      
I'll believe memristor tech when I see commercial products of it available. I don't think it's far fetched that a company is slow to shift it's mainstay, profitable business product to an unproven new technology that has no mainstream commercial products out yet. Cutting edge tech is always rife with delays and shortcomings that weren't anticipated until the rubber actually hit the road.

While many in these comments are touting "double storage & reliability for the same price", that would be for raw manufacturing costs, not including all the R&D money and the cost of fabrication plants that must be built or retooled. It is highly unlikely consumers will see those benefits anytime soon.

Tagbert 10 hours ago 2 replies      
Could they not charge a premium price for memristors early on to prevent flash cannibalization and then bring the price down as they transition their output?
adaml_623 4 hours ago 0 replies      
This catchy headline is essentially saying that company X is going to watch their growth and expenditure and maximise their profits in a manner which won't benefit consumers.

The headline is interesting because it's stating the obvious, standard business practice but possibly they have hit on the only way to glam up the incredibly boring story of 'product will be delayed by 12 months'.

zrail 10 hours ago 1 reply      
I don't see why this is a big deal. It's a year, so what? We've waited decades, a year isn't going to hurt anything. Hynix is a big business that is heavily invested in flash and it takes a lot to completely change the core of your business over, especially when you have to possibly build new fabs and fab technology.
ck2 11 hours ago 0 replies      
This might also explain why SSD prices are dropping fast.

50 cents per GB now and falling.

ricardobeat 10 hours ago 1 reply      
It only makes sense to continue selling the older, higher margin products until a competitor comes up. They are asking for it.

I would even root for Samsung on this (also because it just finished my laundry).

mikecane 11 hours ago 0 replies      
When I first heard of HP's memristors, I was looking forward to them using it in the TouchPad, to really compete against Apple. Now this. FAIL all around, HP.
3.7X speedup from removing a call to sleep webkit.org
132 points by nexneo  17 hours ago   62 comments top 14
edw519 15 hours ago 3 replies      
True story...

Client gives me a Help Desk Ticket. User claims batch job use to run in 5 minutes but now runs for hours. The logs confirm this. The commit logs show that an offshore programmer forgot to remove a 10 second sleep command from inside an iteration (for debugging I presume) before promoting to production. I removed it and got a 100X improvement in throughput.

My client said that now his user loves him; what did I do to fix it so fast?

When I told him that I removed the Sleep, he said, "No! No! No! Change it to a 5 second Sleep so I have something to give him the next time he complains!"

othermaciej 9 hours ago 1 reply      
I am the one who originally added this sleep call to WebKit, when we first imported TCMalloc to use as our custom allocator. It was indeed there for a reason, but that reason is not applicable to WebKit's allocation patterns. TCMalloc was designed for a server workload, over time we have adapted it more to the unique needs of a browser engine. This change may help other operations, but probably not as much as the GC benchmark in question.
pmjordan 16 hours ago 3 replies      
It's interesting how spinlocks are making a comeback in user space code. They are widely used in kernel code, but their use has previously been discouraged in application code. As this example illustrates, it's pretty easy to get them wrong...
cpeterso 4 hours ago 0 replies      
Why did they roll their own spinlock implementation in the first place? They could encapsulate the lock primitives provided by, and optimized for, each platform: CRITICAL_SECTIONs on Windows, futexes on Linux, and pthread_spin_locks elsewhere. iOS and OSX have some Mach-specific spinlocks (OSSpinLockLock?), but I don't have any experience using them.
osivertsson 16 hours ago 0 replies      
It was "only" on one particular benchmark, still it shows that profiling often give you surprising results that nobody familiar with the code anticipated.
tlrobinson 3 hours ago 0 replies      
aristidb 13 hours ago 0 replies      
There was a long comment before the sleep. It seems likely that the sleep was there for a reason.
zizee 16 hours ago 2 replies      
So if I understand this correctly the 3.7x speedup is a speedup in garbage collection and will have a positive effect on the speed of the browser which is nice.

But what sort of effect will this have on page rendering? Anyone in the know care to comment on the specifics?

edit: improvements to my shoddy wording.

swah 13 hours ago 2 replies      
Are sleeps() always necessary in multithreaded programming?
grimebox 16 hours ago 0 replies      
Anyone wanna supply some context? I have no idea what is going on here
chris_wot 8 hours ago 0 replies      
What's with the old comment "// Sleep for a few milliseconds", followed by the sleep function?
subhro 16 hours ago 3 replies      
Awesome!! I wonder when this is going to get pushed to mainstream Safari.
duncanwilcox 15 hours ago 0 replies      
The code is, appropriately, in the WTF directory.
khet 16 hours ago 0 replies      
Reduced the number of lines in the code base AND improved the product? nice.
Show HN: Create pretty charts on the web directly from excel excharts.com
4 points by mmac  1 hour ago   1 comment top
deyan 21 minutes ago 0 replies      
I am potentially interested in this product, but I couldn't help but have a negative reaction to the bare bones website that is currently up. I understand that entrepreneurs need to build MVPs, etc. but a screen with two very small (i.e. unreadable) images and basically no information whatsoever is just frustrating. I hope that feedback helps to make a more informative page.
Samsung Executives Waited Four Days to Tell Chairman elasticrat.com
4 points by ichiro  1 hour ago   2 comments top 2
jcollins1991 1 hour ago 0 replies      
Would rather post this on the blog, but since I can't I'll put it here...

Given that the "chairman" refers to the chairman (Lee Kun-hee) of Samsung Electronics, consider these numbers:

Samsung Electronics revenue in 2011: $149 billion USD

Samsung Telecommunications revenue in 2011: $21 billion USD

Even in a North American style business he would probably have more pressing concerns than a ruling that affects only a very small part of the business he is overseeing. Given that Samsung is a Korean multinational, we're talking about a completely different culture and business style, where the top few layers of people probably don't need to know about this sort of thing right away.

Dystopian 1 hour ago 0 replies      
I'm thinking it's probably a myth - just like the "announcement" that Samsung paid Apple in trucks of small change.
Knockback.js: Knockout.js magic for Backbone.js github.com
65 points by chinchang  13 hours ago   19 comments top 8
FuzzyDunlop 6 hours ago 0 replies      
In all honesty, I think that the way javascript apps are created needs to be rethought when you have to combine your MVCs with your MVVMs to change a View that's somewhat like a Controller into a ViewModel that can maintain state and pass it to a View.

I'm going to propose the FuckThis paradigm until I can come up with something that makes some sort of sense.

Kototama 8 hours ago 1 reply      
ViewModels differ from Views exactly in this: they own states, methods and properties allowing the views to be logic-less.

If you consider, in Backbone.js, that your templates are the views, then the Backbone Views are a kind of controllers. What are the advantages of Knockback.js here? I'm sure there are but it's not clear to me given the description.

Alternatively in backbone.js you can create object models related to View but that are not part of your business logic.

recuter 11 hours ago 2 replies      
It is a lot more clear from the description on https://github.com/kmalakoff/knockback what this is about:

When I was evaluating client-side frameworks, I liked lots of the pieces, but wanted to "mix and match" the best features. I started with Backbone.js and really loved the Models and Collections, and used Brunch to get me up and running quickly.

After a while, I found the view coding too slow so I wrote Mixin.js to extract out reusable aspects of my views. When I was looking for my next productivity increase, an ex-work colleague suggested Sproutcore, but at the time, it wasn't yet micro-frameworky enough meaning I would need to learn something big and "to throw the baby out with the bathwater" as they say (it is hard to give up Backbone models and collections!). Then, I discovered Knockout and knew it was for me!

Knockout provided just the right building blocks for a layer between my templates and data. As I used it more, I built additional functionality like Backbone.ModelRefs for lazy model loading, localization helpers for truly dynamic views, and most recently, an easier way to sync collections and their model's view models.

So here it is...the refactored and shareable version of my Backbone bindings for Knockout: Knockback.js

pbiggar 3 hours ago 0 replies      
I've tried both Knockout and Backbone with Circle (https://circleci.com), I can totally see the use case here. Knockout is great for syncing between the client and server, and does history and routing well. However, it's not great at model form synchronization.

However, Knockout really excels at that, and allows you to do great live updates.

In the end, we chose Knockout + Sammy.js (history and routing) + manual templating (pretty easy really).

highace 7 hours ago 3 replies      
Am I right in saying backbone and knockout do not solve the same problem? I fancy dipping my toes into a javascript framework, and being primarily a .net dev I'm being pushed towards knockout. However backbone seems to be better adopted and more mature. Is knockback the best of both?

Can another provide examples of where one would be superior over the other?

Todd 9 hours ago 1 reply      
Another alternative: https://github.com/toddlucas/uppercut

This is a simplified version of Backbone designed for use with Knockout and other client side rendering frameworks.

Disclosure: I'm the author

krosaen 3 hours ago 1 reply      
for background, here's a stackoverflow answer explaining how knockout and backbone serve different purposes:


bergie 7 hours ago 0 replies      
Another option of data binding: http://viejs.org/

Uses RDFa annotations to automatically construct views and models.

       cached 3 September 2012 04:02:01 GMT