hacker news with inline top comments    .. more ..    23 Sep 2012 News
home   ask   best   5 years ago   
Apple's HTTP POST caching is a bug mnot.net
46 points by igrigorik  48 minutes ago   2 comments top
AffableSpatula 17 minutes ago 1 reply      
browser vendors should be pressured to up their game with regards to web caching. The cacheability of HTTP is the majority of its value as a network protocol and browsers - the defacto HTTP client - just aren't pulling their weight in this regard.

Support for new cache-control directives like stale-if-error, and some kind of API for negotiating/allocating cache storage for web apps would be a good place to start, imo.

Show HN: My Guild Wars 2 data viz weekend experiment using D3 and SVG filters guildwars2viz.com
26 points by enoex1  1 hour ago   6 comments top 4
DanBC 9 minutes ago 0 replies      
This looks nice! And I like the link between the rings and the bars.

I'm a bit thick. It took me a few minutes to realise that the rings of the circle are independent. For example, there can be female humans, all the humans are not male.

On the bar charts you start at 0% (which is good), but you don't finish at 100%, nor on a similar number. One chart finishes at about 15% and another at about 30%. So at a glance the hight of the bars looks similar. I guess the width of the bars compensates? (And the race bar chart numbers add to 99%. Is that a rounding thing?)

garraeth 28 minutes ago 1 reply      
Very cool!

GW2 is currently my favorite game and am really surprised to see it on HN frontpage.

I'm astounded at the amount of data ANet is making available. And, how cleanly and robustly they've done it - all in JSON.

Another bit you might want to play with is their TP (Trading Post) JSON data. These guys (http://www.gw2spidy.com, source is on github) do a really good job, but I'm sure there are tons more things, and ideas, that can come from all that data!

Good luck!

kmfrk 50 minutes ago 1 reply      
It looks great, but I noticed that the black outline animation lags heavily behind the rest, which makes the site feel incredibly sluggish, even though it's probably only just the one animation that ruins the perception.

Great job all in all.

ksec 24 minutes ago 0 replies      
Loads slow in Firefox. Sigh.....
Fokus - new UI concept that emphasizes text-selection hakim.se
13 points by adunk  39 minutes ago   discuss
Winning on Product, not on Price iamvictorio.us
49 points by iamvictorious  2 hours ago   13 comments top 5
csense 1 hour ago 2 replies      
Really, your desired point on the price/performance tradeoff depends on the specific circumstances of your business.

(a) Being able to compete on price means you have to be really hard-nosed about optimizing every part of your operations to stay lean. Good relationships with suppliers and competent execution of your strategy are paramount -- think Walmart.

(b) Competing on product means you have to be able to blow away your competition. It's easier the smaller your niche is. It's often about having a talented designer -- Steve Jobs, Shigeru Miyamoto, Notch -- together with a support team.

A lot of software is easier to compete on product, because (1) the zero-marginal-cost structure of the industry means that competing on price is a race to the bottom that will ultimately be won by open source, and (2) a lot of software is specialized.

Companies that win in (a) tend to be driven by business types who are good at the stuff they teach in MBA school. Companies that win in (b) tend to be driven by creative types and have a culture where the creatives' vision is the core competency, and the business side is seen as more of a support role. The good news for bootstrap-stage startups is, if you're in category (b), your business chops don't matter as much; the product can make up for a lot, if it's good enough.

So if your team is thin on business talent but has a ton of technical talent, you should use strategy (b). The reverse should use strategy (a) in theory, but in practice is rare among startups; business types who are good at what they do tend to pull down very good compensation, and mostly hang around companies that can afford it. Of course, if you're in YC or funded, you may have access to business-knowledgeable people; but getting to that point usually means you already have a great product and a lot of technical talent.

zaidf 1 hour ago 2 replies      
Makes sense for your case - however, keep in mind there is plenty of cases where the innovation is the very ability to offer something for a lower price. This is more true in manufacturing than software I think though I wouldn't be surprised in the long run if it also begins applying to software.
notatoad 1 hour ago 1 reply      
>To win on product, you have to differentiate

to be clear, you have to differentiate by being better. just being different doesn't count. (I'm looking at you, phone manufacturers)

erikb 47 minutes ago 0 replies      
Of course to really beat all the other people at the price margin and still make plus is a skill not everybody has. cheap is a value like any other, too.

That said, most people do it wrong, when they attempt to do it cheaper. This might make a false impression about that pricing is the problem, were actually the people doing it, are what's wrong.

shasta 2 hours ago 1 reply      
And if you can't win on either, win on patents.
Do Web 2.0 Companies Really Have The Best Technical Talent? bozho.net
22 points by bozho  1 hour ago   28 comments top 13
patio11 1 hour ago 1 reply      
What would your job ad say if it was designed to solicit applications from 20 to 22 year old males willing to work long hours for peanuts?

"Come be a frankly mediocre programmer writing pedestrian code to solve unimportant problems"?

A job, like any other product, has to be sold. Beer ads don't say "makes you stupid and fat; occasionally ruins lives; it's going to kill some of you; no, really, you will not have beautiful women drape themselves over you, that virtually never to men sitting at home playing Xbox and drinking beer does not make it substantially more likely." For a reason. Similarly, there's the intimation of a cultural/status/etc component to the product offering for working as an engineer at one of these places. (n.b. Not unique to working at them, incidentally. i-bankers, Japanese salarymen, nurses, and PhD candidates in English all get a sales pitch -- in no case is it totally 100% representative of the actual job because, hey, sales pitch.)

andrewvc 0 minutes ago 0 replies      
Talent takes many forms. For web 2.0 companies engineering matters, but hustle product vision and teamwork come together in a unique way. Talent is not measured on a single axis.

Most startup engineers couldn't build design and build something like hadoop, but I doubt many database authors would perform well at a startup.

In addition, quick and dirty hacks can be a virtue in a startup (Say leading up to an important product demo). Some engineers have a good sense of what it takes to meet a tough deadline. A language designer does not face that challenge.

stevoski 1 hour ago 1 reply      
It would be unfortunate if the best technical talent went to "web 2.0 companies". I'd hope the best are working on our operating systems, our programming languages and their APIs and virtual machines, our database servers, and other technology that "web 2.0 companies" and untold others rely on.
zxcdw 18 minutes ago 0 replies      
I'm afraid to mention(considering many might feel offended) that IMO most "web 2.0 startups" are full of ill-competent self-claimed "software developers" who have 3 months of experience in PHP and JS and then build "cool apps", who then are considered "technically talented" just because their "cool app" has had some exposure on different media which has brought in some users.

There's much more to actual technical talent when it comes to building software than what you get from year or two of experience, let alone 3 months. Though, in this era when you can actually build usable software products in weeks, if not days(as there have been many interesting "weekend projects" mentioned here in HN) this doesn't matter as much than it did in the past. For example game development has never been as easy it is today, there are thousands if not millions of young people creating games for mobile devices. Some of them succeed in making money but that is most certainly not due to technical talent, as many may easily think.

Back in 90's kids(under 20 yo) literally wrote their own 3D renderers in software because there was no hardware acceleration available. Now 20 years later, how many teens would be able to write their own 3D rendering engine even when using graphics APIs, let alone when doing it all in software? Although the amount of programmers has gone up, I think amount of competence has gone down simply because technical competence is far less needed these days.

These days there's far less need for technical talent when it comes to "building web apps" than what one would've needed say 10 years ago to create a simple client-side desktop program. Yet, now more than ever there's talk about technical talent. Strange.

moocow01 1 hour ago 0 replies      
Probably the truth is that most Web 2.0 companies have a mixed bag but by the nature of their business and the startup ecosystem every web 2.0 company has to hype their business and consequently hype the talent they have.

Just about every startup will tell you they have A-players, rockstars, etc etc - its called marketing.

carbon8 8 minutes ago 0 replies      
Another strong possibility is that it's a management problem. The engineers at these companies could be pulling their hair out over this kind of stuff, but it's not sexy or "product" enough for the PHB segment to want to do anything about it or even understand why it's a problem.
halis 21 minutes ago 0 replies      
Well I think many of these startup companies are looking for "rockstar", "ninja" code warriors..whatever that means. I think they are trying to attract the most promising talent in the next generation of up and coming programmers. They try to search these candidates out by making a github profile (or some other open source contribution) mandatory.

I think experience does matter though. Even a talented young programmer can get into big trouble by trying to do cute and clever things. They may not have the experience to know that generic database design is generally a really horrible idea, etc.

rayiner 1 hour ago 2 replies      
His whole article is premised on a false dichotomy:

> So is it really the case that these silicon valley/web 2.0 companies have the best developers, or they are just regular companies that have average developers doing stupid things?

Third option: silicon valley has (many) of the best developers, but great developers do stupid things al the time.

robdoherty2 1 hour ago 1 reply      
One possible explanation for this is that talented though start-up engineers may be, they are in many cases over-worked and under pressure to put out revenue-generating product.

As basic and important as salting + hashing passwords is, it is the sort of task that ends up on the technical debt list or scrum back-log, never to be looked at until it becomes a problem.

HaloZero 1 hour ago 1 reply      
I think the biggest factor that might come into play though is that even top developers write bad code sometimes (weird behaviors that they think might work out but prove to be hard to maintain) especially when working on things they've had no experience in the past writing. (The first facebook api was probably not written by somebody who's done a lot of work with APIs). Or some other legacy issues caused the API to be written in non-optimal way.
true_religion 42 minutes ago 0 replies      
> I took a look at the code of reddit, which (even though I'm not a python developer) struck me with some really odd stuff (won't go into details).

Please, please go into details on this because it is actually a code base that we can all look at and review.

mempko 51 minutes ago 1 reply      
Ask yourself a simple question about web 2.0 engineers. Are gold diggers the paragon of quality motivated engineers.
elchief 1 hour ago 1 reply      
Re Salesforce: we were evaluating them, and I discovered you have to pay extra if you want non-company customers. I realized it wasn't a ploy for extra money, they just hadn't used the Party model, and thus had to do extra work to have individual and company customers. We didn't go with them.

I have a feeling the best technical talent still goes to big blue for the "impress mom" rep and R&D, or NSA for the spook factor

Toomâ€"Cook multiplication wikipedia.org
32 points by ahanjura  2 hours ago   4 comments top 2
psykotic 5 minutes ago 0 replies      
Daniel J. Bernstein has a nice survey paper covering all the major multidigit multiplication algorithms:


What I like about it beyond the comprehensive coverage is that it explains the mathematical structure that underlies each of the algorithms. Unfortunately for HN readers, the paper's intended audience is mathematicians, but with a bit of background in algebra you might still be able to glean some useful insights.

As an example, I'll try to describe the Karatsuba trick.

We want to multiply two linear polynomials p = a + bx and q = c + dx. That is, we want to calculate the coefficients e, f, g in (a + bx)(c + dx) = e + fx + gx^2 in terms of a, b, c, d. The standard algorithm is e = ac, f = ad + bc, g = bd. This has 4 multiplications and 1 addition.

Here's the Karatsuba trick as usually presented. The word 'trick' is apt because this makes it seem like pulling a rabbit from a magician's hat. Let u = (a+b)(c+d) = ac + ad + bc + bd. Then f = u - ac - bd = u - e - f. Thus Karatsuba's trick calculates u = (a+b)(c+d), e = ac, g = bd, f = u-e-f. This has 3 multiplications and 4 additions. We've saved 1 multiplication at the expense of 3 extra additions.

If we now apply Karatsuba's trick recursively in divide-and-conquer fashion to the left and right halves of higher-degree polynomials, we get an algorithm that is faster than the standard algorithm even if we assume that scalar additions and multiplications have the same cost. The standard algorithm has cost O(n^2), where n is the degree of the polynomials, and Karatsuba's algorithm has cost O(n^(lg 3)), which is around O(n^1.6).

So what's the structure underlying Karatsuba's trick? Well, you might have noticed that u is p q evaluated at x = 1. You can evaluate a product of polynomials without multiplying them out first because evaluation is a homomorphism, so u = (p q)(1) = p(1) q(1).

Evaluation is a lossy (non-injective) mapping, so we will have to evaluate their product at three different points (since the product is quadratic and hence has three coefficients) to recover the original product uniquely. We've already evaluated at x = 1. Two other obvious candidate points for cheap evaluation are x = 0 and x = infinity. Evaluating at x = 0 just gives the constant term, so (p q)(0) = a c. Evaluating at x = infinity (make the substitution w = 1/x, clear denominators and evaluate at w = 0) gives the highest-degree term, so (p q)(infinity) = b d.

Now that we've evaluated the product at three points, all we have to do is interpolate between them with the Lagrange formula to recover the product.

This evaluate-and-interpolate approach is also what underlies FFT-based multiplication algorithms. More generally, rather than use just evaluation, you can apply more general homomorphisms and that's how you get Toom's algorithm. After all, evaluation at a point t is just the quotient homomorphism for the maximal ideal (x - t).

jheriko 47 minutes ago 1 reply      
this is not exactly news... there are fft methods which are even better for very large numbers and karatsuba is adequate for the most common ranges.
Node.js podcast streaming live - right now nodeup.com
12 points by cjm  1 hour ago   discuss
FBI wants more back doors in your software cnet.com
27 points by d0ne  3 hours ago   16 comments top 9
luu 2 hours ago 1 reply      
Even if we (dubiously) assume that the FBI's proposed solution is technically impeccable, and can't be comprised, what solutions exist to prevent people from social engineering their way to the data? Heck, what prevents corrupt FBI agents and federal IT staff from just accessing the data themselves?

Of course, the same worry exists for the data at each individual company, but at least those breaches are limited to a single company's data. And, from what we've seen, externally, it seems like at least some companies are more interested in protecting privacy than covering things up. When Google found that an engineer was using his access to stalk someone, he was fired, and the indecent wasn't covered up. It's not uncommon for companies to tell users about security breaches in their own product that would otherwise have gone completely unnoticed (e.g., Pinterest announced a security flaw they had rather than just silently fixing it).

Conversely, in most cases of police and government corruption I hear about, the news breaks after a failed cover-up. No doubt I don't even hear about most cases, because they're swept under the rug. I don't have a particular fondness for Google's employees or process, but, given their track record, I trust them with my data a lot more than I trust some random government employee.

Moreover, if this law gets passed, why would serious criminals continue to use any of these services? This strikes me as having the same impact as most anti-piracy measures: highly inconvenient to non-criminals (in this case, when data gets leaked to actual criminals), but completely ineffective against real criminals. Not to mention the effect on the companies themselves -- I'm certainly not going to use a Chinese email service, because I don't want the Chinese government reading my email. What's an EU citizen going to do if this law is passed?

jasonkolb 32 minutes ago 0 replies      
The fbi sounds like the product manager from the deepest circles of hell. I'm sure this will do wonders for innovation in the economy.
3pt14159 1 hour ago 0 replies      
How the hell are they going to do that?

Even if it were possible, and legal, and secure, what about the other 95% of the world's population that can make apps outside of the US?

Zigurd 1 hour ago 0 replies      
The shortsighted aspect of this is that our government wants to order businesses to become spy-friendly to foreign governments that have a track record of stealing economic, industrial, and scientific data. Foreign governments will model their laws after ours, and specify the same interfaces.
jevinskie 1 hour ago 0 replies      
This is why strong, asymmetric crypto is necessary. I'm worried that the US government will try to put the genie back in the bottle and go back to the 90's where strong crypto was considered a munition not suitable for export [0] and when they wanted all "secure" telecommunication to include an NSA backdoor [1].

[0]: http://en.wikipedia.org/wiki/Pretty_Good_Privacy#Criminal_in...
[1]: http://en.wikipedia.org/wiki/Clipper_chip

smoyer 1 hour ago 0 replies      
I won't echo everyone else's privacy concerns, though I agree wholeheartedly. But does anyone else think it's ironic that the FBI's internal policy is named the "National Electronic Surveillance Strategy"? That's abbreviated NESS and has to be an homage to one of the FBI's more controversial lawmen.

Ness started his career trying to enforce prohibition ... 80 years later our privacy is being prohibited.

zoowar 2 hours ago 2 replies      
One person's backdoor for police is another person's backdoor for criminals.
alttag 1 hour ago 0 replies      
Part of this push by law enforcement is likely due to the increasing recognition of courts regarding the privacy expectations of email. Until recently, for example, U.S. courts have considered a service provider a "third-party", thus certain privacy protections were not available. However, the increasing ubiquity of electronic messaging has caused courts to rethink their position. It is natural law enforcement agencies would want to "push back" to effectively maintain the level of access they've enjoyed previously.
Ntrails 1 hour ago 1 reply      
Is it so unreasonable for the FBI to want to be able to 'wiretap' a facebook conversation, with a warrant, as easily as they can do so to a traditional phone line?

This is not to say I approve of the idea of an insecure back door into my online behaviours, more that I wonder whether there is not at least some validity in their desire to replicate land line style monitoring for currently untraceable online communications.

Glacier CLI github.com
17 points by AdamGibbins  2 hours ago   1 comment top
pronoiac 23 minutes ago 0 replies      
Oooh, I like the git-annex integration.
Torvalds' quote about good programmer stackexchange.com
143 points by lx  11 hours ago   63 comments top 27
robomartin 3 hours ago 0 replies      
Absolutely right. I was lucky enough to learn this in college. Although, I did not learn it from the CS professors but rather my physics prof. He was a champion for a language called APL and he actually cut a deal with the CS department to accept credits for taking an APL class he was teaching as a substitute for the FORTRAN class. APL was an amazing mind-opening experience.

Throughout the APL 101 and 102 courses he would repeat this mantra: "Work on your data representation first. Once you have fully understood how to represent your data start thinking about how to manage it with code."

He would throw this at us permanently. At the time it sounded like our Physics prof had lost his marbles (he was a very, shall we say, eccentric guy). It would take a few years after college for me to realize the value of that advise.

Put another way, our business is managing data of some sort. Whether you work on embedded systems or web applications, you are always dealing with data. You can make your programs far more complicated than necessary by neglecting to choose the right (or a good) representation of your problem space (data).

I equate it to designing an assembly line. Anyone who's watched a show like "How it's Made" cannot escape the realization that efficient manufacturing requires engineering an efficient assembly process. Sometimes far more engineering work goes into the design of the manufacturing process and equipment than the part that is actually being made. The end result is that the plant run efficiently and with fewer defects than alternative methods.

In programming, data representation can make the difference between a quality, stable, bug-free and easy to maintain application and an absolute mess that is hard to program, maintain and extend.

antirez 8 hours ago 2 replies      
This is one of the few programming quotes that is not just abstract crap, but one thing you can use to improve your programming skills 10x IMHO.

Something like 10 years ago I was lucky enough that a guy told me and explained me this stuff, that I was starting to understand myself btw, and programming suddenly changed for me. If you get the data structures right, the first effect is that the code becomes much simpler to write. It's not hard that you drop 50% of the whole code needed just because you created better data structures (here better means: more suitable to describe and operate on the problem). Or something that looked super-hard to do suddenly starts to be trivial because the representation is the right one.

michaelochurch 5 hours ago 2 replies      
The problem with code quality is that there's so much AND-ing that most people give up on understanding this massively difficult problem that is as much social and industrial as it is technical.

One of the first things you learn about systems is that parallel dependencies (OR-gates) are better than serial dependencies (AND-gates). The first has redundancy, the second has multiple single points of failure. That's also true with regard to how people manage their careers. Naive people will "yes, sir" and put their eggs into one basket. More savvy people network across the company so that if things go bad where they are, they have options.

To have code quality, you need people who are good at writing code AND reasonable system designs AND competent understanding of the relevant data structures AND a culture (of the company or project) that values and protects code quality. All of these are relatively uncommon, the result being that quality code is damn rare.

mtkd 6 hours ago 6 replies      
You can normally fix bad code - fixing bad data structures is not usually easy or even possible.

It's why I've still not fully bought in to 'release early release often'.

I prefer to defer releasing for production use until really satisfied with the structures - this way you have no barrier to ripping the foundations up.

If not 100% comfortable with the model - prototype a bare metal improved one (schemaless DBs ftw) - if it feels better start pasting what logic/tests you can salvage from the earlier version and move on.

barrkel 5 hours ago 3 replies      
This is approximately the same reason as why I start out writing most of my programs by creating a bunch of types, and why I find dynamic programming languages uncomfortable to use.

I'm less and less a fan of the ceremony of object orientation, but I think there's a lot to be said for having a succinct formalized statement of your data structures up front. Once you understand the data structures, the code is usually easy to follow. The hardest times I've had comprehending code in my career, apart from disassembly, have been from undocumented C unions.

InclinedPlane 7 hours ago 0 replies      
To paraphrase, if I may, a novice imagines that the goal of programming is to create code which solves a problem. This is true, but limited. The real goal is to create an abstract model which can be used to solve a problem and then to implement that model in code.
hermannj314 6 hours ago 1 reply      
Next week on Hacker News:
Bad Programmers worry about their code. Good programmers ship.

"Bad programmers [technique A on programming KPI metric N1]. Good programmers [technique B on programming KPI metric N1]."

Responses: Someone will ask, "What about metric N2?" And someone will say, "What about technique C?" Someone will post a personal anecdote showing that people really underestimate the value of A. Someone will respond to that by posting a hyperlink to an anecdote that shows technique B really is what matters.

6ren 56 minutes ago 0 replies      
I see a program as a theory, a theory of the problem it solves. You can see how well it generalises, if it is needlessly complicated (Occam)... and in some magical moments, you'll find it predicting phenomena you hadn't explicitly anticipated.

So I think a program's conceptualisation of a problem is the most important thing - more important than data structures or code. Though, data structures are usually closer to it, by representing the concepts you model.

However, it's really hard to get these things right. Linus created both his great successes (linux and git) after experience with similar systems (unix/minix and bitkeeper). Being able to play with an implementation, experience its strengths and weaknesses, gives you a concrete base to use, reason with, push against, and come up with new insights - it's enormously helpful in seeing the problem.

But that's a grand vision - I wonder if Linus is also talking about programming in the small, each step of the way, as a low-level pragmatic guide. I don't like git's interface or docs much, but the concepts are great, it is implemented really well, very few bugs, and even on ancient hardware boy is it fast.

east2west 34 minutes ago 0 replies      
This brings up a burning question that I have been pondering for a while. I still don't get how to properly design good APIs. I have been programming but as a scientific research not as a professional developer, and I have found I cannot remember how to use my code a day after writing that code.

Take my current project as an example. I have some samples, each of which are observations along a sequence of non-overlapping segments. My objective is to extract observations over arbitrary intervals for all samples. So I have a segment defined as a pair of start and end position plus its observation, a sample as a vector of segments, and all samples as a dictionary of samples. There are various utility functions to make segments, collect samples, and scan individual samples. The problem is I have to remember all three levels of data structures to use this code. I wonder whether it is better to define an interface for those data structures as well so I just need to remember the interface. My objections to formal definition of interfaces is that everything is so simple and obvious and formal interfaces smack of overengineering.

I got to this point because in my previous projects I put every identifiable thing as a class and found too much coupling in classes and convoluted interfaces.

Xcelerate 29 minutes ago 0 replies      
I believe Rich Hickey believes in keeping data really simple and writing the code around that. Could someone explain this philosophy to me?
jedbrown 3 hours ago 0 replies      
This is Normalization rearing its head. A properly normalized database can be extended without needing refactoring and does not have modification anomalies. There is a formal process to normalization. There is no such equivalent in code, but a poorly normalized data model virtually guarantees that any code wrapped around it will be messy. Conversely, mediocre code wrapped around a clean data model (less common in the wild) is much more amenable to incremental improvement.
KevinMS 3 hours ago 0 replies      
I think this can be distilled down to: "bad programmers worry about how, good programmers worry about why"
seanalltogether 5 hours ago 1 reply      
My only problem with this quote is it equates "new" programmers with "bad" programmers. Yes if you still have these problems after 10 years of professional work, then you're a bad programmer, but if you show these symptoms after 6 months it just means you're still learning. There's got to be a better way of stating this.
maxwell 2 hours ago 0 replies      
This seems to apply to all kinds of "writing" (symbol sequence generation), from math to poetry, though the terms differ, e.g.:

  Bad novelists worry about the plot. Good novelists worry   
about the characters and their relationships.

jiayo 48 minutes ago 0 replies      
This is the first time I've clicked on a StackExchange thread and not seen "closed - not a real question".
mathattack 5 hours ago 0 replies      
I've seen this is numerous areas.

Think about business analysis... Let's say you're analyzing the best way to support Marketing at a consumer products company like Colgate. If you start with fancy windows, or the latest technology, you won't help the business nearly as much as if you think through the data needs of the business first, and then worry about presentation later. The data model outlasts the presentation software.

Consider doing a webpage. It's much better to think about what you want to show (the HTML) first, and worry about style (CSS) and scripting (pick your tool) after.

This isn't to say coding isn't important too. You need programming skills to get what you need in the end. It's that data is the foundation. A poor data model reflects either an overly complex or unthought business model.

lttlrck 4 hours ago 1 reply      
Algorithms + Data Structures = Programs

A 1976 book written by Niklaus Wirth, designer of Pascal


JoelJacobson 2 hours ago 0 replies      
I find myself writing new functions and updating existing ones a lot more often than having to add new columns or tables to my database schema.
This is probably true for all well designed systems. If you need to change your schema every time you need to change your code, its a clear sign you are a very bad developer.

quality of code indicator = [code changes] / [database schema changes]

Related quote:

"Show me your flowcharts and conceal your tables, and I shall continue to be mystified. Show me your tables, and I won't usually need your flowcharts; they'll be obvious." (Fred Brooks)

zalew 8 hours ago 1 reply      
there's quite a similar quote on photography

"Amateurs worry about equipment, professionals worry about money, masters worry about light"

kobolt 2 hours ago 0 replies      
This is similar to the rule of representation from the Unix philosophy, covered here: http://www.faqs.org/docs/artu/ch01s06.html

"Rule of Representation: Fold knowledge into data so program logic can be stupid and robust."

simonb 4 hours ago 0 replies      
This can be viewed as a corollary to Perlis's "It is better to have 100 functions operate on one data structure than 10 functions on 10 data structures." [http://www.cs.yale.edu/quotes.html]
lifeisstillgood 4 hours ago 0 replies      

functional data structures - how to bend your brain in a functional world - nearly twenty years old. Still have to take a run up to read it

joefarish 9 hours ago 0 replies      
"good data structures make the code very easy to design and maintain, whereas the best code can't make up for poor data structures."

Quite a nice summary, courtesy of http://programmers.stackexchange.com/a/163195/31774

dschiptsov 6 hours ago 0 replies      
Algorithms, data-structures and code quality (ability to quickly understand and change) are three dimensions of the same thing, neither could be neglected.

Or to put it another way - algorithms are the plot, representation are your characters, and code is your language. One or two of three aren't enough.

Code is very good investment, it must be as short and simple as possible, but not too much. There must be a balance, compromise between volume, verbosity and meaning, as in a good poetry. The heuristic here is 'less is more'.

This must be not confused with languages. Bad, unreadable code could be written in any language, but some over-hyped languages are bloated and messy from the birth.

Roybatty 39 minutes ago 0 replies      
He's dissing OO there.
chris_wot 7 hours ago 1 reply      
Posting these stackexchange questions to HN is the best way of getting them closed as "unconstructive", whatever that means.
nirai 3 hours ago 0 replies      
I think he meant that it is better not to worry about the code since it will stink no matter what you do...
Show HN: Online C/C++ to assembly visualizer [Weekend Project] ynh.io
113 points by ynh  10 hours ago   43 comments top 20
erichocean 0 minutes ago 0 replies      
It'd be nice to make, e.g. the for loop clickable, so that it "stuck" and you could scroll down and see where it ended.

As it is now, as soon as you move your mouse, it unhighlights.

munin 3 hours ago 0 replies      
LLVM has a similar page: http://llvm.org/demo

you can compile C/C++ to LLVM IR and also 32 or 64 bit intel assembler. you can also see the effects of different optimizations, etc

berkut 9 hours ago 0 replies      
Side by side would be a better layout like:
option_greek 9 hours ago 2 replies      
Nice project. Will be more useful if there is an explanation of code generated. For example, making the various assembly instructions clickable and displaying information about them. Also a tiny bit of explanation about code and data sections might be of help too :)
jrajav 5 hours ago 0 replies      
Silently fails if you make a syntax error. I accidentally pasted this without newlines and at first I thought it wasn't responding:

  #include <stdio.h>int main() { printf("Hello World"); }

winter_blue 7 hours ago 1 reply      
This could be a great educational tool! It could be used to teach Introductory Computer Organization at the college level.

I have a suggestion:

- Would you be able to separate the web interface, so that it could used to display Java -> JVM assembly, or any other language. Basically, so that it would be re-usable.

- Putting this up on GitHub would be great.

Finally, great work for just a weekend!

pgbovine 1 hour ago 0 replies      
excellent -- my unsolicited advice is for you to ping some professors and TAs who teach low-level programming to get them interested in using this in their teaching.
unix-dude 6 hours ago 0 replies      
Definitely interesting! Lots of us unix people would just run objdump -d for a quick ASM extraction, but this is probably much friendlier to the user just starting to learn C/ASM.

Overall, Good Job!

b3b0p 3 hours ago 0 replies      
I love it!

My only suggestion right now is to have it so you can have the assembly on the left or right side. Maybe it's just me, but my focus seems to be on the left side more than the right.

Since this is a focused on assembly it seems, wouldn't you want the focus on the assembly, not the C? It could be different for different people though.

akavi 2 hours ago 0 replies      
Does anyone have a link to a good beginner's guide explaining the output?
jheriko 40 minutes ago 0 replies      
don't work really - some guidance on use might be acceptable. does it need to be a whole program? my snippets give confusing errors about variable identifiers not being types o_0
mappu 7 hours ago 1 reply      
Can i recommend adding -masm=intel to your gcc flags? I know AT&T syntax is the default for gcc and gas, but not why. Intel syntax seems clearer to read and is used by most assembly tutorials.
daurnimator 6 hours ago 1 reply      
Error: Command failed: /bin/sh: 1: c-preload/compiler-wrapper: Permission denied
ajhai 6 hours ago 0 replies      
Interesting project. As an improvement to the UI, you can consider adding syntax highlighting to the editor. Take a look at ACE editor (http://ace.ajax.org/).
Aardwolf 2 hours ago 1 reply      
In the hello world example, no matter what text you put there, the assembly is the same. Where does the string content go?
vvnraman 8 hours ago 0 replies      
This really removes any hurdle I might have of inspecting the assembly code due to sheer laziness of opening up my Visual Studio and start debugging. Awesome work. If you add some type of interactivity to the generated assembly, it can be made more visual.
gtklocker 5 hours ago 1 reply      
Error: Command failed: /tmp/test969394043.c:1:0: fatal error: can't open /root/assembler/ccMEBJ3B.s for writing: Permission denied
compilation terminated.
The bug is not reproducible, so it is likely a hardware or OS problem.


tisme 7 hours ago 0 replies      
Very neat! Suggestion: add a 'flags' field where you can specify stuff like -std=c99 and any other flags you might need for your code to compile, and a horizontal rule or something like that after every line of input code.

If you do the 'flags' thing make sure you filter it to avoid escapes to the underlying shell.

alpb 9 hours ago 2 replies      
Try this on iPad. Generated assembly looks very bad in terms of layout.
bostko 2 hours ago 1 reply      
Can't you add a function for memorizing the code written?

Thank you!

Nvidia Tegra With Open-Source Graphics Is Coming phoronix.com
9 points by mtgx  1 hour ago   discuss
Sweet.js Hygienic Macros for JavaScript github.com
51 points by philf  7 hours ago   12 comments top 3
saurik 4 hours ago 1 reply      
Despite spending a lot of time both in the design Wiki and in the talk discussing the importance of being able to determine whether a / indicates a regular expression literal or a division operator entirely within the lexer (as opposed to using the parser, which is how JavaScript is generally defined), the algorithm that this developer implemented does not actually work.

First off, an example where it works:


If you run this through sjs you get:

    a / 5 / 7;

This is because, in JavaScript, statements continue across line boundaries until they are either explicitly terminated by a semicolon or a syntax error, in which case the parse is retried at that point as if a semicolon had been provided. In this case, that means we have a single statement that is a division of these three expressions: a, 5, and 7.

However, let's take a more difficult case:

    a = function() {}

This is also a single statement: you are entirely allowed to attempt to divide a function literal by a number, you will simply get the value NaN as output. If you take this file and run it through node, adding a "console.log(a)" to the end, that is in fact what you will get: NaN. However, when first run through sjs, you instead get "[Function]".

The reason is that sjs translated the code to:

    a = function () {

This is incorrect, and demonstrates how difficult some of these underlying issues are when parsing languages that have intertwined lexer and parser state. :( Attempting some other test cases involving regular expressions (but not semicolon insertion) also failed: it seems a lot more work will need to be done on this before it will be able to process general input (and it is not 100% clear to me that the shortcut required is even possible: I haven't thought enough about it yet to say for certain, however).

(I work on the JavaScript parser for a compile-to-JS language used by people doing jailbroken-iOS development for live introspection of running processes, and thereby that was the first thing I was interested in: how well the parser worked. ;P I have intentions to add reader macros, and then replace all of the extra Objective-C syntax I added with them, but I haven't gotten around to it yet. FWIW: I actually found and fixed a bug in my parser while writing this comment. ;P)

SchizoDuckie 3 minutes ago 0 replies      
OK, either your description page doesn't get your point across at all, or you're missing a definition somewhere.

'Wish the function keyword in JavaScript wasn't so long? What if you could define functions with def instead?'

Erh, no? Why in the lord's name would I ? Is that your big selling point? 'I don't like function() because it's too long?'

'Want a better way to make "classy" objects?'

<provides example that looks like php vomited on Javascript after mating with ruby>

Why would you want to make javascript less like javascript, introduce a dependency to javascript that can read your language, and then compiles back into javascript in realtime, in a way that will obviously make debugging nearly impossible (like coffeescript)?

Am I the only one that doesnt understand the use case of this? Or is it just another lexer/parser?

saurik 4 hours ago 1 reply      
There seemed to be some confusion during the question and answer segment regarding the relative hygiene of macros in Clojure near the end of the motivation and design talk; while it was totally off-topic for the video (and it thereby made sense to take it offline), I personally wish I had been around afterwards to ask the guy who seemed so adamant that syntax-case was fundamentally better than the Clojure solution (which he claimed didn't do it correctly) why that was the case.

I'm totally willing to believe it, but based on my understanding (which sadly is somewhat limited for Scheme, but fairly in-depth for Clojure) it isn't intuitive to me: it would seem like the way you escape hygiene in Clojure (which by default achieves correct hygiene by attaching namespaces to symbols read for macros or inside of quasi-quote) is quite similar in semantics--but simpler in practice due to being exceedingly less verbose--than using syntax->datum and datum->syntax.

Facebook's Gen Y Nightmare (the future of privacy) mondaynote.com
5 points by donmcc  1 hour ago   discuss
Show HN: Hosting my WordPress blog entirely on Amazon S3 slowping.com
20 points by iqster  4 hours ago   7 comments top 3
joshbaptiste 1 hour ago 2 replies      
Holy cow, $60/month for one VPS strictly used for educational projects? You can find and build 10 VPS's (US based) on http://www.lowendbox.com for less than that, granted you wont be on Amazon's network, but for small projects and testing distributed services they are great.
yuvadam 2 hours ago 2 replies      
Nice hack.

However you really should look into static website generators such as Jekyll, Octopress or Hyde. They provide a much cleaner interface for that kind of stuff.

callmeed 2 hours ago 0 replies      
Cool but the URL structure seems crappy from an SEO standpoint. I wonder if there's any flexibility there when generating static files from WP.
Mark Shuttleworth: Amazon search results in the Dash markshuttleworth.com
73 points by kracekumar  7 hours ago   72 comments top 22
jellicle 5 hours ago 5 replies      
Look, Mark - can I call you Mark? - this idea of yours sucks and you should flush it down the toilet.

I'm typing this on stock Ubuntu 12.04 and let me describe what your search does and doesn't do.

I have a lot of music files on this computer with the phrase "indigo girls" in the title and metadata, okay? So let's see.

Searching for that phrase under Home: "Sorry, there is nothing that matches your search." (!!!) Fail.

Searching for that phrase under Applications: "Sorry, there are no applications that match your search." Okay!

Searching for that phrase under Files and Folders: "Sorry, there are no files or folders that match your search." (!!!) Fail.

Searching for that phrase under Music: Gives me 15 results that are "Available to purchase", but clicking on them results in Banshee media player coming up, with no file playing. I have no idea where these results are coming from and can't do anything with them. My local files do not come up. Fail.

Searching for that phrase under Videos: Gives me results from "Online", some of which appear to have indigo girls in the title and some of which do not. Apparently these are movies on Youtube that I can rent for $2.99. Most of them don't have anything to do with the Indigo Girls, but I guess you'd get a cut if I rented any of these random movie selections. The top result you suggest for "indigo girls" is "Ladies vs Ricky Bahl", which is some Bollywood movie that has nothing to do with the Indigo Girls. At least they actually work, unlike the Music suggestions. Fail.

So in sum, Mark - your lens search utterly, utterly fails at searching for the couple hundred files that are on my computer that match it, and it also fails at monetizing my search results with Youtube and wherever the Music search is supposed to send me to. It literally does not work, at all, in the slightest. At no point did ANY of my local files come up in that search. (Searching for "Indigo" and "indigo" had identical results - none of my local files found. In fact file search doesn't work at all for any search.)

Again, this is stock Ubuntu 12.04.

>We're interested in feedback in what sorts of things would be useful to search straight from the home lens, and how to improve the search results, as well as provide better control of the process to you.

I'm going to suggest that you search for local files. Apparently this is crazy stuff. But I think it would be an improvement over sending me to rent unrelated Bollywood videos on Youtube. What do you think, Mark?

kijin 6 hours ago 3 replies      
> The Home Lens of the Dash is a “give me X” experience. You hit the Super key, and say what you want, and we do our best to figure out what you mean, and give you that.

That argument would be a lot more convincing if the Dash actually displayed results from a lot of places, not just Amazon. What if I want to search the Web? What if I want to search my social networking services? What if I want to look up directions to a location? What if I want to look up a word in a dictionary? (Remember, Ubuntu is popular in schools in some countries.)

Shopping is just one of the many, many things that people want from their computers. Generally speaking, when I'm looking for something on the Internet, Amazon is seldom the first place where I go look for it.

If you really want to turn the Dash into the ultimate "give me X" experience, at least add Google, Twitter, and Wikipedia to the list. That would make a nice replacement for Firefox's search bar. It might even increase your affiliate revenue. You might also consider providing an API so that third-parties such as DuckDuckGo can develop and distribute their own search integration add-ons. (Extra points if you can correctly guess whether I'm looking for web search results or shopping results at any given time.)

> We are not telling Amazon what you are searching for. Your anonymity is preserved because we handle the query on your behalf. Don't trust us? Erm, we have root. You do trust us with your data already.

That statement sounds suspiciously like the other Mark that we all know and love/hate. You know, the guy who is trying his damnedest to make privacy obsolete.

cs702 4 hours ago 1 reply      
My initial reaction to non-local search results in the Dash by default was one of dismay, but after reading this post by Shuttleworth, I've decided to reserve judgment until Canonical has worked out all the kinks. The source of conflict, I believe, is that Canonical is trying to serve three distinct market segments which will react very differently to the new feature:

* Enterprise customers deploying hundreds or thousands of desktops. They will love this feature, because it will allow them to customize which external and internal online sources employees will be able to search, and then they will be able to track all employee searches.

* Regular people -- that is, the kind of people who don't even know about HN. These people will also love non-local search in the Dash. They already search for everything on Google, buy everything through Amazon, and readily hand over all their intimate, personal information to FaceBook... without ever giving their own privacy a second thought.

* Power users who're aware of the privacy issues involved. Virtually everyone in this market segment, including me, feels strongly that non-local search should be offered only as an opt-in feature, if at all.

Viewed in this light, Canonical's decision to implement non-local search can at least be understood: they're trying to make their customers happy, but they've unintentionally pissed off the smallest of the three market segments above: power users. (Sorry for the harsh language; I can think of no better way to convey how a lot of Ubuntu power users feel about this.) Alas, power users may be the smallest of the three market segments above, but they have disproportionate influence over the other two. Disregarding the concerns of power users may not be a good idea in the long run.

In retrospect, Canonical could have -- indeed, should have -- handled the announcement of this feature much better. There was really no announcement; the news was just 'dumped' on the community on a third-party blog. Is this really how Canonical wants to treat power users?

vibrunazo 6 hours ago 2 replies      
Why does he have to pretend so hard that he is not trying to make money? Why the dodging? There's nothing wrong in making money. The only thing wrong here is this dishonesty.

As many others here said in the other thread, we would gladly pay for Ubuntu, if there was an easy, straight forward, transparent way to do so. I would much prefer that instead of Amazon ads (sorry, I mean "integration").

I already pay to be a "friend of eclipse" just because it's easy to do and they deserve it. Or even better, if Ubuntu one services were worth a thing, I would love to pay for it.

cooldeal 5 hours ago 1 reply      
>It seems to me that the FLOSS community is facing M$ 2.0. But then, hey, "please don't feed the trolls."

I like how he was dismissing the concerns as FUD. I wish he just said it was about money, since it hard to employ 500 people to work on Ubuntu, instead of making it sound like Ubuntu is trying to enhance the user experience by forcing Amazon searches on people.

fierarul 5 hours ago 1 reply      
>These are results from underlying scopes, surfaced to the Home lens, because you didn't narrow the scope to a specific, well scope.

This sounds kinda buzzword-ish and ignores the fact that people, basically, don't need Dash to do any of that stuff.

>Don't trust us? Erm, we have root. You do trust us with your data already.


All I trust Ubuntu is they don't intentionally leak data. It's not like I store it on their hard drives.

Frankly I was more relaxed about this whole Amazon thing before Mark bothered to write the article. Now, I think it might not be so bad to consider another distro in the future.

quesera 5 hours ago 2 replies      
Given the wide array of Linux distro options, why would anyone choose one with this sort of junk included?

Remember all the hate for Windows with crapware preinstalled? (they stopped doing that, right?) How is this less distasteful?

Why would anyone download Ubuntu avec crap when Ubuntu sans crap will be available via BitTorrent within hrs after release? Someone will fix this bug, obviously, immediately.

Clearly, Shuttleworth is getting tired of self-financing his charitable enterprise. But what is he thinking?

Disclaimer: I don't always run FOSS Unix, but when I do, I prefer FreeBSD.

staunch 7 minutes ago 0 replies      
Wow. What utterly transparent bullshit. I would have much preferred honesty about his desire to generate revenue. I could respect that. Now I just think he's a liar or completely self-deluded.
comice 5 hours ago 1 reply      
"These are not ads because they are not paid placement, they are straightforward Amazon search results for your search."

both ads and affiliate links make money when you click on them and buy something (nobody would buy ads if they didn't result in sales in one way or another)

The precise details of how they are paid for is almost irrelevant. I appreciate affiliate links in search results like this are much better than paid placement ads, but to say they're totally different things is untrue.

If Amazon didn't offer an affiliate scheme, would Ubuntu still be so keen to integrate their search results in the same way?

f4stjack 6 hours ago 1 reply      
"We are not telling Amazon what you are searching for. Your anonymity is preserved because we handle the query on your behalf. Don't trust us? Erm, we have root. You do trust us with your data already."

Sorry but you have root to what exactly? I am not using ubuntu one and I don't trust my data with ubuntu, thank you. And I don't find his arguments convincing seriously. Some of us are old enough to remember "BonziBuddy" who "helped" with our searches and as far as I remember it also Just Worked, except it was an irritating pervasive spyware.

Privacy and Functionality are two different spheres, I can't accept something very functional if it invades my privacy.

tomp 6 hours ago 3 replies      
He's basically saying that they're integrating Amazon because (besides the obvious reason that that's how they make money) it's the most useful service for the user besides searching their local files&apps. Really? We're that deep in the consumerism mindset that any search I do you're immediately trying to sell me something?

I'm resisting as hard as I can. Besides food, energy, and transportation, I could count all the purchases this year and I would probably barely reach 20.

cooldeal 5 hours ago 1 reply      
>We are not telling Amazon what you are searching for. Your anonymity is preserved because we handle the query on your behalf. Don't trust us? Erm, we have root. You do trust us with your data already.

Wait, so my personal local file search keywords are sent to both Ubuntu and Amazon? If anything, that's only slightly better than just Amazon having them along with the IP address.

And, no, you don't have root or control of my data, unless you're telling us about some backdoors you're inserting into Ubuntu.

>Here's a quick Q&A on the main FUD-points.

FUD? Really? Is he trying to imply the outrage is manufactured by Microsoft or Oracle?

This is crossing a line that an OS should not cross. What next? Showing me local grocery results when I make a note to buy milk?

People are smart enough to pull up a browser to search for things to buy.

michaelhoffman 4 hours ago 1 reply      
The use of "please don't feed the trolls" to dismiss widespread, valid concerns about your product is pretty obnoxious. I don't think this situation is a huge deal, but I don't think it is trolling either to point it out.
motters 4 hours ago 0 replies      
The privacy and liability issues here are significant - disclosure of searches of the local OS potentially revealing the names of documents to third parties - so I think that Ubuntu/Canonical needs to take those onboard and do some re-engineering. If I was running a business, school or government department and thinking about using Ubuntu I'd be reconsidering after reading this.

We'll just have to see what turns up in 12.10, but if Canonical are insistent upon heading in this direction then sadly that's a deal-breaker for me, and I won't be able to recommend Ubuntu to others.

spinchange 4 hours ago 0 replies      
As much as I admire and appreciate Ubuntu and all the hard work, time, and money Shuttleworth & Canonical have committed, you simply don't end a "setting the record straight on user concerns" piece with a veiled threat/warning about having everyone's root.
comice 5 hours ago 0 replies      
I'd like to see the real thinking (and data if there is any) behind this decision. Did they do some tests where they asked new users to find and buy a book, and did those users try searching in unity for it?

Or did someone say "How can we get affiliate links into Ubuntu as a source of income?"

He seems to be suggesting this is user led (or at least, aimed at making a better experience for the users). Which I suspect is bullshit, but I could be wrong. I'm not an average user.

RexRollman 6 hours ago 0 replies      
No thank you sir. No thank you.
takluyver 6 hours ago 1 reply      
All the problems with this could be trivially avoided if it was a separate lens, rather than showing up in the home lens, which we expect to be searching our own computer.

I have opened a bug report for this:

jagira 2 hours ago 0 replies      
This feature is totally useless for Ubuntu users from countries like India where Amazon is not present.
tomrod 5 hours ago 0 replies      
Rather than address concerns, straight up snark on Twitter. https://mobile.twitter.com/ubuntu/tweets
pasbesoin 2 hours ago 0 replies      
Ubuntu: EOL-ed at 12.04?
k_bx 4 hours ago 0 replies      
Ok, but where will we go from here? If there will be 100 sources of online search, will it make 100 HTTP requests on every search? Or will there be yet another (I'm sure it will be closed-source) search engine that will "aggregate" other search engines?
Vim Koans sanctum.geek.nz
96 points by janogonzalez  13 hours ago   10 comments top 4
rauljara 55 minutes ago 0 replies      
I found "The slow student's despair" truly uplifting. It applies to so much. I'd love to read the original Koan (Koans?) it was modeled after.
ralph 4 hours ago 0 replies      
I realise the point is to show sometimes vim is inappropriate. However, the commands given in the first one are long-winded and not equivalent, I'm not sure the readers who stand to gain from the article will notice this. As I wrote elsewhere...

Having done :v/tcp/d there's no need to match again with :g/tcp/s... since :%s... will do, besides :g//s... would have done. Also, :g/tcp/s/\S\+\s\+\(\S\+\)\s.*/\1/ mandates a whitespace after the second word so it's not equivalent to the given :%!awk '/tcp/ {print $2}'.

nsns 5 hours ago 1 reply      
Seems a little confused between koans ("gong'an" = "public record") and the Taoist stories of Zhuangzi. But never mind, both are Chinese.
anonymouz 8 hours ago 5 replies      
Seems to be somewhat incoherent flamebait about Vi, mainly giving contrived examples where other tools do a better job (how many people think that vim is the right tool for editing CSV? Or vimscript the right language to write a Markdown processor?). If only it were at least funny...
The End of the Future (2011) nationalreview.com
5 points by simonster  2 hours ago   1 comment top
wam 33 minutes ago 0 replies      
"Today's aged hippies no longer understand that there is a difference between the election of a black president and the creation of cheap solar energy; in their minds, the movement towards greater civil rights parallels general progress everywhere. Because of these ideological conflations and commitments, the 1960s Progressive Left cannot ask whether things actually might be getting worse. I wonder whether the endless fake cultural wars around identity politics are the main reason we have been able to ignore the tech slowdown for so long."

Any sufficiently deluded rhetoric is indistinguishable from trolling.

The "progressive left" has no such blinders. Nor does the "right," in my experience, or most other people. But you could easily reverse this and say that to Peter Thiel, technological progress automatically leads to progress in quality of life for everyone. Or, more disturbingly, that social justice is an irrelevant invention of identity politics. No: those are in actual conflict. The trick of identity politics is to subsume and trivialize real social problems, and it's Thiel who has fallen for it. This is essay is just weird.

Thiel sees economic growth being appropriated to justify misguided and counterproductive social policies. That's not even wrong, it's just mundane to the point that it's disingenuous to say that anybody doesn't see that.

Is this guy going to stockpile real nuclear waste under his house until a better solution is invented? No, he actually wants to build a floating city where he can make his own rules. A ten year old's fantasy, described by a man who can afford to ignore the practical consequences of his ideal future.

This is antithetical to the hacker ethic. Good hackers look at big social and technological problems and think "how do we hack that into something better" instead of whining about how nobody else accepts their top-down solution.

Real problems to attack: The pitfalls of implementing representative democracy via legislators who develop strong social ties with each other; how to build infrastructure without becoming Robert Moses; techniques for balancing competing basic needs in a large population; p=np; obesity epidemics; vaccination against destructive ideological memes like this one.

Startup = Growth paulgraham.com
541 points by moeffju  1 day ago   197 comments top 68
grellas 22 hours ago 2 replies      
A few thoughts:

1. This is a superb essay delineating the attributes of a fast-growth, all-or-nothing type of startup. No surprise here. Who besides pg has had the depth and breadth of quality first-hand experience with such ventures over such a sustained period and in such an explosive context as that of recent years? He has here given us a classic analysis of the prototypical, Google-style startup.

2. I think the idea of a startup should not be so narrowly defined, however, and the big reason is this: many founders set out to build ventures that are tech-based, innovative, aimed at winning key niches via hoped-for rapid growth and scaling, positioned for outside funding as suited to their needs, and aimed at liquidity via capital gains as the primary ROI for their efforts . . . but who also place a huge premium on minimizing dilution and maximizing founder control. These are the independents. The ones who, by design, want to defer or even avoid VC funding so as to build their ventures on their own timing and on their own terms. Now this is not the Google startup model. It is, in a sense, its opposite. But it is not the model of a small business either. It is just a different type of startup.

3. The trend over this past decade has moved decidedly toward greater founder independence in the startup world. Back in the bubble days, as a founder, you had very little information available to learn how startups worked, you often had heavy capital needs (e.g., $2M to $4M) right up front to do such things as build your own server banks, and you would almost certainly have little leverage by which to minimize dilution or loss of control at the time of first funding. Today, this has completely flipped. Vast resources are extant teaching founders how startups work. Initial capital needs are often minimal. And it is relatively easy to get reasonable funding on founder-friendly terms. What this means is that, today more than ever, the independent-style startup is more open to founders than ever before.

4. Given the above, it seems to me that this is not the time to say that the only style of startup worthy of the name is that of the super-rapid-growth type. The rapid-growth type may be more glamorous by far but it really defines only the tip of the startup world. Beneath it is a vast world offering incredible opportunities to founders who want more control over the timing, scale, and management of their ventures and who seek to realize gains and manage risks accordingly.

ChrisNorstrom 1 day ago  replies      
- Venture Capital pouring millions into untried businesses.

- The crazy valuations.

- The recent complains of VCs that "Entrepreneurs aren't working on enough big ideas".

It all actually makes sense now. It's all in the name of Big Risk = Big Reward style ventures. Especially after defining a "startup" as a company meant to grow rapidly and to massive proportions. Not necessarily a tech business. Not an online store for your company. But instead an extreme-expantion-potential style company. I wish Startups were defined like this from the beginning.

It also further pushes me away from the whole Startup / Silicon Valley thing. Growing that fast means a lot can go wrong and there's very little time to learn from mistakes. I'm a slower thinker, I like to analyze and enjoy, learn and understand, build and live, not sacrifice my life and grow my company like a lunatic.

Thank you Paul. You've actually freed me from a dream that I now realize will never make me happy. I can finally let go of my plan to abandon my family, move to the bay area, drain my life savings, live in a shoebox, stumble from one conference and event to the next hoping to network and find my messiah & co-founder, try to get funded, grow my business to someone else's expectations, all for a tiny fraction of a chance to succeed and be either a slave to my own company or lose control of my baby and walk away with diluted equity. I think I'll stay here in St. Louis with my aging family and build my businesses slowly and calmly.

You freed me Paul. You gave me back my life, my real one.

cperciva 1 day ago 1 reply      
A good growth rate during YC is 5-7% a week. If you can hit 10% a week you're doing exceptionally well. If you can only manage 1%, it's a sign you haven't yet figured out what you're doing.

This is, to me, the most interesting thing here: I've seen lots of people talk about "traction", but this is the first time I've seen someone in the startup world give hard numbers for what a "good growth rate" is.

Another way to look at these numbers: A good growth rate during YC means that you're doubling every 10-14 weeks. An exceptional growth rate is doubling every 7 weeks, and if your doubling time is more than a year, it's a sign you haven't yet figured out what you're doing.

This fits pretty well with the rather imprecise commentary that "a startup measures the time to double in size in months, except for wildly successful ones, which measure it in weeks".

ryanwaggoner 1 day ago 3 replies      
The discussion of expected return sounds good from an investors perspective, but founders have no diversification, so a 1% chance of $100m or 99% chance of wasting five years sounds pretty lousy. This is my biggest issue with the VC world from a founder's standpoint. The situation gets even worse once you throw the decreasing marginal utility of money into the mix, because now the expected value of $100m is not worth 10x as much to me as $10m. In terms of ability to change my life, $10m provides probably 50-70% of the value that $100m provides.

So if my odds of succeeding with a $10m payout from a bootstrapped business are 10%, and my odds of succeeding with a $100m payout from a VC business are 1%, those expected returns are equal in math terms, but not utility terms, and I'd be crazy to raise money.

asknemo 1 day ago 6 replies      
Am I the only one who think pg's view points appear to be getting more and more extreme, in some sense rather biased compared to his previous essays?

Zynga is definitely all about growth. It is fiercely focused on metrics, fiercely focused on growth. But as someone from game industry, we cannot agree that this model is THE model that gives the world and everyone value. If the game industry worked like the way pg describes in the essay decades ago, we would never have Diablo, Baldur's Gate, Grim Fandango or Minecraft. We would all be left with choices like Farmville, Monsterville, Mineville, forever and ever.

"Growth drives everything in this world."? Does it? All fads grow like wildfire too, but does it drive everything in world? Or a better question would be: should we allow it to?

edanm 1 day ago 3 replies      
This article is, for me, more proof of a general phenomenon that's happening recently - startups are no longer considered the best vehicle for hackers to become wealthy.

Maybe it's just my own history and confirmation bias speaking here (recently switched from startups to a Consulting business). But lately, the whole "bootstrap" movement is getting much more popular around here. More and more, I'm seeing articles and comments from tptacek, patio11, and others talking about how programmers could make vastly more money, especially by doing freelancing. I think the message is starting to sink in - the kind of people who read this site can start very profitable businesses, make loads of cash, and do this without the high risk of startups. No chance of a working 5 years and then striking a goldmine of an exit, but much higher chance of working 5 years and putting aside large amounts of money.

This pg article is a great one, and a very honest one too. To me it reflects the changing times, and the changing understanding of what a startup means. No longer, like in previous articles on wealth, is pg very clearly advocating that all hackers should be starting startups. This essay, to me, reads as a much more precise explanation of what someone can expect if they start a startup. And it makes it much clearer when people should not start a startup.

pixelmonkey 1 day ago 4 replies      
One of my favorite pg essays of all time. Loved this:

"Almost every company needs some amount of funding to get started. But startups often raise money even when they are or could be profitable. It might seem foolish to sell stock in a profitable company for less than you think it will later be worth, but it's no more foolish than buying insurance. Fundamentally that's how the most successful startups view fundraising. They could grow the company on its own revenues, but the extra money and help supplied by VCs will let them grow even faster."

Took me awhile to realize this as a founder.

Profitability is a great goal (and makes the business very "real" by cutting away vanity metrics), but self-funding growth from profitability pretty much guarantees you are locked into a relatively slow growth rate. pg's simple charts show why being locked into a lower growth rate could mean being blown away by your competitors.

paulsutter 1 day ago 3 replies      
I gotta say, "a company designed to grow fast" is not only more concise, but broader and more on point than Steve Blanks' definition ("an organization formed to search for a repeatable and scalable business model"[1])

An epic essay with tremendous depth. Love the ending:

"A startup founder is in effect an economic research scientist. Most don't discover anything that remarkable, but some discover relativity."

[1] http://steveblank.com/2010/01/25/whats-a-startup-first-princ...

nemesisj 1 day ago 1 reply      
Recently there was an article floating around where a VC asked why there aren't more B2B startups. This mindset is why - there's simply no way you can grow at these rates in the early days with most B2B products, particularly the ones in very hard to solve areas like ERP or the like. Anything with a longer sales cycles seems to be instantly disqualified by this definition, which is why we're relegated to so many photo apps and twitter thingies.
kemiller 1 day ago 1 reply      
I think seeing it put so clearly, it's convinced me that I don't even want to found a startup. I'd like to own a business, but that's different, and I should behave accordingly. That might make it the most useful thing I've read in years.
gfodor 1 day ago 3 replies      
This essay highlighted something for me, you actually end up having a 2x2 matrix for "work for" vs "invest in" and "startup" vs "non-startup."

For example, a certain person may try increasing their wealth by investing in startups, but prefer working in a non-startup. Or another person may prefer investing in non-startups (safe, dividend paying stocks or bonds), but try increasing their wealth by working for startups.

For people with talent in creating products, best to focus their investing in safe, low maintenance non-startup investments and their wealth creation in working for startups. For people who have access to capital and a knack for choosing winners, they should work for non-startups (or philanthropy, or whatever, since they are probably already fairly wealthy) and invest in startups they think can win.

For the really talented, who both know a thing or two about building products, and also can pick winners, then you should work for a startup that invests in startups. See: pg.

d4nt 1 day ago 0 replies      
"The best thing to measure the growth rate of is revenue. The next best, for startups that aren't charging initially, is active users. That's a reasonable proxy for revenue growth because whenever the startup does start trying to make money, their revenues will probably be a constant multiple of active users"

This, for me, is the weak point in an otherwise excellent article. A lot of investments, valuations and jobs rest on this assumption. Facebook's PE ratio is currently 127, this assumption is the reason that it isn't around the same level as other entertainment companies like, say, Disney (17) or News Corp (56). And when Facebook's valuation rises like that and people invest at that level then there's a whole lot of money for paying engineers >$100k salaries and buying up pre-revenue businesses like Instagram. So, even if you're Google and you're bringing in real money, the application of this assumption to a few big cases permeates through the whole system and means that you too have to pay engineers >$100k and you too have to pay more to get hold of someone like Nik (makers of Snapseed).

PG logic appears flawless, but as a seed fund manager in the middle of this ecosystem, he's working several layers of abstraction up from some big applications of this assumption. So much so that it probably doesn't feel like an assumption to him. After all, he didn't value Facebook[1] at that level.

I genuinely hope this assumption is correct, because a lot of people and livelihoods are depending on it.

[1] I'm using Facebook here as an exemplar, I'm sure there are lots of other companies out there with valuations that are due in part to the assumption that users = revenue. My argument is that when someone sets a valuation based on this assumption it has a knock on effect to the whole ecosystem.

nhangen 1 day ago 0 replies      
I'll admit that I was a little bummed after reading this article. Every single PG essay I've read left me feeling stoked, lit on fire like I could take on the world. I felt I could identify with the man, and like I belonged here.

This essay, on the other hand, left me feeling like I don't belong here. I feel as if the YC philosophy has evolved into something different than it was, or perhaps, that this is more honesty than we've ever seen before.

Either way, I couldn't be happier to use revenue as my measuring tool, and not free users. Racing to give my product away at a rate of 5-7% weekly growth would require me to completely change my product development philosophy. I build what I build because I see something missing, not because I hope to flip it in a year.

outside1234 1 day ago 0 replies      
PG's definition of startup is self selecting. Increasingly, startups do not need VCs nor Angels as the cloud (Azure, outsourcing what used to be IT for pennies, etc.) quashes the cost curve of startups.

This is pushing angels, seed round, and VCs farther and farther up the enterprise growth curve where costs become something that the founders can't bootstrap. For virtual enterprises, this is leaving them with a smaller and smaller set of companies as software eats all of the historical infrastructure costs.

Basically, he is defining startup in a way that YC is a necessary component - but increasingly, it isn't.

sandee 1 day ago 1 reply      
"We usually advise startups to pick a growth rate they think they can hit, and then just try to hit it every week. The key word here is "just." If they decide to grow at 7% a week and they hit that number, they're successful for that week. There's nothing more they need to do. But if they don't hit it, they've failed in the only thing that mattered, and should be correspondingly alarmed."

This is the gem.

lifeisstillgood 1 day ago 0 replies      
So, is "b) reaching all the people in the Market" a function of converting a decentralised market to a centralised model?

Facebook is a successful startup because it took a decentralised model (talking to your friends) and centralised it.

Barbers are decentralised - but after I build a robo-barber for every home, then suddenly one company can cut everyones hair.

So is it possible that growing a startup fast is about increasing the slope between a decentralised (diffuse players, low margins) and a centralised model.

I suspect there are good counter examples but really startups that grow fast seem to optimise for one central point for doing what they do - dropbox, airbnb readthedocs

davidw 1 day ago 0 replies      
Excellent piece. A few comments:

> The constraints that limit ordinary companies also protect them. That's the tradeoff. If you start a barbershop, you only have to compete with other local barbers. If you start a search engine you have to compete with the whole world.

This is one of the reasons why 37 Signals' "small Italian restaurant" is not pertinent to many web businesses.

On PG's concept of startup, which I feel is spot on:

At this stage in my life, I'm more interested in... the "stay small" type of business, ala Rob Walling or patio11's bingo card thing. I think there's something to be said for a niche that's small enough that it's not interesting to larger companies, but can be served with a business that's mostly automated enough to mostly run itself. Perhaps you won't make zillions of dollars, but if it works, it's a good path to more freedom, which for some of is, is what it's about.

nwenzel 1 day ago 1 reply      
Based on required growth rates and measurement intervals (5-10% per week), there would seem to be a pretty heavy bias towards the consumer space.

B2B or so-called Enterprise Companies, especially industry-specific new companies, would have a hard time qualifying on several fronts (market size, growth rate, growth interval). I am particularly interested in the B2B style of startup because I run an Enterprise Startu-er... Enterprise New Company focused on serving the insurance industry. A B2B Startup, it would seem, would have to link customer charges to something that can grow without a new sales contract. I guess, the trick to achieving high growth rates is to create viral growth inside an existing account. Growth in the B2B space will be large jumps (with a new contract) followed by organic growth or not (within the bounds of the existing sales contract). It seems to follow then that since the purchase agreement is the painful and tough and slow part, a B2B Startup would want to have a Freemium or some other type of contract with low/no startup costs and higher per user/GB/account/server/unit costs.

Gives me some new direction on pricing.

On an unrelated note, I'll disagree with other comments of "favorite pg essays" and say that "Wealth" was the best by an order of magnitude. http://paulgraham.com/wealth.html

geuis 1 day ago 2 replies      
What about in my case, where I run a free service that currently has served over 9.3 million requests in the last month? (That service is http://jsonip.com)

Its a utility service that a lot of people are finding useful and has a lot of traction, but no way that I can see to monetize it. Not even sure I have a desire to try and monetize it, since it costs me almost nothing to run.

I would love to be able to build that resource into something more, but I'm not even sure where to start. It has a lot of usage and growth, but its not something I would remotely call a startup.

I made a short post a few hours ago about the current state: http://news.ycombinator.com/edit?id=4556711

carsongross 1 day ago 1 reply      
OK. But remember the old Paul Graham, who talked about things like this:


Seems like there's a lot of non-startup in that startup curve, if we are using the new Paul Graham's definitions.

nadam 1 day ago 0 replies      
I did not think about the definition of a startup too much but in fact I always thought about it as 'product business' as opposed to 'service business'.

Given this definition of a startup I don't want to start a startup anymore. My aim is a product business. I am into products which make me money when I sleep. Even if the income rate stops growing at $100.000 per year because of the relatively small market.

eranation 1 day ago 0 replies      
I don't think good ideas are ideas that are overlooked. I think we saw many times that being second is better. Google vs Altavista, Facebook vs Myspace, Github vs SourceForge, Google maps vs Mapquest, Foursquare vs Yelp, Stackoverflow vs ExpertsExchange, and the list goes on.

I think the ideas are overrated, what matters is the small twist in the idea that makes you better than the competition, and most importantly, execution. Once you are out with your product, getting the right talent, getting the right design, getting features out there faster than your competition (like your own company in the past based on one of your posts) is what matters.

I think there are not that many ideas that are overlooked, just people that don't think they can do something with them.
the chance that an idea was not thought by someone else, that reads the same blogs, have a similar lifestyle, and is a smart person like you, is very low (or the problem you are trying to solve is not a real problem), but the chances of that person to have the courage, time, effort and perhaps money to start a business, and find good co-founders, (and willing to risk his marriage and apply to YC) is what's a bit more rare in my opinion.

On the other hand, I have zero experience relative to you, and it's a little weird for me to disagree with one of the biggest startup mentors of our time, but still, this is just my opinion.

wamatt 1 day ago 0 replies      
Not a fan of praise for the sake of it either, but having said that, this is one of the most succinct and focused essays I've read on startups, in a long time, and hard to fault it's fundamental message.

As founders it's easy to do things other than push every day to get customers and/or active users. Some founders are so focused on other less stressful activities, that they outsource the entire function to a 'growth hacker'. Let someone else deal with it... Yikes!

Grow is core to the startup's success, and happy to be reminded of it.

tubelite 1 day ago 0 replies      
Consider an alternate universe with far worse odds: 1 startup in a field of 100,000 makes $100 billion, the rest make zero. Expected value is $1 million.

Should you be one of the 100,000 founders who "rationally" choose to buy a startup lottery ticket for $100b?

Well of course! If you could live for long enough to run through thousands of iterations of the startup game, that is.

In the real world,
- Founders are typically limited to 1 startup at a time
- It takes time - years perhaps - for a startup to fail
- Founders can do only a few startups in their lifetimes

Given low odds like 1%, with < 10 iterations per lifetime, expected value is the wrong metric. From a purely financial POV then, it appears only rational to do growth startups
- If you are on the investing side (a "parallel" entrepreneur, running tens or hundreds of iterations in your lifetime)
- You are a founder (a "serial" entrepreneur) with reasonable financial security and none of your life goals would be irrevocably damaged by the most likely outcome - a string of failures.

(There are of course several non-financial reasons and payoffs. For instance, an Idea takes demonic possession of you, and the only way to exorcise from your tortured brain is to do a startup..)

photon137 1 day ago 0 replies      
"For a company to grow really big, it must (a) make something lots of people want, and (b) reach and serve all those people"

Very valuable insight.

However, (b) in its own right, can serve as a fast growth business model - where the delivery or "clearing" of value between those who demand and those who supply is the value proposition itself - because everyone wants delivery (making it, by default, a big market).

Most banks work on this principle, in an abstract sense. A business like FedEx or UPS is a more physical example of this.

Online takeout-ordering services are examples of this - the customer wants the food and the "online ordering website" startup does not produce food - but what it produces is "clearing" ie matching demand to supply.

This - as an idea, in my experience, always scales and grows fast as well while falling into the category (b) that pg mentions.

hasenj 1 day ago 1 reply      
So basically a startup is a company whose goal is not to create a profitable business, but a company whose goal is to grow a large userbase rapidly and lure VCs to pump more money into it, because VCs just dream about finding the next google or apple and funding it in an early stage.

This just emphasizes why I do not want to be a part of this scene.

As DHH says: fuck doing a startup.


ef4 1 day ago 2 replies      
pg makes his point clearly at the cost of oversimplifying his definition. Scalability is a continuum. There is a continuum between barbershop and search engine.

VCs have every incentive to hit the far high end of the continuum. But a young, hungry entrepreneur probably gets higher expected value by not straying quite so far out.

josephlord 23 hours ago 0 replies      
It may be sacrilege round here to disagree with pg but I dislike the appropriation of the word startup for this niche of companies. I think of all new businesses as startups for the first couple of years. I think a better name for speculative high growth companies would be 'venture companies' which could have venture founders to go with the VCs that may fund them.

That way the term would make sense for older companies if they hit a high growth phase.

ChuckMcM 23 hours ago 0 replies      
One of the things we don't see often, and I wonder about are situations where the Angel funds 40% of the company at a seed round at one valuation, and then contributes half their interest in exchange for cash in the series A.

From a practical standpoint this means that the company needs to raise n + A where the "+ A" part pays back a return to the angel. But it allows the Angels to shoulder risk early to weed out the non-viable players and still make money at it.

So imagine this scenario. Alice wants to start Woohoo and Bob funds 40% at a $250K valuation (since Alice is the only person so far, that is $100K invested. Alice works hard and gets out an MVP and goes for a Series A where she wants to raise $1.5M at a post raise valuation of $5M. Bob sells into the round half his shares (its taking money off the table for him) with a contractual net return of xx% (probably anywhere from 20 - 100, that being the negotiating rub) and does not participate in future rounds, he retains y% at the end of the series A (optionally converted to Common stock) and dilutes going forward. (remember he's got 'free' stock at this point, he's made his bit with the angel round)

The two negotiation points on Bobs term sheet are the net return and the retained interest portion. The people coming in after Bob are Ok with it because Bob is out now and paid his 'finders fee' or however ever you want to describe his return. The net return will affect the series A amount needed by the founder, the retained interest would be a function of how many rounds the founders think they will need to exit.

Given the seed to series A or bust cycle is usually at most 24months, Bob has a good idea of his risk profile, and by stepping out from an equity point at the Series A he's not an obstruction to new partners coming in, more of an adviser at that point.

plinkplonk 1 day ago 0 replies      
pg, just out of curiosity what was the (approximate) weekly growth rate of Viaweb ? Did you focus on this metric when you were building Viaweb?

Just curious if you were aware of this factor when building Viaweb.

philwelch 1 day ago 0 replies      
> But the two connections are distinct and in principle one could start a startup that was neither driven by technological change, nor whose product consisted of technology except in the broader sense.

Arguably Netflix was one of these. DVD's may have been the "technological change" driving the company, but in principle you possibly could have done the same thing with VHS.

Most logistics-based big companies, like Fedex, Walmart, or even Costco, could be cited as examples. Jetliners predated Fedex by decades, for instance. Aside from the novelty of selling PC's, Dell was a similar instance.

This indicates that, just as in war, solving logistical problems often provides the biggest wins in business as well.

Combining innovation and logistics was powerful enough to make humble Apple erupt into the world's most valuable business. In fact, Apple's growth rate during Jobs' tenure as CEO was ultimately exponential despite the company's age--does Jobs-era Apple somehow qualify as a startup due to this?

> I once explained this to some founders who had recently arrived from Russia. They found it novel that if you threatened a company they'd pay a premium for you. "In Russia they just kill you," they said, and they were only partly joking.

I remember that in the 90's, many of Microsoft's acquisitions had this same sinister undertone to it. I'm reminded of a Simpsons episode where Microsoft "buys out" Homer's startup, but all that happens is a bunch of goons smash up the house while Bill Gates quips, "I don't get rich by writing a lot of checks."

scribu 1 day ago 0 replies      
This is a foundational essay. It's hard to believe the whole Y Combinator ecosystem has gone on for so long without it.
staunch 1 day ago 1 reply      
> And the probability of a group of sufficiently smart and determined founders succeeding on that scale might be significantly over 1%. For the right peopleâ€"e.g. the young Bill Gatesâ€"the probability might be 20% or even 50%. So it's not surprising that so many want to take a shot at it.

Which means that even Bill Gates may have failed repeatedly at creating a successful startup.

Just as in poker even great players will lose if it's not in the cards and even the mediocre player can win big on occasion.

The lesson is that even a founder as good as Bill Gates can only expect to succeed if she is willing to make multiple attempts.

jusben1369 17 hours ago 0 replies      
What defines a startup for me is utilizing technology with a very good chance it will be a spectacular failure. You have a great feeling that you'll have users but there's a very real chance that you'll get none to 10 and go down in tremendous flames. You're swinging for the fences. That's because you're doing something that no one has done or your entering a market with huge dominant players whom you hope to usurp. It's either very audacious or insane. Your 10 professional friends split down the middle when asked.

You know if you open a barber shop you will get people come in off the street to cut their hair. You just worry you'll have enough over time to sustain it. If you start a consulting business you know you'll get clients (at least I assume you do) as your skills are in demand. You probably got commitments from at least 3 before you made the plunge. It's just whether you'll get enough and a steady flow to go with. If you're bootstrapping a lifestyle tech company it's probably because you know exactly what the market needs and have your first 5 customers from your previous gig so you know you can get through the first year. All of these still have a large amount of risk but they're not really true startups to me.

It's true it's very hard to have a startup without (rapid) growth as the central tenant . And you can have many businesses that are not startups that aren't focused on growth over anything else. However, you can have many businesses that are not startups hyper focused on growth (SuperCuts, Pappa Johns) to make the definition in this essay too weak for me.

seunosewa 1 day ago 0 replies      
A 'startup' is simply a new business. That's what the word means. You can't just take a word that has an existing meaning and say it means something else.
shashashasha 1 day ago 0 replies      
It's a little Pop Science, but as Geoffrey West notes, other things that follow an S-curve growth are people. We start small and weak, at some point start rapidly growing, and as we reach a certain age we level off. West then extends this to the inevitable deaths of corporations:


Choppen5 1 day ago 0 replies      
Seminal article - fantastic. I immediately calculated growth rate for Mightbuy.it - posted them here: http://blog.mightbuy.it/2012/09/22/user-growth/

The total #s are pretty embarrassing but the growth rate actually looks good, at around 10% currently - except the rate of growth of the growth is decreasing.. how meaningful is growth rate for such short time periods?

In my case I'm closer to the first stage of just making a product people want, but seems useful to keep track of this rate already.

PS - please sign up and help a brother out.

smalter 1 day ago 1 reply      
Question: w.r.t. weekly revenue growth, are we talking about growth rate of monthly revenue run rate (assuming billing happens monthly) or something else? For instance, something like this: new signup revenue in the current week x historical conversion to paid divided by revenue of paying customers + sum of estimated revenue of recent previous weeks who haven't yet hit the date of conversion? (if that makes sense)

I'd love to hear thoughts on how to calculate the referenced number.

tristan_juricek 1 day ago 0 replies      
This article might just be my point of reference whenever I get a bit too much "valley hate". Like, when I start reading a bunch of the ideas coming out of the blog from 37 Signals.

I still think that for every serious genius, there are a lot of copycats. These copycats create this gigantic echo chamber of ideas. It's a ton of noise from a lot of likely failures. That noise can feel dishonest.

After reading this essay, I'm left with a sense that startup land is probably the best, dare I say more honest way, of turning investment money into new technology and jobs.

I'm actually left wondering about the relationship between "normal small businesses" and bankers. It feels adversarial in comparison.

mmmmax 1 day ago 0 replies      
If this essay had an abstract, this would be it:

"If you want to understand startups, understand growth. Growth drives everything in this world. Growth is why startups usually work on technologyâ€"because ideas for fast growing companies are so rare that the best way to find new ones is to discover those recently made viable by change, and technology is the best source of rapid change. Growth is why it's a rational choice economically for so many founders to try starting a startup: growth makes the successful companies so valuable that the expected value is high even though the risk is too. Growth is why VCs want to invest in startups: not just because the returns are high but also because generating returns from capital gains is easier to manage than generating returns from dividends. Growth explains why the most successful startups take VC money even if they don't need to: it lets them choose their growth rate. And growth explains why successful startups almost invariably get acquisition offers. To acquirers a fast-growing company is not merely valuable but dangerous too."

tstyle 1 day ago 1 reply      
"It's the same with other high-beta vocations, like being an actor or a novelist. I've long since gotten used to it. But it seems to bother a lot of people, particularly those who've started ordinary businesses. Many are annoyed that these so-called startups get all the attention, when hardly any of them will amount to anything."

I see so many comments on HN poking fun at VC backed startups with underdeveloped business models. Sure, a lot of the criticisms are well deserved, but I wonder if a part of it is because people are bothered by the power law phenomenon.

gojomo 1 day ago 0 replies      
Some community-oriented/UGC startups nowadays seem to prefer modest growth, at least for a while, to make sure the right standards/user-expectations can be maintained. (Hypergrowth can mean a change in community makeup faster than desireable norms can be maintained, which can spoil the dynamics in a way that's hard to un-spoil.)

Compare HN's own revealed preference for limiting certain surges of new users. (Corollary: HN is not a 'startup'.) In a way, even Facebook's initial campus limitations served this purpose, getting certain mechanics (and corporate practices right) before facing the challenges of a larger userbase.

I wonder: have YC companies had to face an explicit decision: grow faster or defend/consolidate the culture of the existing userbase, and if so what advice would PG and the other partners be likely to give?

vegashacker 1 day ago 1 reply      
Question about measuring weekly growth rate: A lot of YC startups are centered around an iPhone app. If it rides up the charts, it will likely get a huge bump in whatever metric you are trying to measure. But the bump's very often temporary, since more than likely the app will slide back down the charts in a week. How do you measure growth when this happens?
jusben1369 21 hours ago 0 replies      
The problem is a barber shop can be a startup. It can have fierce local competition and global aspirations (Supercuts anyone?) It feels like a brave attempt to solve the question that regularly to torments these boards but I don't feel any closer.
batgaijin 1 day ago 1 reply      
I think it's missing the idea of bootstrapping. I think it makes a much rougher environment for fledgling startups, but I think it should be considered more like a hot forge. The more the odds are stacked against you, the better you get. I guess people sometimes miss that when they are aiming for Twitter/Facebook level revenue accountability.
nreece 1 day ago 1 reply      
If you write software to teach Tibetan to Hungarian speakers, you'll be able to reach most of the people who want it, but there won't be many of them.

It may not scale, but the chances of it making (any/more) money are much higher ("riches are in the niches") than a "fast startup" as it's very rare for a startup to grow fast and monetize quickly at the same time, because most people just won't buy immediately, sometimes even if it solves a problem for them.

For example, I've been using Evernote and Dropbox for years, but haven't had the need to buy a premium account. I believe there are many others like me who are happy with the free (or open source) software that does solve an itch. Are these "fast startups"? Is their business model (freemium) scalable?

sreitshamer 1 day ago 0 replies      
I thought 'startup' was defined as a business-like organization in search of a business model. Once it finds/chooses one, it becomes a business.

The essay seems to define 'venture-backed startup'.

bdr 1 day ago 1 reply      
I don't understand this part: "For founders who are younger or more ambitious the utility function is flatter." Does flat mean O(1) or O($)? I'd guess the former, but the latter seems to fit more with the conclusion.
rdudekul 1 day ago 0 replies      
Paul Graham is a hero all technology startup founders need to look up to. Who else can say "the constraint of growing at a certain rate can help define a startup", so well?

It is rare to find a blog post that has so much insight that you will need to read it slowly and then read it again even more slowly.

rsheridan6 1 day ago 1 reply      
I don't understand why he chose restaurants and barbershops as examples of non-startups. Both make things lots of people want, and some of them are big chains that have wide reach, so they fulfill criteria a and b. Either these are startups, or there are more criteria pg didn't include, such as rate of growth - even the most successful chains usually have 10 years or so from the opening of the first store to becoming huge (inter)national chains.
dave1619 1 day ago 2 replies      
For a startup measuring users (not revenue), what's the right thing to measure to know your growth rate?

Is is DAUs, MAUs, daily sessions, length of session, total signups?

nreece 1 day ago 0 replies      
Lot of insight, but I'd rather breath outside the bubble and build a slow company, and I'll still call it a startup - http://www.fastcompany.com/3000852/37signals-earns-millions-...
Tycho 1 day ago 0 replies      
I'd be interested to know the volatility of these growth rates. PG talked as if the might be fairly constant over the course of one or two years, but my gut feeling is that they'd be anything but.
eyoel 1 day ago 0 replies      
I've read most of the points made here in an essay Joel Spolsky wrote a while back: http://www.joelonsoftware.com/articles/fog0000000056.html

It was very insightful at the time I read it.

smoyer 1 day ago 0 replies      
Isn't a start-up that's bootstrapped and designed with a clear path to monetization desirable anymore? You can obviously design something to grow quickly, but doesn't that also increase the risk of it being a flash-in-the-pan?
monty_singh 1 day ago 0 replies      
Fred Wilson's followup on PG's growth essay: http://www.avc.com/a_vc/2012/09/growth.html
001sky 1 day ago 0 replies      
But at the moment when successful startups get started, much of the innovation is unconscious.

-- This is an interesting bit. The notion of what is intelligence.

At any given time, there are 1001 questions that may be [intelligent]. But ask right question, and the answer is often [easy] to see. So Framing. Observation. Caring. Passion. Perserverence. Non-ovious. Yet invaluable. Forms of intelligence.

c3d 1 day ago 0 replies      
Illustrated over nearly 75 years of growth for Hewlett-Packard here: http://www.youtube.com/watch?v=FCOt03Y_0SM.
tlogan 1 day ago 0 replies      
The part "and how to reach those people" should be written in bold.
It seems like the difference between lifestyle business and startup is sometimes (or in majority of cases?) just in figuring out scalable way to acquire new customers.
allenbrunson 1 day ago 0 replies      
minor corrections:

"What matters is not the abolute number of new customers ..." should be "absolute"

in footnote 13:

"Though nominally acquisitions and sometimes on a scale …" should be "acquisitions ARE sometimes" I guess?

willrobinson 1 day ago 0 replies      
spellcheck: abolute

My favorites were the last paragraph and the last footnote.

How did those Russians get on anyway?

The only other thing I would add is that these high growth companies, so-called startups, are all utilizing the web. They are relying on what it provides. I guess that's implicit, maybe it need not be stated, but historically could older forms of media have supported the type of growth rates discussed? How popular was the term "startup" before the web? And did it mean the same thing?

The startup: 1. Trying to solve a harder problem than existing businesses are willing to take on. 2. Being equipped for rapid growth. 3. Utilizing the web.

ekm2 1 day ago 1 reply      
pg listed writing novels and tech startups as "high-beta" occupations.What are other high beta and high alpha occupations?
kaiwen1 1 day ago 0 replies      
This is probably the most insightful article on startups ever written. I was blind, and now I see. Thank you, Paul.
quasistar 1 day ago 0 replies      
Indeed, this cogent essay has been a long time coming and should be a pre-requisite for anyone thinking of getting in the game. "A barbershop isn't designed to grow fast. Whereas a search engine, for example, is." Brilliant.
happypeter 1 day ago 0 replies      
I learned a lot, thx pg!
Munksgaard 23 hours ago 0 replies      
mcartyem 1 day ago 0 replies      
"the same thing that makes" -> "the same thing makes"
timpeterson 1 day ago 0 replies      
more blah, blah on what made google, fb, etc. great,

what's the point?

Why Kids Should Grade Teachers theatlantic.com
88 points by jseliger  17 hours ago   57 comments top 10
tokenadult 15 hours ago 1 reply      
This is one way this process has been validated, from the submitted article: "The responses did indeed help predict which classes would have the most test-score improvement at the end of the year. In math, for example, the teachers rated most highly by students delivered the equivalent of about six more months of learning than teachers with the lowest ratings. (By comparison, teachers who get a master's degreeâ€"one of the few ways to earn a pay raise in most schools â€"delivered about one more month of learning per year than teachers without one.)

. . . .

"The survey did not ask Do you like your teacher? Is your teacher nice? This wasn't a popularity contest. The survey mostly asked questions about what students saw, day in and day out.

"Of the 36 items included in the Gates Foundation study, the five that most correlated with student learning were very straightforward:

1. Students in this class treat the teacher with respect.

2. My classmates behave the way my teacher wants them to.

3. Our class stays busy and doesn't waste time.

4. In this class, we learn a lot almost every day.

5. In this class, we learn to correct our mistakes."

Here is earlier reporting (10 December 2010) from the New York Times about the same issue:

Here is the website of Ronald Ferguson's research project at Harvard:


And here are some links about the project from the National Center for Teacher Effectiveness:


LAST EDIT: I'm amazed at how many of the comments in this thread appear to be about issues thoroughly discussed in the submitted article, but unresponsive to what the submitted article said. On this kind of issue, it's an especially good practice to read the fine article before assuming what is being discussed. We all know about school, but specific proposals for school reform have specific details that make some worse than others, and can be empirically tested.

hooande 15 hours ago 7 replies      
Basing teachers' pay and job security on surveys from students seems like a good idea, especially given the numbers mentioned in the article. One problem is that it might give too much power to students.

I was a dick back in high school. The hacker I was back then would have figured out exactly how the testing and metrics were set up (public information) and organized a union of students to manipulate it. I can't do much with standarized test scores, they reflect on me. But a teacher quality survey? That's just a weapon.

Things like this make me wish that we had some kind of Hacker in Chief, to figure out how to circumvent new systems before they get implemented.

dubiousjim 1 hour ago 2 replies      
I know more about student evaluations at the university level than at the grade levels discussed in this article. Here is an excellent overview of some of that research: http://home.sprynet.com/~owl1/sef.htm

The message one takes away from that is that (i) yes, student evaluations are a good predictor of some objective properties of a class (and other measures don't even achieve that much), but those properties aren't what teachers should be optimizing. I'd grant that (ii) it does seem worthwhile for students to see that their interests make a difference to what happens in the classroom. I'd also grant that (iii) some classroom situations may be so bad that optimizing student satisfaction may, even if not educationally ideal, still be a big improvement. And for all I know, this may be widely true at the pre-university level; but on the other hand, for all I know, giving these evaluations a big institutional role at the pre-university level could also be counter-productive...the evidence cited in the article hardly enables us to say. Any deliberation about giving student evaluations an institutionalized role should take the evidence behind (i) seriously.

One promising message from the research reported in the Atlantic article is that (iv) the specific tests being discussed have been designed in ways that seem novel and especially revealing. But the article mixed that together with an indiscriminate enthusiasm for student evaluations quite generally. And I think many people will read this and say, "Duh, that's a no brainer." Yes, it is a brainer! These kinds of policy issues aren't settleable from the armchair. Even if we cleared all the political hurdles and made someone the educational policy dictator, he or she isn't going to be able to tell just from the armchair what the results of rolling out one policy rather than another is going to be. So I get frustrated with articles like this one, that report some interesting evidence but mix it together with the kind of insensitivity to the details exhibited in comments like "That research had shown something remarkable: if you asked kids the right questions, they could identify, with uncanny accuracy, their mostâ€"and leastâ€"effective teachers. The point was so obvious, it was almost embarrassing."

Neither does this inspire confidence: "Some studies...have shown that professors inflate grades to get good reviews. So far, grades don't seem to significantly influence responses to Ferguson's survey: students who receive A's rate teachers only about 10 percent higher than D students do, on average." I hope that readers of this site don't need an explanation of why the clause after the colon is only barely relevant to whether grades get inflated because that leads to better evaluations. It's almost irrelevant. In the first place, teachers needn't be aware of the cited fact; they may experience grade-inflation pressures differently. Also, the cited fact is compatible with the majority of current A-getters scoring their teachers in ways that are largely insensitive to the grades they get, but a minority of current A-getters and a majority of current B-getters being extremely responsive to the grades they get. The cited fact is just not what we need to know.

rumcajz 12 hours ago 0 replies      
As with any self-sustaining system (politics, economy, ecology) you need a feedback loop to form among the parties with conflicting interests. If the feedback loop is broken, the system deteriorates and ultimately falls apart. If it's inefficient, the system tends to be inefficient as well. (Examples: soviet-style planned economy, rabbits in Australia etc.)

This kind of thing (teachers grade students, students grade teachers) could improve the efficiency of the feedback loop and thus efficiency of the education system as a whole.

PaperclipTaken 15 hours ago 1 reply      
I am worried about adding another metric to the way that we measure performance in schools. It's already been shown that there are problems with the standardized tests. Not that they are terrible things, but students in the US tend to disagree with their proliferation. (Perhaps they would feel less this way if there were less tests)

Having students grade the teachers I think is also the same way. One of the problems is that students don't know what makes a good teacher. The tests can gear students in that direction (ie "I feel challenged but not overwhelmed in this classroom"), but if we look for too much insight from the students I think we will be misdirected. Just like we are misdirected when we pay too much attention to standardized tests.

My fear is that some schools would start to look at these performance measurements as golden bullets sort the way that we've started to look at standardized tests as golden bullets. Most of jr. high was geared towards getting perfect scores on the state exams. My Junior and Senior years of high school were almost 100% (a few teachers went outside of the scope, but it was a teacher decision and not an administrative decision) geared around AP tests and the ACT.

The ACT and the AP tests have their place. And I think that student evaluations of teachers have their place as well. Both are very useful when applied appropriately.

I just don't want to see the system (d)evolve in such a way that too much emphasis is placed on empirical data.

philwelch 15 hours ago 4 replies      
This is probably the root cause of grade inflation in American universities. Terrible, terrible idea.
alttag 14 hours ago 1 reply      
If the student evaluations correlate strongly with the standardized tests' measures of student progress, then:

a) What's the purpose of having both?

b) If the evals are preferred over the tests, will "good" teachers continue to teach to a predictable, standardized curriculum?

c) Is the correlation additional evidence in favor of "differential compensation", that is, a compensation program based at least in part on exam scores?

d) Even if the information supplied is similar, doesn't this extra test/survey administration detract from instructional time? Is the information gleaned sufficient to compensate for the loss of instructional time?

e) Atlanta (Georgia, USA) is still reeling from a years-long cheating scandal. If such evaluations become "high stakes" (and there will likely be a push to do so, despite likely union opposition), won't these results be exploitable as well? (And perhaps even more so, through campaigning, social engineering, etc?)

jakejake 15 hours ago 4 replies      
Some kids would be excellent at grading teachers. Unfortunately some would just have a grudge, or just have a power trip and try to sabotage teachers. I think on average they would be a good reflection of how well-liked the teacher is, but maybe not how effective they were at teaching.

For example the tough math teacher would get a bad evaluation, but the cream-puff teacher who never assigned homework and gave everyone A's would get a good evaluation.

LH_Chapman 1 hour ago 0 replies      
This survey was part of the larger study funded by Bill Gates. The claim of "prediction" is not supported by the evidence.. the correlations with test scores were low.. and in any case, this is another example of reifying test scores and stupid concepts such as "a months worth of learning" as if all learning is the same, regardless of the subject, the grade level, the prior experience of the kids and so on. The very low correlations with test scores are not surprising and a by-product of the survey questions, designed to check whether the "good" teacher gets kids to comply with rules, defer to the teacher's authority, think of learning as not making mistakes (and for the sake of Gates, stay on task all the time). Perfect conditioning for students being taught that education is a matter of doing well on fill-in-the bubble tests that Gates and this researcher seem to value as the single best measure of great education. See this and the links within it. http://voices.washingtonpost.com/answer-sheet/guest-bloggers...
camus 3 hours ago 0 replies      
next step , let kids grade there parents , and the administration calls social services when scores are low.
Coding Horror: The Last PC Laptop codinghorror.com
116 points by janerik  10 hours ago   134 comments top 24
linker3000 8 hours ago  replies      
"Want the smallest most portable device you can get away with? "

"Want to be always connected to the Internet? Sure you do;"

"Just try opening a laptop on a crowded subway train or bus."

Someone needs to stop trying to arrange their life around their tech and try it the other way around.

If you are so overloaded with work, or so keen at your work, or so behind in your work, that you need to open your laptop on the train and start tapping away then some part of your work-life balance is in serious need of adjustment.

If you're THAT valuable to yourself or the business that you MUST be hunched in front of the keyboard, or online, every waking moment then why are you on a train? Your value/benefit demands that you be driven everywhere so you can work in the back of the vehicle and that you have a PA to handle messaging. Right?

I am an IT Manager for a high-tech company. Here's my solution, based on my personal values and my value/responsibility/benefit to the business:

Samsung Galaxy Note

It's a stupidly-big phone AND a moderately-sized tablet. I can keep in touch with emails, speak to people(!), do video calls and Skype/VoIP, check messages and do remote support via RDP/VNC/SSH/OpenVPN if there's a serious crisis when I'm on the move. If needed (rare), I can use the phone for reading/editing documents and PDFs, minor coding work and, if I really want to fill another pocket, it will work with a bluetooth (or miniature USB wired) keyboard.

Acer Aspire 5735 laptop.

It's a dual-core, 15.6" screen model, about three years old. The laptop comes with me very occasionally when I need it for a meeting or if I need to hook up to some kit when I'm on site. As it happens, the Galaxy Note has an MHL (HDMI) video connector and so I can use it for meeting presentations anyway. Laptop for use on a train? Maybe - on the extremely rare occasion that I need to type up a report on the way back home as it's needed first thing the next day. Typing stuff on the way TO work or site? That was done in the office the day before, or maybe at home if things get sprung on me at very short notice. If I don't grab the laptop, an original Acer Aspire One AA150 netbook comes with me.

In a nutshell, I have boiled down my tech-demanding activities to:

Desk-based: Use a conventional PC, or my laptop, or my phone (for calls and simple networking diagnostics)

On the move: Phone for about 95% of the time. Take laptop or netbook if needed.

Considering that the phone is always with me, that means I'm automatically setup for almost all my work and personal tech needs all the time.

Would I consider buying a UX31A or similar ultrabook (or a tablet for that matter) - sure, if cost/benefit was not an issue, but the device would probably spend most of its life on a desk, and all the things that make it what it is (size, lightness etc.) would be wasted - unless of course, I turned things on their head and altered my work/lifestyle to fit around the technology - mind you, that would mean taking the train to work instead of driving - and, no, I don't have a chauffeur!

edanm 6 hours ago 6 replies      
I wanted to buy a new windows laptop that was light, strong, and looked good. I really, really did. Then, because of work reasons, I ended up getting a MacBook Air instead.

Now, I've been dead-set against switching to a Mac for a long time, because I am incredibly invested with Windows, more so than almost anyone I know. I didn't want to have to relearn everything, plus rewrite all the scripts I have on my Windows machine.

Still, I'm pretty happy with the switch so far. The laptop is everything you could want - small, fast, amazingly easy to pull out and turn on for quick use. I haven't actually tortured the laptop with too much programming yet, but OSX has a lot of small advantages over Windows. I'm not finding it too hard to switch, much easier than I thought I would find it.

I can't help feeling that, unless the hardware makers get their act together, I'll never be going back to a Windows-based laptop. The MacBook is just too good.

jawngee 8 hours ago 2 replies      
Interestingly, I'm having the opposite thing happen. Due to this MacBook Pro Retina (I have the fully spec'd version), I firmly believe that my Mac Pro will be the last desktop computer I own.

While the Mac Pro (about a year old) still smokes this laptop in a lot of areas, it's not enough smoke that I find myself turning it on half as much as I used to after I got this laptop. Occasionally, if I'm doing some heavy duty OpenFrameworks hacking, I'll switch over.

I've moved almost all my video editing to the MBP, mostly because I went with the G-RAID thunderbolt RAID (wrongly assuming they'd introduce a Mac Pro line with Thunderbolt sometime soon). And render times aren't so bad that I miss it all that much, though when I do I will begrudgingly shuffle some files around and fire up the Mac Pro. (I do video installation art as a hobby).

I think I will miss how warm the Mac Pro keeps my office in the winter though.

I also can't wait to see what next year's line refresh brings.

Derbasti 9 hours ago 1 reply      
This very much reminds me of the Steve Jobs quote that in the future, laptops and desktops will be like trucks. They will always be around for those who need them, but most people don't need trucks.

Personally, I don't quite see that yet with today's hardware and paradigms. The processing power is certainly getting there. My tablet today is significantly more powerful than my laptop in 2006 but still not quite fast enough for compiling stuff etc. Also, tablet user interfaces are still a bit too unwieldy and rigid for programming work and there is something to be said for huge monitors.

But then, Windows 8 looks like it could be the solution: A tablet that can run Windows on the go, but transforms into a regular desktop when connected to a keyboard/mouse/screen. The future.

zalew 9 hours ago 3 replies      
> And even programmers, the audience who would most need all the power and flexibility of laptops, are switching to tablets.

programmerS? the link is the only guy I heard of so far, and I even know die-hard ipad fans who laugh at the idea of a tablet being a serious workstation.

abecedarius 9 hours ago 4 replies      
Along with this, I'm considering the Thinkpad X1 Carbon or the Samsung Series 9 -- they're all advertised as lightweight, powerful, and with more vertical resolution than the usual piddling 768px. Anyone tried them and want to weigh in?

(edit: previous discussion at https://news.ycombinator.com/item?id=4428262 )

I'm leaning towards the Thinkpad since I get the impression Thinkpads usually go well with Linux. (I might need to run Windows too, alas.) This will replace a Mac laptop.

mark_l_watson 4 hours ago 0 replies      
It is possible that a better future tablet would have good enough programming tools - it will be fun to wait and see.

I got rid of all three of my desktop systems (1 Mac, 2 dual Linux/Windows boxes) a few years ago. I don't plan on buying any more Apple gear, but will use what I have until it needs to be replaced: MacBook Pro running Ubuntu, Air 13', a Toshiba laptop Linux/Windows, and a iPad 2. My wife and I live in a very small house in the mountains and not having as much computer clutter is great.

The biggest recent technological change in my work/life mix has been the purchase of a Samsung Galaxy III S super sized phone (1280x720 screen resolution). I am amazed by how much I use this device: for just about everything but coding and writing. I do most of my programming in Ruby and Clojure so I don't need much horsepower on my development systems so I can imagine a future where a more powerful phone with a docking station might cover most of my work, writing, and entertainment needs: skip the tablet!

All that said, I am likely to purchase a Surface when they are released and the early adopter bugs are ironed out. I used to have fun bashing Microsoft but there are a few things that have softened my view: Bill and Melinda Gate's good works for making the world better, the fact that the surface would fit my needs, and I still have pleasant memories of writing a Windows 1.03 app (the SAIC ANSim neural network tools that had a nice UI for managing training, etc.)

jacobr 8 hours ago 1 reply      
He has a desktop computer with a bunch of monitors, so I guess his laptop is mostly for email, entertainment and web browsing. Of course a tablet can do that just fine. If you're using a laptop as your main machine, you have different needs.

Personally I am planning to retire my desktop computer and buy an ultrabook, probably a Samsung 900X3C. The Samsung is tinner and lighter, and I read it's a bit more solid in the build than the Asus.

Anyone tried both the 900X3C and a UX31A? Anyone tried Linux on them?

jbk 8 hours ago 0 replies      
I have been using the Ux31 (previous version, only 1600x900) on Arch and Windows 8, and I am quite impressed too.

I have been a heavy Thinkpad (T42 and T61p) and Macbook Pro user, and I had the chance to test other Ultrabooks, like the Samsung S9 and some cheap Acer. I was quite prejudiceed against Asus, but this machine is really the best I had so far.

john2x 9 hours ago 0 replies      
My biggest problem with PC laptops is getting Linux to run on it painlessly (their trackpads are a close second). How does Linux run on this one?
brador 9 hours ago 5 replies      
Is this an ad or genuine? I'd trust it more if it didn't come with an amazom affiliate link.
RexRollman 5 hours ago 2 replies      
He does bring out a point about something I have never understood: why does the PC industry love putting stickers on laptops? It is a really annoying and ugly practice.
Tichy 7 hours ago 0 replies      
I don't see myself typing long texts on a tablet or phone in the foreseeable future. For consuming on the road, it turned out the 10.1" tablet is actually too big and the phone is sufficient.

What about the Asus tablet with the detachable keyboard? I saw it in a shop recently and it looked surprisingly solid.

Another, perhaps stupid thought: with screen resolutions blowing up, could it become possible to use glasses to see a phone display in a size equivalent to 27"?

dbbolton 50 minutes ago 1 reply      
What's with the `?tag=codihorr-20` at the end of the Amazon link?
trotsky 7 hours ago 0 replies      
The first thing I did when I got the laptop was wipe it and install the Windows 8 preview

That's why it won't be your last laptop.

madprops 10 hours ago 1 reply      
Being and UX31A user myself I share your joy. I ditched windows and installed archlinux and tailored it to my liking. This is indeed one fine laptop.

I think that until we get something that makes keyboards and screens obsolete there will be room for a laptop.

roel_v 9 hours ago 1 reply      
Guess he didn't look very hard for the specs he's quoting. I had a Dell Dimension D430 which was awesome a few years ago (actually it was great until a few weeks ago when I dropped it) and got 6+ hours of battery life easily, and the Z-series Vaio I'm typing this on are just a teeny bit heavier than what he's talking about and also get similar battery life. Plus, for longer times without access to power outlets, there's always the second battery. Better to go from 5 to 10 hours by taking a second battery than going from 5 to 7 by adding extra weight that most of the time you don't use, imo.
DanBC 8 hours ago 0 replies      
It seems he wants a super-powerful tablet. That'd be reasonable for on the go. He'd have a good docking system at home to connect to proper monitor and keyboard, with big storage. There'd be a portable keyboard too, for proper work on the go.

He'd still have to take a phone.

And, of course, he'd have a struggle with OSs not knowing if they were using horrible tablet interfaces or desktop interfaces; and having to adjust between mouse and touch all the time.

lumberjack 8 hours ago 0 replies      
It is a bit strange that there is a differentiation between battery life and portability. As far as I'm concerned portability is (battery-life * (1/weight)), assuming that you give them the same importance.
mathattack 5 hours ago 0 replies      
Very good article. I think the leadup lends to a profound conclusion - all this work reaching perfection, and laptops may cease to exist in a few years.
webreac 6 hours ago 0 replies      
My conclusion of this article is that even after so many years, it is still not possible to find a decent computer that fits normal needs even when money is not the issue.

I completely agree with the needs of the article: something usable in a subway, with a good screen, 10 hours of battery life, a descent CPU, an (optional) keyboard for productivity and a descent OS.

danmaz74 7 hours ago 0 replies      
I have a UX32VD and I really like it - except for the fact that the integrated display sometimes doesn't wake up when it goes to sleep. Did anybody else notice the same problem?
jpkeisala 9 hours ago 1 reply      
I have considered Zenbook over my MacBook Air but I have not changed yet. I am somewhat reluctant moving back to Windows but in the same time I am not much of a Apple fan either. I guess I am on deadlock here, waiting a linux miracle. So, it all boils to software.
wildc4rd11 8 hours ago 1 reply      
By emphasizing that this is his last "PC" laptop, I assume he might still buy a laptop in the future but not a PC one. However, all the points listed in the article are reasons for why tablets are more awesome.
ASK HN: I made tarbackup.com for you, what do you think? tarbackup.com
13 points by nanch  6 hours ago   25 comments top 4
Cyranix 3 hours ago 1 reply      
Your Git repo doesn't inspire confidence for usage by the public at large. If you want to continue using it, I would recommend that you change it over to friends-and-family access while you improve it.

A few things that don't impress me right off the bat:

  * text file for queue of users to create
* not accepting passwords with non-alphanumeric characters
(and also not having the knowledge to recognize a use case
for a regular expression)
* using a static salt

To be honest, I just stopped looking at this point. That's all without actually looking at the front-end of it (which has several typos, by the way -- when you're selling yourself to a very large audience, every detail counts).

genwin 3 hours ago 2 replies      
Thanks! I understand what this service is: free "cloud" backup, no guarantees. I have no problem with the latter part, given the former part. I'll use it accordingly and appreciate your gift.
tisme 6 hours ago 1 reply      
See: tarsnap.com
ysleepy 3 hours ago 1 reply      
I see that you are passionate about this.

After I wrote a paragraph about how you could use the hardware for a shell server, static (octopress/jekyll) website hoster etc. I realised, that free backups are actually really awesome.

For people who cannot afford another service or dont have access to a credit card or else. Also fpr my most important data like code as a second backup if all goes wrong.

Thanks for the service, I will probably actually use it.

But it being a fun project and only behind 5mbit upload, this is obviously not a professional solution or could be offered for money.

Why Amazon.com associates can't make money even when Amazon does artymiak.com
20 points by surfingdino  5 hours ago   19 comments top 6
davidw 3 hours ago 2 replies      
I've always found Amazon a bit weird/frustrating with regards to i18n/localization.

As a user, I should be able to decide, independently,

* What language I want to use the site in.


* Where orders should come from.

In other words, if I'm an American living in Austria or Italy, I should be able to just go to Amazon.com, have the site in English, and have it route books/whatever from the nearest/lowest tax place. In some cases, that'd be Amazon.it, in others, Amazon.co.uk, and in some cases, it'd have to send things from the US, incurring higher taxes/tariffs.

Instead, Amazon.it and Amazon.de are very much only in "the" local language (Italy actually has areas where German and French are official languages), and routing does not seem intelligent at all.


gav 2 hours ago 1 reply      
As Amazon gets more and more established, the value of their affiliate program drops. They end up giving up margin for a sale they are likely to capture anyway.

I predict that as time goes on, they will cut back their affiliate program. Once you have the data to determine on a per-product-category basis what kind of affiliates are worth it and which aren't they next step is to prune back the latter.

lazyjones 1 hour ago 0 replies      
We have the opposite problem, our income from the associate program is growing at such a rate that we're becoming increasingly dependent on it (i.e. changes in their terms could have a huge impact on our company). Amazon is cornering many markets because other merchants have inferior websites and are often under (illegal) pressure from vendors to keep prices high (specifically in my country).
ljd 1 hour ago 0 replies      
I've been reading more blogs on how people are upset or confused by Amazon's economics. I'm not necessarily taking an opinion either way but it is important to realize that Amazon is commoditizing every product they can to increase their complementary product: the market itself.
duiker101 3 hours ago 4 replies      
I think the title is a bit misleading, it should be "Why blggers can't make money by linking to books". This is a very specific case. I know people who actually their only business model is Amazon association. Risky if you ask me(as always when you depend entirely on someone else), but it works.
paulhauggis 1 hour ago 0 replies      
Amazon is a terrible company. When will people realize this? They have been abusing their Internet marketplace monopoly for years because we don't really have any other choice.
       cached 23 September 2012 19:02:01 GMT