6ren

  easy to use
  easy to buy

Id did open source some games, but only well after the main burst of sales had dried up.

Counterpoint: reddit and HN are open sourced (are they games? they have points...)

Just some data points, not advice.

BONUS: for next year, the G6630 gives 50% performance increase + the expected node shrink allows a further ~30% overclock. 1.5*1.3 = 1.95 ~ 2.0.

For the following year, they could go for G6630MP2 (the 6 clusters can be multicored as a whole). Next (three years from now), they could go G6630MP3 plus another node shrink. By that time, there could be 8/10 clusters available G6830, G6A30 etc, and an entirely new architecture in the works, though probably not yet available.

Portuguese is the main language of Brazil, and adapting to those speakers and customs may lead to Portugal and the other seven Portuguese-speaking nations http://en.wikipedia.org/wiki/Portuguese_language I expect it might also lead to Spanish speaking countries better than does English. BONUS: Brazil's developer community was strong enough to produce Lua - other non-English speaking countries with a similar claim include Japan (Ruby) and the Netherlands (Python). (Though to be fair, internet collaboration overshadows local community.)

Counterpoint: the popularity of Google's Orkut in Brazil didn't lead to other markets... so it might be a tricky strategy to pull off.

  The iPhone 5S (A7) has x1 the graphics power as the iPad 4.
  The iPad 5    (A7) has x2 the graphics power of the iPad 4.

I think the reason to buy an iPad 2 (or indeed an iPad 5) is for the larger display.

Have they just stopped using the A7X style name? Have they just x2 the GPU frequency? (seems implausible, but it's Rogue, which is apparently extremely efficient, and maybe could handle x2 frequency?).

edited above for clarity

BTW: The anandtech iPhone 5S review (http://www.anandtech.com/show/7335/the-iphone-5s-review/7) noted that the A7's rogue G6430 is more powerful than was utilized in the iPhone 5S:

  At 200MHz that would give the A7 twice the peak theoretical performance of the GPU
  in the iPhone 5. And from what I’ve heard, the G6430 is clocked much higher than
  that.

But they do editorialize on top (and cough and sniffle)... fortunately, only occasionally.

For Melbourne, Australia (same timezone as Sydney), it's at 4am (4 hours from now).

This site shows the hours until the event: http://www.macrumors.com/

Did this stop you or how did get past it?

How does Learn Haskell Fast and Hard compare?

Not better enough - and no new platform to carry it unlike JS, Perl, Ruby, Python, PHP, Java (web); Objective-C, Java (mobile); Java, Ruby, Python (cloud).

The problem is that people don't like to switch. They'd prefer their present tool to improve, and they'll support that. So the only way a language can achieve ascendency is when users don't switch. That is, new users. During rapid adoption of a platform, a significant majority can be new users. And that is the only time a new language can take over.

NOTE: it may appear there's an exception with flavour-of-the-month frameworks etc, but this is a combination of rapid turnover of developers/projects in specific areas, plus the enthusiast/techie kind of developers who love to switch (the ones always excited about the newest tech for its own sake, not for its long-term practical benefits: think inventor, not investor). These developers cannot form the basis of language's future, because they will quickly switch to the next new thing.

  > A rabbi named Zusya died and went to stand before the judgment seat of God. As he
  > waited for God to appear, he grew nervous thinking about his life and how little
  > he had done. He began to imagine that God was going to ask him, "Why weren't you
  > Moses or why weren't you Solomon or why weren't you David?" But when God appeared,
  > the rabbi was surprised. God simply asked, "Why weren't you Zusya?"

Also striking is the lack of photos - I suspect it wasn't literally "like sailing through a garbage tip" (as opposed to occasional debris, much more than usual).

But (I think) he means as a product, not a service (i.e. for bounties, only the asker pays, so it's a service; though it's true that everyone else benefits too).

I like this idea. The big problem, as always, is the boundary or interface of the module/component. (1). The buyer has to specify it, which is usually the difficult bit and writing it is usually easy. (2). For a market, you need to find modules that many people need, that people can search for and find, and that are difficult enough to implement that people would buy rather than build.

Maybe it's possible, for things that seem one-off or too specialized to a person, if you reduced friction enough, in a SO way. They've solved the search problem.

I guess an example would be a square-root function. I wouldn't know how to code it, I'd have to look it up, test and debug. It would take at least hours, I'd think. Of course, it's such a useful one, it's already done. That's the fate of all generally useful functions. The exceptions are new ones and niche ones.

I guess you could just add a little "debugged code" box for SO, and the person keeps coming back to it, to support it. This might be a useful addition; but I think you're right that money would spoil it. However... if you imagine a spectrum from one-liner to massive library, there's probably some level where the amount of work would make it reasonable to pay.

Conclusion: to see the value in the idea, leave money out (build something people want). How would you modify SO, or start a new one, to provide this value, or reusable functions/code snippets?

Discussion (though might be dying): http://discuss.joelonsoftware.com/?biz

Withholding Tax (if you are non-US, selling to US) "Components" are taxed at 5%; "desktop tools" (like compilers, editors, IDEs etc) are taxed at 0%. The tax is an admin hassle all round, so this is one plus for not selling components. (NB: this is an oversimplification, there's tax treaties, 30% tax if not under a treaty, byzantine IRS forms, tax rulings on the definition of "royalty" for components, etc).

ASIDE an idea: "Software Components as a Service". That is, it's in the cloud and paid for like a SaaS; but the service it provides is not an end-point (like an API to a business), but a transformation - the kind of thing you'd normally call a library for. e.g. text->pdf. Sounds inefficient, but you co-host with AWS/dropbox/heroku/GoogleApps etc so bandwidth is fast and free. (Alt: let the customer install a copy in the cloud, but you charge like a SaaS business model - an advantage for them is they can expense it monthly, not taxed like a capital purchase).

I've seen some evidence of this, in the form of cloud integration services e.g. http://www-03.ibm.com/software/products/us/en/castiron-cloud... (2010), but it's still mostly about integrating end-points, not the bits in between.

I accomplished this. But it was by definition a stationary target, and competitors (in this case, open source) largely caught up. But worse, I lost interest, because I was no longer "the only/best in the world" at it.

Success: I not only made enough to live on, I made enough to invest in index funds giving a return that was itself enough to live on, and "retire" (for small values of retire, i.e. a student-level lifestyle - "ramen retirement"). Business began in 2000, able to retire in 2004 or so IIRC. Have lived off the business/investments since then.

  lessons learned:

- I tried to make it too much a business; if I'd stayed truer to my roots, it might have been OK. If you have an open-source version (or, at least, free), it starves off the competition, leaving them with no oxygen to get started, because no one needs it, they get no interest. And everyone's happy, which actually is very important for one's personal happiness too.

- I actually made a lot of money selling to the enterprise; but I hated it. Like, it really really got to me. In one particular sale, there was 5-6 months of negotiation, admin, international tax laws, etc. Ugh. And it all feels dishonest, pointless, and interacting with people who aren't competent because aren't interested and don't care. (the standard story of a developer dealing with business issues). So, you can do it this way, if you keep going up-market. The enterprise likes old stuff, that is trusted, field-tested, known etc.

- turns out I don't enjoy making a fixed standard. There's beauty in it as a business model; but for a technologist, the fun part is making it (and for me, supporting and debugging was also fun). But when it's all done, all working perfectly, it stops being fun. It's a funny old world.

- the underlying factor is that progress marches on. It's good for the world, and it's fun to participate. But not so good for a self-sustaining business with a competitive moat (Warren Buffett likes candy, chewing gum, soft-drink, bricks, furniture etc - stuff that doesn't change fast). You have to keep investing in it - developing new features, new products, new users (analogous to the capital investment required by the cotton mills of Berkshire Hathaway, making them a terrible investment). But this is all OK, if it's fun for you. And it can be.

That is, like a vector display, but using the LEDs of a flat screen. There's no electron ray scanning over the phosphors as there was in a cathode ray tube.

--

The LEDs can switch only so fast - but does this latency prevent them from being switched independently?

The LEDs need to stay on for long enough for the human visual system to perceive them (and without flickering etc) - but this could be managed in other ways.

I think the main reason, apart from inertia, is that the larger market for displays is as TVs, where the concept of updating the whole frame (frames per second) is even more entrenched - though, there's no reason why video couldn't be displayed in the same way, it's just pushing out a kind of compression to the display itself. Light from real objects is not emitted one frame at a time.

"On the other hand, the world is littered with thousands of spare words which spend their time doing nothing but loafing about on signposts pointing at places. Our job, as wee see it, is to get these words down off the signposts and into the mouths of babes and sucklings and so on, where they can start earning their keep in everyday conversation and make a more positive contribution to society. http://en.wikipedia.org/wiki/The_Meaning_of_Liff http://folk.uio.no/alied/TMoL.html

  KENTUCKY (adv.)
  Fitting exactly and satisfyingly. The cardboard box that slides neatly into an exact
  space in a garage, or the last book which exactly fills a bookshelf, is said to fit
  'real nice and kentucky'.

I was questioning your analogy with Apple. I'm sure they could win the low-end; but they don't want to.

Addressing the question of whether Musk can win it... well, firstly it is uncertain (and he might get hit by the proverbial bus), so we have to factor that in. But, some first mover factors in his favour: he's further along the engineering learning curve than anyone else; and further along winning market headspace; and further along getting every little business thing right to make it all happen. That is, he has a high probability of succeeding.

Secondly, I think Toyota or General Electric are better comparisons than with GM (whose cap reflects its terrible shape), which are 5 times bigger. (see the figures in my other comment: https://news.ycombinator.com/item?id=6567057) That is, the potential prize is larger than given in the article.

Whether Tesla is over-priced or not I couldn't say - except that the market overreacts to both good news and bad news, so it's probably best to wait for a dramatic stuff up that doesn't actually affect its real business prospects, and buy the depressed stock.

But Musk doesn't want to be high-end. He wants to change the world, get us to all buy electric cars. That means more affordable cars, which is the direction he's going. He only started high-end because that was the best place to start. Of course, there are many challenges before he's done.

(Goodness, why is this being downvoted?)

  TSLA	Tesla Motors Inc	182.14	-1.42	-0.77%		 22.13B
  TM	Toyota Motor Corp...	129.95	-0.29	-0.22%		205.93B
  GE	General Electric ...	24.30	-0.06	-0.23%		247.50B

GE is not a bad comparison: founded by Edison, based on electricity, created power station infrastructure...

The striking thing is that, per Hz, x86 is still far more powerful than ARM (like, integer factors, I've heard x5). If they can get the power consumption low enough and/or more performance is needed in phones/tablets, they could possibly win (though it seems unlikely).

Their best bet is to lead the next disruptive category - watches or glasses or whatever it is - but unfortunately their bias towards performance over power consumption doesn't bode well for them.

Keeping them seems one benefit to dual OS (reusing the Chrome kernel). It's nice to be able to watch HD video without hiccups.

Not sure if that's fast enough, but at least it's better than the typical 20MB/s for USB 2.0.

I think your marketing approach is what will make or break it: "non-sweaty to dates/interviews" and the phone-charger could be the killer app (perhaps look into what segment actually experiences this pain most? Many of today's phones have adequate battery life for some people's usage; target those phones + usages that don't).

  $ time sort < t > t-sh
  real    0m0.016s
  user    0m0.008s
  sys     0m0.008s

  $ time py -c "import sys; open(sys.argv[1],'w').writelines(sorted(open(sys.argv[2]).readlines()))" t-py t
  real    0m0.088s
  user    0m0.056s
  sys     0m0.012s

  compile time with -O3 is roughly the same as Python VM startup and has to be done only once

But I like his emphasis on large standard libraries, enabling compact scripts, esp string processing, and memory managed. WHy couldn't you do this trick with C (i.e. compile & run)? It's mainly libraries, though memory management isn't natural. Actually, I could believe that many scripting languages actually started like that, but shifted to their own syntax asap.

This trick can also be done with java, by keeping a server in the background to run it to cheat the VM startup tax (and auto-compiling as needed). Java verbosity is a problem, but you can write C-like code in Java. The biggest problem is the detail of Java libraries - they give you a lot of control, but a scripting language should give you less control, in return for quick functionality (like unix `sort`).