[...] there are some academic papers showing that heterogeneous cores are a good idea.
That's what the IBM/Sony/Toshiba "Cell" architecture is all about. It has one "real" PPC core and eight mini-PPC (SPEs) cores all in one processor. You can read more about it here.
On the subject of Nick Park and Aardman Animations, surely you're familiar with his first short, 1989's Oscar-winning Creature Comforts. Well, in 2003 they made a series of thirteen ten-minute Creature Comfort shorts. I discovered this quite by accident the other night; I turned on the TV and there it was! Comedy Central is showing 'em at apparently random times, two shorts in a half-hour time slot.
This is another example of the "Broken Windows" fallacy. The audits and the blogging and all had an opportunity cost; doing it meant not doing something else.
Look at it this way: if the SCO lawsuit hadn't happened, would the Linux community have spent their energies the same way? They could have, but they wouldn't have, because their time would have been better spent elsewhere. So the SCO lawsuits caused the Linux community to derail its energies into projects of diminished benefit.
if (!ptr) and if (ptr != NULL) are exactly the same thing. It's like asking "which is cheaper, six eggs or a half-dozen eggs?"
Show me the compiler that generates different code for those two snippets.
No, the reason he's got "lots o money" is because he won the AOL jackpot.
Nullsoft was a tiny company (just Justin Frankel and his dad iirc) selling shareware licenses to Winamp for $10 a pop. Then AOL, who following their IPO had more money than sense, decided they needed to own a media player. So they ambled over and offered to buy Nullsoft for $100m. Well, next thing you know, ol' Justin's a millionaire.
After that, Lycos said "Hey, I want one too!" and bought the Sonique project. Personally, I have yet to fathom the reason why these companies felt they had to own media players. It's not like they ever leveraged off 'em; they just gave it away for free, and didn't do anything with 'em. Maybe it had something to do with Microsoft announcing Windows Media Player, and keeping-up-with-the-Joneses. Who knows.
Anyway, let me point out that Steve Wozniak is a smart guy, has oodles of money to throw at new projects, and has yet to found a profitable company after Apple. And Woz has been at it for twenty years.
I wish Mr. Frankel success, but I suspect his new company will be much more like Cloud 9 than Nullsoft.
If I, as an American consumer, have $50 and want a pair of shoes, and I have the choice between a $50 pair of shoes made in America and a $30 pair of shoes made in China, then I'm clearly better off buying the $30 shoes--at the end of the transaction, I still have a pair of shoes, but I also have $20 more dollars.
Don't talk to me about "nations", or "trade imbalances". The fundamental unit in economy is the consumer, and anything beyond that is an abstraction. In this scenario, the consumer is better off because they can buy cheaper shoes. Period end of sentence.
What of China, who has artificially subsidised these shoes? If it costs them more money to make the shoes than they'll get by selling them, then they'll eventually run out of money and fail. Meanwhile, they have transferred their wealth to the American consumer. Subsidies are good for consumers, and rotten for the business attempting them.
Let's assume for a moment that they're not losing money; perhaps they made a small profit by employing cheap labor. That being the case, now the shoe manufacturer in China is richer as a result of the trade. And that's a good thing. History clearly shows that as a society becomes richer, it becomes freer. The Chinese shoemaker with extra money can now afford better school for his child, or to take time off to protest, or whatever he or she thinks is best. It affords him choice, and with choice inevitably comes freedom. If we want China to become a free society, we should trade more with the Chinese, not less.
You still haven't said how other people using crack and heroin can impact you.
The founding tenet of Libertarianism is this: nobody can use force anybody else to do anything. All exchanges between adults therefore must be achieved by persuasion, and any exchange between adults is therefore voluntary. A system that enforces only voluntary exchanges is one where everyone gets richer, because any time you make a voluntary exchange, both sides become richer (otherwise they wouldn't make the exchange)...
But I'm getting off track. This "no force" foundation of the Libertarian party has a lot of natural consequences, that makes predicting how a Libertarian will react to a question of policy pretty easy. The Drug War is an easy one: you simply don't have the right to force someone else to take drugs, or to not take drugs.
Honestly, making hard drugs legal was the hardest thing for me to accept when I started considering the Libertarian party a decade ago. I thought, "but those drugs are terrible! what of the crack babies in the hospitals?" I've come to accept that I simply don't have the right to tell others not to use such substances. And even if I tried, drug prohibition still wouldn't work, because buying and using drugs are victimless crimes.
The Drug War is one of the vilest things our government is involved in. In order to enforce these victimless crimes, the Government has granted itself incredible invasions, paving over the Bill Of Rights in the process. Mandatory sentencing and asset forfeiture in particular are heinous assaults on our civil liberties.
However, making drugs illegal hasn't reduced demand for them any. All it's done is restrict the supply. And when you restrict the supply, you drive the price up. Illegal drugs are now so expensive that it's created whole new mafias, filling our streets with armed drug dealers engaged in drive-by shootings.
But it doesn't even stop there! Doctors live in fear of procecution in case some bureaucrat decides they've "overprescribed" pain medication. Sick people who could benefit from medicinal marijuana are forced to hide from the federal government, flee the country, or do without. Junkies who are addicted to these artificially-expensive drugs are forced to steal to maintain their habits.
This is the result of making drugs illegal. Are you certain that drugs are so terrible that this (the inevitable outcome of their prohibition) is better? Are you certain that drug prohibition is the most compassionate policy?
Just as legalizing alcohol after the repeal of Prohibition didn't turn America into a country of drunks, so shall legalizing drugs not turn America into a country of potheads. And with drugs legal, their cost would drop, the incentives for "pushing" would disappear (you don't see people "pushing" cigarettes or alcohol, do you?), as would the whole villainous drug cartel, people would be more willing to seek treatment... all sorts of wonderful things would happen, honest.
I for one think the Drug War is simply an untenable position; sooner or later the government will abandon it, and then you'll see how much better off we are without it.
By the way, in case you think I'm kidding about the "no force" policy being the bedrock of the LP's position, take a look at the form to join the party. There's a special declaration that says (quote approximate) "I (the undersigned) declare that I believe no person should be able to use force to compel another person.". If you don't sign that, you can still be a member of the party, but without nominating or voting privileges.
So, lemme get this straight. You view him as nearly perfect when it comes to his plan for leadership of this country, but you're not going to vote for him because he has some funny conspiracy ideas?
Who do you regard as the superior candidate, the person who will therefore get your vote?
Personally, I agree, that Badnarik has some funny ideas. His website was expunged of some really odd stuff just after he got the nomination. But that won't stop me from voting for him, as a) they wouldn't affect his performance on the job, and b) there's not a chance he'll win anyway. (It'd be great if he did win, but surely it's not realistic to expect that he could, this year.)
NT 4.0 was faster than 3.5/3.51, on the same hardware.
Windows 2000 was faster than NT 4.0, on the same hardware.
I was there, I used them, I remember.
These speedups were sometimes achieved by moving away from NT's originally-vaguely-microkernel approach to a more monolithic kernel. For instance, I'm told that for NT 4.0 they moved the display driver into ring 0. This meant fewer ring transitions to blit to the screen, hence, it ran faster, at the cost of robustness (buggy display drivers could now BSOD your machine).
Of course, I expect that most speedups were gained through algorithmic optimization, or fresh approaches to problems. For instance, Linux 2.6 is faster and more responsive than 2.4, even though it does more. Unless you assert that the NT 3.1 code was perfect and could not be optimized (an assertion I suspect you wouldn't make), surely you must allow that their team of bright programmers working year-round on the OS could figure out ways to speed it up.
Please don't make up blind assertions and claim they are fact, "period".
Honest, DirectX does allow you to write portable code. I've been working on a game for two years, and my game is totally platform-agnostic. Don't blame DirectX, blame the game programmers, and/or the box designers.
But DirectX also allows you to write part-specific code, if you so choose. DirectX provides a rich set of diagnostics that let you peer into exactly what capabilities the card provides. If you really cared about compatibility, your renderer would examine this information then include/exclude rendering features to get the prettiest graphics by using the best features of the card. However, if you were under time pressure, and/or just lazy, you could ignore this information, and just code one rendering path that worked well on the card(s) you had handy. Or code two paths, and only enable the good one on parts that meet your requirements (which happen to conform to one manufacturer's parts but perhaps not the other's).
Anyway, I suggest to you that a) when "most games" specify specific parts on the box, they'll run fine on the next generation of cards from any manufacturer, and b) the reason that "most games" specify parts is because consumers don't know the innermost secrets of their cards' abilities. Consumers are more likely to understand "GeForce 1 or better" than "must support hardware T&L".
As for Windows 95 letting games run on "more than a couple of graphics cards"... they did claim that, and hey! that's 'cause it was true. Writing your game for Windows 95 was way better than writing it for DOS. You didn't have to understand each graphics card and sound card for yourself, Windows 95 would abstract it away for you. But it still gave you a low-level and pretty fast way of blitting onto the screen (see the WinG library). Remember telling Doom what sound card you had? Sure, we all do. Remember telling Doom 95 what sound card you had? Oh, that's right, nobody had to.
Perhaps not in the initial revisions. But it would be wonderful for desktop machines, the big new market for LCD. Especially for l33+ gamerz like myself, who want to drive their displays at more than 60Hz.
Anyway, widespread adoption of the technology will drive down price, and likely someone will invent a similar technique for low-power situations like laptops, and everybody wins. Hooray for the free market!
Golly! And here I thought they'd been working on their "Object File System" for about ten years now, predating even the original release of BeOS. (See "Cairo".) But you state, as absolute fact, a) that the sum of their labors is only a duplicate of the clever-but-not-overly-ambitious functionality of the BeOS file system, and b) they are so dumb they haven't been able even to accomplish that.
I haven't seen WinFS, and neither have you. So, unlike you, I'm not going to state facts about it. But I suspect there's a lot more to it than the BeOS FS and ReiserFS database stuff. Whether that's good, bad, or indifferent, I also can't say. My pet theory is that the project has kept slipping over the past decade because it is ill-defined... if you don't know what you want, it's hard to get it.
Remember also that Microsoft has so much money they can hire shedloads of smart people. Ten years of smart people working on a problem is likely to result in something more interesting than "relational database as filesystem".
If you really hate Microsoft, the worst thing you could do is underestimate them.
Remember "San Francisco Rush: 2049 Edition"? I think the official release was built on Windows, but they later ported it to Linux. Its never-released sequel "Hot Rod Rebels" was Linux only.
At CA Extreme over last weekend I also saw a game called "Crossfire Extreme Paintball" that had crashed to a Linux prompt.
It makes a lot of sense to run Linux on arcade machines. You won't have to extend it much (if at all) to get at all the machine's hardware, and it'll save you some cash per-unit.
What's not true? You don't actually refute the quoted statement. You seem to be refuting something to do with locals and tourists, but nobody (including you) made a statement to that effect.
See, you have to set up your paper tigers before you knock them down so triumphantly.
For capitalism to work as per the definition of capitalism, consumers must be "perfectly informed".
For capitalism to work perfectly the market would have to have perfect knowledge. But, gosh, we seem to limp along pretty well with imperfect knowledge, huh?
Anyway, yeah, capitalism as a means of allocating production sucks. It just happens to suck less than all the others.
The solution is for the government to mandate a warning tag like the warnings on cigarette boxes.
Given the dismal failure that warning labels on cigarette packs have been, I'm surprised that's the program you want to emulate.
What's so wrong with letting the market regulate itself? You seem to be pretty well informed about RFID tags, and yet "the government" has enacted no RFID-related legislation to date. Seems to me like the system (aka "capitalism") is working.
Why not vote with your dollars, rather than turning once again to the Men With Guns?
Does the name of the scripting language contain the letters 'P', 'y', 't', 'h', 'o', and 'g', in that order? Score one point for every correct letter. Perl gets 1 point--oh, too bad! Python, with 5 points out of 6, is our highest-scoring language.
Personally I think that's a better metric than the author's simple-minded "the program in this language was eight characters shorter so it gets a higher score". This was apparently an attempt to measure the expressive power of a language, but done in such a way as to ensure Perl scored well.
I think there are more important things to keep in mind when selecting a scripting language, like
how easy is it to learn?
how easy is it to remember how to use everything?
how easy is it to read other people's programs
how easy is it to read your own code six months from now?
On these counts Perl loses big time. Its "every combination of punctuation characters does something" language design saves you typing, at the expense of pretty much everything else.
If the only thing you use the scripting language for was one-off half-page scripts that you use once then throw away, and you wrote those eight hours a day five days a week, then I could see using Perl. For any other purpose I think you're better off using something else. (And, yes, I have a suggestion in mind, and it rhymes with Crython.)
larry
p.s. I have a nomination for a shorter Python program to remove #-characters from a file:
import sys system ("sed -i -e \"s/#.*//\" " + sys.argv[1])
After all, it is precisely as "legal" for the Python program to call sed as it is for the sh program to call sed.
Slashdot Mirror
So apply for CEO!
larry
larry
larry
Look at it this way: if the SCO lawsuit hadn't happened, would the Linux community have spent their energies the same way? They could have, but they wouldn't have, because their time would have been better spent elsewhere. So the SCO lawsuits caused the Linux community to derail its energies into projects of diminished benefit.
larry
if (!ptr) and if (ptr != NULL) are exactly the same thing. It's like asking "which is cheaper, six eggs or a half-dozen eggs?" Show me the compiler that generates different code for those two snippets.
"The question of whether computers can think is precisely as interesting as the question of whether submarines can swim."
Isn't that -2?
Nullsoft was a tiny company (just Justin Frankel and his dad iirc) selling shareware licenses to Winamp for $10 a pop. Then AOL, who following their IPO had more money than sense, decided they needed to own a media player. So they ambled over and offered to buy Nullsoft for $100m. Well, next thing you know, ol' Justin's a millionaire.
After that, Lycos said "Hey, I want one too!" and bought the Sonique project. Personally, I have yet to fathom the reason why these companies felt they had to own media players. It's not like they ever leveraged off 'em; they just gave it away for free, and didn't do anything with 'em. Maybe it had something to do with Microsoft announcing Windows Media Player, and keeping-up-with-the-Joneses. Who knows.
Anyway, let me point out that Steve Wozniak is a smart guy, has oodles of money to throw at new projects, and has yet to found a profitable company after Apple. And Woz has been at it for twenty years.
I wish Mr. Frankel success, but I suspect his new company will be much more like Cloud 9 than Nullsoft.
Remember, folks, think before you post!
larry
Don't forget, in two months we'll get "Doom 3 Now Available On Linux" as submitted by Roland Piqualle!
If I, as an American consumer, have $50 and want a pair of shoes, and I have the choice between a $50 pair of shoes made in America and a $30 pair of shoes made in China, then I'm clearly better off buying the $30 shoes--at the end of the transaction, I still have a pair of shoes, but I also have $20 more dollars.
Don't talk to me about "nations", or "trade imbalances". The fundamental unit in economy is the consumer, and anything beyond that is an abstraction. In this scenario, the consumer is better off because they can buy cheaper shoes. Period end of sentence.
What of China, who has artificially subsidised these shoes? If it costs them more money to make the shoes than they'll get by selling them, then they'll eventually run out of money and fail. Meanwhile, they have transferred their wealth to the American consumer. Subsidies are good for consumers, and rotten for the business attempting them.
Let's assume for a moment that they're not losing money; perhaps they made a small profit by employing cheap labor. That being the case, now the shoe manufacturer in China is richer as a result of the trade. And that's a good thing. History clearly shows that as a society becomes richer, it becomes freer. The Chinese shoemaker with extra money can now afford better school for his child, or to take time off to protest, or whatever he or she thinks is best. It affords him choice, and with choice inevitably comes freedom. If we want China to become a free society, we should trade more with the Chinese, not less.
larry
The founding tenet of Libertarianism is this: nobody can use force anybody else to do anything. All exchanges between adults therefore must be achieved by persuasion, and any exchange between adults is therefore voluntary. A system that enforces only voluntary exchanges is one where everyone gets richer, because any time you make a voluntary exchange, both sides become richer (otherwise they wouldn't make the exchange)...
But I'm getting off track. This "no force" foundation of the Libertarian party has a lot of natural consequences, that makes predicting how a Libertarian will react to a question of policy pretty easy. The Drug War is an easy one: you simply don't have the right to force someone else to take drugs, or to not take drugs.
Honestly, making hard drugs legal was the hardest thing for me to accept when I started considering the Libertarian party a decade ago. I thought, "but those drugs are terrible! what of the crack babies in the hospitals?" I've come to accept that I simply don't have the right to tell others not to use such substances. And even if I tried, drug prohibition still wouldn't work, because buying and using drugs are victimless crimes.
The Drug War is one of the vilest things our government is involved in. In order to enforce these victimless crimes, the Government has granted itself incredible invasions, paving over the Bill Of Rights in the process. Mandatory sentencing and asset forfeiture in particular are heinous assaults on our civil liberties.
However, making drugs illegal hasn't reduced demand for them any. All it's done is restrict the supply. And when you restrict the supply, you drive the price up. Illegal drugs are now so expensive that it's created whole new mafias, filling our streets with armed drug dealers engaged in drive-by shootings.
But it doesn't even stop there! Doctors live in fear of procecution in case some bureaucrat decides they've "overprescribed" pain medication. Sick people who could benefit from medicinal marijuana are forced to hide from the federal government, flee the country, or do without. Junkies who are addicted to these artificially-expensive drugs are forced to steal to maintain their habits. This is the result of making drugs illegal. Are you certain that drugs are so terrible that this (the inevitable outcome of their prohibition) is better? Are you certain that drug prohibition is the most compassionate policy?
Just as legalizing alcohol after the repeal of Prohibition didn't turn America into a country of drunks, so shall legalizing drugs not turn America into a country of potheads. And with drugs legal, their cost would drop, the incentives for "pushing" would disappear (you don't see people "pushing" cigarettes or alcohol, do you?), as would the whole villainous drug cartel, people would be more willing to seek treatment... all sorts of wonderful things would happen, honest. I for one think the Drug War is simply an untenable position; sooner or later the government will abandon it, and then you'll see how much better off we are without it.
By the way, in case you think I'm kidding about the "no force" policy being the bedrock of the LP's position, take a look at the form to join the party. There's a special declaration that says (quote approximate) "I (the undersigned) declare that I believe no person should be able to use force to compel another person.". If you don't sign that, you can still be a member of the party, but without nominating or voting privileges.
larry
Who do you regard as the superior candidate, the person who will therefore get your vote?
Personally, I agree, that Badnarik has some funny ideas. His website was expunged of some really odd stuff just after he got the nomination. But that won't stop me from voting for him, as a) they wouldn't affect his performance on the job, and b) there's not a chance he'll win anyway. (It'd be great if he did win, but surely it's not realistic to expect that he could, this year.)
larry
NT 4.0 was faster than 3.5/3.51, on the same hardware.
Windows 2000 was faster than NT 4.0, on the same hardware.
I was there, I used them, I remember.
These speedups were sometimes achieved by moving away from NT's originally-vaguely-microkernel approach to a more monolithic kernel. For instance, I'm told that for NT 4.0 they moved the display driver into ring 0. This meant fewer ring transitions to blit to the screen, hence, it ran faster, at the cost of robustness (buggy display drivers could now BSOD your machine).
Of course, I expect that most speedups were gained through algorithmic optimization, or fresh approaches to problems. For instance, Linux 2.6 is faster and more responsive than 2.4, even though it does more. Unless you assert that the NT 3.1 code was perfect and could not be optimized (an assertion I suspect you wouldn't make), surely you must allow that their team of bright programmers working year-round on the OS could figure out ways to speed it up.
Please don't make up blind assertions and claim they are fact, "period".
But DirectX also allows you to write part-specific code, if you so choose. DirectX provides a rich set of diagnostics that let you peer into exactly what capabilities the card provides. If you really cared about compatibility, your renderer would examine this information then include/exclude rendering features to get the prettiest graphics by using the best features of the card. However, if you were under time pressure, and/or just lazy, you could ignore this information, and just code one rendering path that worked well on the card(s) you had handy. Or code two paths, and only enable the good one on parts that meet your requirements (which happen to conform to one manufacturer's parts but perhaps not the other's).
Anyway, I suggest to you that a) when "most games" specify specific parts on the box, they'll run fine on the next generation of cards from any manufacturer, and b) the reason that "most games" specify parts is because consumers don't know the innermost secrets of their cards' abilities. Consumers are more likely to understand "GeForce 1 or better" than "must support hardware T&L".
As for Windows 95 letting games run on "more than a couple of graphics cards"... they did claim that, and hey! that's 'cause it was true. Writing your game for Windows 95 was way better than writing it for DOS. You didn't have to understand each graphics card and sound card for yourself, Windows 95 would abstract it away for you. But it still gave you a low-level and pretty fast way of blitting onto the screen (see the WinG library). Remember telling Doom what sound card you had? Sure, we all do. Remember telling Doom 95 what sound card you had? Oh, that's right, nobody had to.
Perhaps not in the initial revisions. But it would be wonderful for desktop machines, the big new market for LCD. Especially for l33+ gamerz like myself, who want to drive their displays at more than 60Hz.
Anyway, widespread adoption of the technology will drive down price, and likely someone will invent a similar technique for low-power situations like laptops, and everybody wins. Hooray for the free market!
I haven't seen WinFS, and neither have you. So, unlike you, I'm not going to state facts about it. But I suspect there's a lot more to it than the BeOS FS and ReiserFS database stuff. Whether that's good, bad, or indifferent, I also can't say. My pet theory is that the project has kept slipping over the past decade because it is ill-defined... if you don't know what you want, it's hard to get it.
Remember also that Microsoft has so much money they can hire shedloads of smart people. Ten years of smart people working on a problem is likely to result in something more interesting than "relational database as filesystem".
If you really hate Microsoft, the worst thing you could do is underestimate them.
At CA Extreme over last weekend I also saw a game called "Crossfire Extreme Paintball" that had crashed to a Linux prompt.
It makes a lot of sense to run Linux on arcade machines. You won't have to extend it much (if at all) to get at all the machine's hardware, and it'll save you some cash per-unit.
larry
See, you have to set up your paper tigers before you knock them down so triumphantly.
But they'll go great with my Sennheiser blade server and Sony laptop! Oh, wait.
For capitalism to work perfectly the market would have to have perfect knowledge. But, gosh, we seem to limp along pretty well with imperfect knowledge, huh?
Anyway, yeah, capitalism as a means of allocating production sucks. It just happens to suck less than all the others.
Given the dismal failure that warning labels on cigarette packs have been, I'm surprised that's the program you want to emulate.
What's so wrong with letting the market regulate itself? You seem to be pretty well informed about RFID tags, and yet "the government" has enacted no RFID-related legislation to date. Seems to me like the system (aka "capitalism") is working.
Why not vote with your dollars, rather than turning once again to the Men With Guns?
Personally I think that's a better metric than the author's simple-minded "the program in this language was eight characters shorter so it gets a higher score". This was apparently an attempt to measure the expressive power of a language, but done in such a way as to ensure Perl scored well.
I think there are more important things to keep in mind when selecting a scripting language, like
- how easy is it to learn?
- how easy is it to remember how to use everything?
- how easy is it to read other people's programs
- how easy is it to read your own code six months from now?
On these counts Perl loses big time. Its "every combination of punctuation characters does something" language design saves you typing, at the expense of pretty much everything else.If the only thing you use the scripting language for was one-off half-page scripts that you use once then throw away, and you wrote those eight hours a day five days a week, then I could see using Perl. For any other purpose I think you're better off using something else. (And, yes, I have a suggestion in mind, and it rhymes with Crython.)
larry
p.s. I have a nomination for a shorter Python program to remove #-characters from a file:
After all, it is precisely as "legal" for the Python program to call sed as it is for the sh program to call sed.Not at all, mon ami. PostScript is a very real, very capable language. It's based on Forth but adds support for real types.
Personally I'm more impressed that someone wrote a TCP/IP stack in PHP, but then I have a pretty low opinion of PHP.
And on /. no less!