Are you sure about that? Getting a paper onto arxiv.org doesn't seem to be that hard, and there's lots of ways to find out about it (RSS feed, etc.). He may not have had any reason to believe that he'd get this sort of attention, as he may have thought everyone involved would simply assume that it wasn't worth much, not having been peer reviewed.
While I love the free and open flow of information that arxiv represents, this is hardly the first time that something has been posted on there and subsequently blown out of proportion. The Internet at large doesn't seem to really understand arxiv.org, that just because someone's got a fancy LaTeX paper up claiming some wild thing doesn't mean it's credible. A paper on arxiv.org shouldn't even be understood as being endorsed by the author, let alone "science". I always love when somebody backs up their argument about physics with a link to arxiv.org, it's like a red flag that it's time to just pack it in, you're not going to get through to this person, because they only understand the trappings of science, not the actual process.
Diablo is a graphical Roguelike; in particular, it's basically a somewhat-dumbed-down*-but-much-prettier Angband. If Hack/Moria/Angband make the cut, there's no way to exclude Diablo.
(*: Not really a criticism. The Roguelikes often take advantage of their essentially textual nature to do things that simply can't be done graphically, or would be fantastically difficult. Anybody want to write the code that morphs any given monster type into any other given monster type for NethackGL? Not me!)
The basic technology is "make an HTTP request from JavaScript"; not exactly bloated.
Who said anything about "bloat"? Bloat is a boogeyman.
Sockets solve two problems, and actually solves them, not just hacks around them. It allows payloads that aren't HTTP, in particular including small payloads of a few bytes without the massive 10,000%+ overhead of wrapping an HTTP request around it, and it allows clean, effective "push" to the browser without it being a hack.
If you can't imagine how that might be useful, I suggest an imagination upgrade. The reason why you currently conceptualize these basic, 1960s-computing capabilities as "overkill" is because you, and apparently everybody else, have so deeply internalized the idea that everything must be HTTP requests with all the associated overhead that you can't even imagine a world without them. But there's hardly a top-end web app that wouldn't benefit from one or both of these features, and by "benefit" I mean real and concrete benefits to the user, with faster, more capable, more responsive interfaces that at the same time put less of a load on the server.
If you are sophisticated enough to load pirated ROMs onto a card for your DS, then you are sophisticated enough to read reviews and figure out whether something is "shovelware" before you buy it. Since the Internet got big, I have not yet bought a true dud of a game. I've taken a couple risks, not all of which panned out, but I knew what I was getting into.
(Yes, you can't just take the score for granted, either, but I also assume that if you can load software onto your DS, you are sophisticated enough to see through that, too. I play a lot of games that only get an 8/10, docked a point or two for "niche appeal", and wouldn't care to play many 9.5/10 games. Again, I don't have trouble with that.)
The problem if you go too far in the "what's useful" direction is that you don't know what will be useful until you know what you're going to do.
That argument applies far more directly to trig than it does to my suggested replacement. Game theory is useful to everyone and also will teach you many things about math that nothing else in the current cirriculum does; trig identities is only useful in very limited situations. (Please note I'm not advocating removing trig itself, which would be something entirely different.)
"Useful" isn't an adequate summary of my point, because you have to define "useful for what". And the very reason I am critical of trig identities is that by the most relevant definition I can come up with, "useful for mathematical education, understanding, and future mathematical study", trig identities aren't useful.
No shit, sherlock; the different spelling should be a dead giveaway. But what population is going to be the biggest users of optics? Hint: There aren't that many telescope designers in the world.
Well then, get ready for the people taking optics courses to be confused when you take out trig identities.
Seriously? That's the best you can come up with? Optometrists? All.03% of them, or whatever?
Here, you've provided me an excellent demonstration of why we need game theory in high school, because your post neglects the vitally important concept of opportunity cost, something that I'd much rather the general populace had exposure to than something as useless as trig identities. The opportunity cost of teaching trig identities when you could be teaching, say, opportunity costs, is way too high. Trig is not even close to the best thing we could be spending our time on.
Besides, optometrists and surveyors are invited to take specialized courses in trig identities, just as the mere fact that I took a course on the mathematics of evolutionary computation doesn't even begin to imply that everybody in high school should learn about that stuff. Time is finite. Opportunity costs are important. Trig identities are too expensive and displacing a lot of stuff that is both useful in real life and more useful to mathematicians, who, like I said, don't consider them important.
(Actually, the disconnect between real mathematicians and mathematical education is truly staggering once you fully understand it. The educational community, and I say this with full consideration to the people involve, wouldn't know math if it bit them on the ass.)
I've often considered how I would re-write the math curriculum if I had a chance, and while I take some things out and put some things back in as time goes by, I have two constants that never seem to change:
Trigonometric identities go bye-bye. Even real mathematicians consider them little more than curiosities. Bring them out when you do Taylor expansions, put them away again when you're done.
Game theory is in. I'd happily trade Calculus for Game Theory for "non-Engineering bound students". Game theory is fantastically useful, even if you don't (or can't) actually "compute" with it in the real world, the concepts serve you in economics, politics (how many maths can make that claim with a straight face?), and business.
If the consultant really expected you to email the data, and expressed even a modicum of surprise that you wouldn't do it, they've already disqualified themselves from being able to securely handle your data.
Do you really think that this is the only flaw in their handling of sensitive data? That, otherwise, they are security conscious and careful, except for this odd flaw where they don't understand how insecure email is?
If you care, it's time to change consultants.
If you don't care, just email it already.
(I'm actually not quite as rigid as this may sound out-of-context. I don't agree that security is all-or-nothing, so please don't strawman me that way. My second paragraph is important; anyone who expects those things emailed to them is so far away from the necessary knowledge and skills that debating whether they are close enough or whether they will be able to take reasonable care is a waste of time, arguing about whether the receiver made a touchdown when they got tackled on the 10 yard line on the wrong side of the field.)
No. Not even close. Slashdot readers happen to identify with a group of people that gives each other points for slagging on the US. The true situation is that they fail to get the "action" taken they desire because the only real change people have experienced in the last 20 years is somewhat stronger airline security, hardly the basis to build a revolution on. Mostly, there are just "concerning trends", which we always have to be on guard against.
But don't take my word or anybody else's; find a reason to visit first, and ask around to get people's real experiences. For such a big step, I'd consider that a necessity.
The more this kind of crap happens, the more I hate the software industry. It's MY computer damn it. If I buy software, I should be able to use it the way I want.
Solution: Install Linux. Recognize that when you "can't do something", like play DRM'ed games, it's a feature, not a bug.
I'm not kidding. I've made the jump to Linx fulltime, and while the ever-encroaching DRM-fiasco is not the reason, it's a very good one.
You can't own your computer if you can't control the code running on it.
(And, mercifully, the idea that you have to audit every line of code yourself is a strawman, since said auditing can be done collaboratively and you don't even have to participate, whereas with commercial software you're just stuck.)
OK... but the blame lies not on the "big telcos", but reality itself. Network effects exist; better to harness them than kvetch about them. What are the big networks supposed to do, pretend they don't exist and screw their customers in the process?
So, today the Bush administration is brilliant, and they came up with a genius plan to make email go away, while appearing incompetent.
Someone be sure to send me the talking points when we're back to "The Bush administration is staffed by morons", k?
(Such amazing IQ swings we see. Genius! Moronic! Brilliant! Ape-like! Bing-bam-boom! Sometimes several flip-flops in one day! One would almost wonder if the problem lies in the observers, rather than the observed.)
"Never attribute to malice what can be adequately explained by incompetence." I think "incompetence" covers it just fine; I'm sure this is hardly the first migration screwed up this way.
What, do you think nuclear reactors are build and atomic bombs are dropped on the large, naturally occurring thorium fields that we all remember playing in as children?
Ah, how I remember passing the days on the bountiful thorium fields of my youth, before they paved them over with asphalt. How will the youth of today grow up to be responsible adults without the healthy, life-giving exposure to thorium we all used to get? Good times, good times.
(It never ceases to amaze me how rationality just goes flying out the window, even here, when any subject even remotely related to radiation comes up. I understand why, but it still amazes me.)
The problem is what to do with the heat pumped out, which could damage the arctic nearby, maybe even melt the foundation. But if the total mass cooled is small (like a few dozen microchips), that byproduct heat could be used to keep some human operators alive.
Scale, scale, scale, scale, scale. Don't let environmentalist mottos fool you; humans aren't actually "heating" the planet. By pumping large amounts of a greenhouse gas into the atmosphere, we are causing the planet to retain more heat. (This isn't news to environmentalists, which is why I explicitly point out that the mottos are wrong. Things tend to get simplified in their trip into the collective consciousness.)
We're nowhere near being able to actually heat the planet to any significant degree. Run the computations on how much energy the planet receives from the Sun every day, compare with the total energy generated by humanity in a day.
You don't need to worry about "damaging" the Antarctic by running some computers and dumping the heat out into the local environment. Heat just dissipates, and since it's at a rate proportional to the difference between the temperatures, it goes away faster the more you try. (That's the fundamental reason why we can't directly heat the planet, because even if we did, it would just radiate away.) You need to worry about the stuff that doesn't dissipate.
Comparing Guantanamo Bay to China is absurd, utterly absurd, and disproves your own point.
Remember why people object to Guantanamo Bay. It has little or nothing to do with the treatment of the prisoners; people who have actually been there, even those very critical of the facility, find no fault there. (What few allegations there are show every sign of having been trumped up by the terrorists if you actually investigate it, and that is literally out of the terrorist playbook... and by literally, I mean that documents telling terrorists to make allegations of Koran abuse if they are captured have been found.) The complaint about Guantanamo Bay is that the people being held there weren't having their human right to a fair trial upheld, they were simply held there without trial.
This is certainly a strong start to an argument that Guantanamo Bay is a place where human rights are repressed.
However, in China, everyone is treated that way and worse. Everyone. Read that link. (The whole thing. It also shows some interesting progress being slowly made. But the rules for defense would be utterly unacceptable to the same people protesting Guantanamo. Read the last sentence, too.)
To hold up Guantanamo Bay as an equivalent atrocity is to betray the very arguments being made against GB. To complain that treating a handful of active, violent enemies of a state that conduct their activities in flagrant and persistent violation of the Geneva Conventions (which as so many people conveniently forget lays responsibilities upon combatants too, not just countries) is equivalent to treating an entire country worse, not even the same but worse, is to display a moral relativism that simply staggers the mind.
If you object to Guantanamo Bay, you should be objecting to the Chinese regime a millionfold more! In China, Guantanamo Bay would be the progressive prison.
It would be neat to see them try the main arc of the first season of Stand Alone Complex, to see the world's premiere meme factory fuck up a story about an prolific, errant meme. Some sort of irony or something.
The first sci-fi miniseries needs to be understood as a filmed theatrical presentation, rather than as a movie. This explains the "pretentiousness"; I believe it was deliberately done in a theatrical style, complete with monologues and heavy symbolism (to the point of incomprehensibility if you didn't read the book).
The second miniseries did this to a lesser extent, but it is still present.
All things considered, while I believe they were forced into it by budget constraints rather than choosing it, I think it was a good choice. However, if you were expecting a modern-style movie presentation, you will be very disappointed.
If you watch it this way, it makes a lot more sense.
Will "Parallelism" be the next new thing? Well... no. That's like asking if "for loops" are going to be the next big thing. It's a tool, to be used when appropriate and not used when appropriate. It's going to be very hard to convert "Parallelism" into a magic VC pixie dust.
I say this as someone who has recently been tuning his career, experience, and personal projects towards learning more about parallel programming in practice, and I still don't see this as a "next big thing". It's just another in a long list of skillsets that is going to be gradually more useful over the next few years. I still think it's a good move to learn about it (and I'm putting my effort where my mouth is), but I don't see it sustaining any sort of bubble.
One thing I did learn from the last discussion is that a lot of people are very far behind on parallelism, though; if you think locks and pthreads are the state of the art, you've got some catching up to do. There's been a lot of practical progress in the past ten years that doesn't seem to have made it into the popular consciousness of Slashdot programmers. (And I emphasize the word practical; it's not just academic wanking anymore, or libraries optimized for such.) I suggest anyone interested in the topic check out Erlang and Software Transactional Memory (in that order). Whether these technologies will "win" I don't know, but Erlang in particular, while it may in some sense not contain a single idea that can't be traced to the 1960s, is unique in being a complete runtime and library based on ideas-that-will-be-new-to-you (most likely, based on the previous/. discussion).
Sometimes it can take a few decades (or centuries) for people to adopt something that has been rediscovered by many people in the past.
As usual, nothing new has been discovered since the 1960s.
The big difference now is that, for instance in Erlang, you can get an entire runtime built on the idea of message passing, with a library built on message passing, and many downloadable libraries also built on message passing, etc. It's not somebody's thesis and proof of concept, it's a set of deployable libraries and a small, but real, set of real-world applications.
If you do get into Erlang, or are thinking about it, you should consider downloading EJabberD and poking at the source for that. I work with that code professionally, and it took me a while to really wake up to how beautiful the code is; this massively concurrent messaging system that is the very definition of a non-embarrassingly parallel problem (all sorts of dependencies), and there isn't a lock as far as the eye can see. And it really does work in real deployments under heavy load; it's not one of those things that says it works when it really means "it ought to work in theory" or one of those open source projects that describe what they hope to accomplish as if they're already the leading library that does it... it really works. The code says what it does and it does what it says. It's really quite impressive.
Oh, and it clusters across multiple machines too. No big deal, right...?
That's why the really interesting work is being done in how to do multithreading without mutexes, semaphores, and all the other lock-based concurrency systems, which I'm surprised to see few people talking about in my skimming so far. Possibly because the old bucks aren't aware of these things, think they aren't fundamentally different, or possibly think they've been tried already.
The two most promising approaches I know of are message-passing based concurrency and Software Transactional Memory.
Message-passing based concurrency has been tried before, but to really reap the benefits it needs to be pervasive and easy. The current market leader in the field of message-based concurrency that gets things done in the real world (not just on paper) is Erlang. (I program professionally in Erlang.) I characterize it this way: Erlang doesn't make the issues of concurrency like deadlock go away, but it takes them from requiring superhuman intelligence to human intelligence.
As I like to say, a large enough quantitative change becomes a qualitative change. Yes, you can do everything in C that you do in Erlang, though you've got a lot of implementation in front of you and the resulting syntax will be awful. But by making the Right Thing easier in Erlang than the Wrong Thing, the experience is qualitatively different.
Software Transactional Memory is still in flux; the Haskell implementation has some relatively unique characteristics, but I have not yet decided how I feel about it. Erlang actually has an embryonic STM system called "Mnesia", which isn't quite the same as the hyper-pure on Haskell has, but on the other hand, we know it works. But it's something to keep an eye on; I'm experimenting with it in a personal project right now.
There are several promising lines of attack on the multicore problem, but none of them look like "semaphores and mutexes". Consensus seems to have developed that those are fundamentally flawed primitives to build software on. (Of course, message-passing is implemented with some locking, and STM can use things like that to be implemented, but saying that we are therefore building our systems on locks anyhow is exactly like claiming that since everything ends up as machine code, we should just program in machine code instead.)
We're getting there, incrementally. We can't jump to it all at once for the reasons cited by others.
If you want to see that world, be sure to support the Futurama movies, and pick up Stargate: The Ark of Truth, too. Right now, that's the most direct way to indicate interest in this.
"Thought multiplicative lighting was the way to go, rather than dealing with the performance hit of additive lighting in Quake 3."
Conclusion: Carmack sucks!
I mean, seriously, what's your point? The man's not actually a God so we shouldn't listen to him? Is there somebody more experienced I should prefer to listen to? Is "n3tcat" the handle for somebody with thirty years experience in first-person shooter engines or something?
Slashdot Mirror
Are you sure about that? Getting a paper onto arxiv.org doesn't seem to be that hard, and there's lots of ways to find out about it (RSS feed, etc.). He may not have had any reason to believe that he'd get this sort of attention, as he may have thought everyone involved would simply assume that it wasn't worth much, not having been peer reviewed.
While I love the free and open flow of information that arxiv represents, this is hardly the first time that something has been posted on there and subsequently blown out of proportion. The Internet at large doesn't seem to really understand arxiv.org, that just because someone's got a fancy LaTeX paper up claiming some wild thing doesn't mean it's credible. A paper on arxiv.org shouldn't even be understood as being endorsed by the author, let alone "science". I always love when somebody backs up their argument about physics with a link to arxiv.org, it's like a red flag that it's time to just pack it in, you're not going to get through to this person, because they only understand the trappings of science, not the actual process.
Diablo is a graphical Roguelike; in particular, it's basically a somewhat-dumbed-down*-but-much-prettier Angband. If Hack/Moria/Angband make the cut, there's no way to exclude Diablo.
(*: Not really a criticism. The Roguelikes often take advantage of their essentially textual nature to do things that simply can't be done graphically, or would be fantastically difficult. Anybody want to write the code that morphs any given monster type into any other given monster type for NethackGL? Not me!)
Who said anything about "bloat"? Bloat is a boogeyman.
Sockets solve two problems, and actually solves them, not just hacks around them. It allows payloads that aren't HTTP, in particular including small payloads of a few bytes without the massive 10,000%+ overhead of wrapping an HTTP request around it, and it allows clean, effective "push" to the browser without it being a hack.
If you can't imagine how that might be useful, I suggest an imagination upgrade. The reason why you currently conceptualize these basic, 1960s-computing capabilities as "overkill" is because you, and apparently everybody else, have so deeply internalized the idea that everything must be HTTP requests with all the associated overhead that you can't even imagine a world without them. But there's hardly a top-end web app that wouldn't benefit from one or both of these features, and by "benefit" I mean real and concrete benefits to the user, with faster, more capable, more responsive interfaces that at the same time put less of a load on the server.
Sockets. Raw sockets. Stop pretending with AJAX, with Comet, and just cut to the chase. Why this isn't the first thing on the AJAX agenda beats me.
If you are sophisticated enough to load pirated ROMs onto a card for your DS, then you are sophisticated enough to read reviews and figure out whether something is "shovelware" before you buy it. Since the Internet got big, I have not yet bought a true dud of a game. I've taken a couple risks, not all of which panned out, but I knew what I was getting into.
(Yes, you can't just take the score for granted, either, but I also assume that if you can load software onto your DS, you are sophisticated enough to see through that, too. I play a lot of games that only get an 8/10, docked a point or two for "niche appeal", and wouldn't care to play many 9.5/10 games. Again, I don't have trouble with that.)
No excuses.
"Useful" isn't an adequate summary of my point, because you have to define "useful for what". And the very reason I am critical of trig identities is that by the most relevant definition I can come up with, "useful for mathematical education, understanding, and future mathematical study", trig identities aren't useful.
No shit, sherlock; the different spelling should be a dead giveaway. But what population is going to be the biggest users of optics? Hint: There aren't that many telescope designers in the world.
Here, you've provided me an excellent demonstration of why we need game theory in high school, because your post neglects the vitally important concept of opportunity cost, something that I'd much rather the general populace had exposure to than something as useless as trig identities. The opportunity cost of teaching trig identities when you could be teaching, say, opportunity costs, is way too high. Trig is not even close to the best thing we could be spending our time on.
Besides, optometrists and surveyors are invited to take specialized courses in trig identities, just as the mere fact that I took a course on the mathematics of evolutionary computation doesn't even begin to imply that everybody in high school should learn about that stuff. Time is finite. Opportunity costs are important. Trig identities are too expensive and displacing a lot of stuff that is both useful in real life and more useful to mathematicians, who, like I said, don't consider them important.
(Actually, the disconnect between real mathematicians and mathematical education is truly staggering once you fully understand it. The educational community, and I say this with full consideration to the people involve, wouldn't know math if it bit them on the ass.)
- Trigonometric identities go bye-bye. Even real mathematicians consider them little more than curiosities. Bring them out when you do Taylor expansions, put them away again when you're done.
- Game theory is in. I'd happily trade Calculus for Game Theory for "non-Engineering bound students". Game theory is fantastically useful, even if you don't (or can't) actually "compute" with it in the real world, the concepts serve you in economics, politics (how many maths can make that claim with a straight face?), and business.
That latter one I particular wish I could get in.If the consultant really expected you to email the data, and expressed even a modicum of surprise that you wouldn't do it, they've already disqualified themselves from being able to securely handle your data.
Do you really think that this is the only flaw in their handling of sensitive data? That, otherwise, they are security conscious and careful, except for this odd flaw where they don't understand how insecure email is?
If you care, it's time to change consultants.
If you don't care, just email it already.
(I'm actually not quite as rigid as this may sound out-of-context. I don't agree that security is all-or-nothing, so please don't strawman me that way. My second paragraph is important; anyone who expects those things emailed to them is so far away from the necessary knowledge and skills that debating whether they are close enough or whether they will be able to take reasonable care is a waste of time, arguing about whether the receiver made a touchdown when they got tackled on the 10 yard line on the wrong side of the field.)
But don't take my word or anybody else's; find a reason to visit first, and ask around to get people's real experiences. For such a big step, I'd consider that a necessity.
I'm not kidding. I've made the jump to Linx fulltime, and while the ever-encroaching DRM-fiasco is not the reason, it's a very good one.
You can't own your computer if you can't control the code running on it.
(And, mercifully, the idea that you have to audit every line of code yourself is a strawman, since said auditing can be done collaboratively and you don't even have to participate, whereas with commercial software you're just stuck.)
OK... but the blame lies not on the "big telcos", but reality itself. Network effects exist; better to harness them than kvetch about them. What are the big networks supposed to do, pretend they don't exist and screw their customers in the process?
So, today the Bush administration is brilliant, and they came up with a genius plan to make email go away, while appearing incompetent.
Someone be sure to send me the talking points when we're back to "The Bush administration is staffed by morons", k?
(Such amazing IQ swings we see. Genius! Moronic! Brilliant! Ape-like! Bing-bam-boom! Sometimes several flip-flops in one day! One would almost wonder if the problem lies in the observers, rather than the observed.)
"Never attribute to malice what can be adequately explained by incompetence." I think "incompetence" covers it just fine; I'm sure this is hardly the first migration screwed up this way.
What, do you think nuclear reactors are build and atomic bombs are dropped on the large, naturally occurring thorium fields that we all remember playing in as children?
Ah, how I remember passing the days on the bountiful thorium fields of my youth, before they paved them over with asphalt. How will the youth of today grow up to be responsible adults without the healthy, life-giving exposure to thorium we all used to get? Good times, good times.
(It never ceases to amaze me how rationality just goes flying out the window, even here, when any subject even remotely related to radiation comes up. I understand why, but it still amazes me.)
We're nowhere near being able to actually heat the planet to any significant degree. Run the computations on how much energy the planet receives from the Sun every day, compare with the total energy generated by humanity in a day.
You don't need to worry about "damaging" the Antarctic by running some computers and dumping the heat out into the local environment. Heat just dissipates, and since it's at a rate proportional to the difference between the temperatures, it goes away faster the more you try. (That's the fundamental reason why we can't directly heat the planet, because even if we did, it would just radiate away.) You need to worry about the stuff that doesn't dissipate.
Comparing Guantanamo Bay to China is absurd, utterly absurd, and disproves your own point.
Remember why people object to Guantanamo Bay. It has little or nothing to do with the treatment of the prisoners; people who have actually been there, even those very critical of the facility, find no fault there. (What few allegations there are show every sign of having been trumped up by the terrorists if you actually investigate it, and that is literally out of the terrorist playbook... and by literally, I mean that documents telling terrorists to make allegations of Koran abuse if they are captured have been found.) The complaint about Guantanamo Bay is that the people being held there weren't having their human right to a fair trial upheld, they were simply held there without trial.
This is certainly a strong start to an argument that Guantanamo Bay is a place where human rights are repressed.
However, in China, everyone is treated that way and worse . Everyone. Read that link. (The whole thing. It also shows some interesting progress being slowly made. But the rules for defense would be utterly unacceptable to the same people protesting Guantanamo. Read the last sentence, too.)
To hold up Guantanamo Bay as an equivalent atrocity is to betray the very arguments being made against GB. To complain that treating a handful of active, violent enemies of a state that conduct their activities in flagrant and persistent violation of the Geneva Conventions (which as so many people conveniently forget lays responsibilities upon combatants too, not just countries) is equivalent to treating an entire country worse, not even the same but worse, is to display a moral relativism that simply staggers the mind.
If you object to Guantanamo Bay, you should be objecting to the Chinese regime a millionfold more! In China, Guantanamo Bay would be the progressive prison.
It would be neat to see them try the main arc of the first season of Stand Alone Complex, to see the world's premiere meme factory fuck up a story about an prolific, errant meme. Some sort of irony or something.
The first sci-fi miniseries needs to be understood as a filmed theatrical presentation, rather than as a movie. This explains the "pretentiousness"; I believe it was deliberately done in a theatrical style, complete with monologues and heavy symbolism (to the point of incomprehensibility if you didn't read the book).
The second miniseries did this to a lesser extent, but it is still present.
All things considered, while I believe they were forced into it by budget constraints rather than choosing it, I think it was a good choice. However, if you were expecting a modern-style movie presentation, you will be very disappointed.
If you watch it this way, it makes a lot more sense.
Will "Parallelism" be the next new thing? Well... no. That's like asking if "for loops" are going to be the next big thing. It's a tool, to be used when appropriate and not used when appropriate. It's going to be very hard to convert "Parallelism" into a magic VC pixie dust.
/. discussion).
I say this as someone who has recently been tuning his career, experience, and personal projects towards learning more about parallel programming in practice, and I still don't see this as a "next big thing". It's just another in a long list of skillsets that is going to be gradually more useful over the next few years. I still think it's a good move to learn about it (and I'm putting my effort where my mouth is), but I don't see it sustaining any sort of bubble.
One thing I did learn from the last discussion is that a lot of people are very far behind on parallelism, though; if you think locks and pthreads are the state of the art, you've got some catching up to do. There's been a lot of practical progress in the past ten years that doesn't seem to have made it into the popular consciousness of Slashdot programmers. (And I emphasize the word practical; it's not just academic wanking anymore, or libraries optimized for such.) I suggest anyone interested in the topic check out Erlang and Software Transactional Memory (in that order). Whether these technologies will "win" I don't know, but Erlang in particular, while it may in some sense not contain a single idea that can't be traced to the 1960s, is unique in being a complete runtime and library based on ideas-that-will-be-new-to-you (most likely, based on the previous
The big difference now is that, for instance in Erlang, you can get an entire runtime built on the idea of message passing, with a library built on message passing, and many downloadable libraries also built on message passing, etc. It's not somebody's thesis and proof of concept, it's a set of deployable libraries and a small, but real, set of real-world applications.
If you do get into Erlang, or are thinking about it, you should consider downloading EJabberD and poking at the source for that. I work with that code professionally, and it took me a while to really wake up to how beautiful the code is; this massively concurrent messaging system that is the very definition of a non-embarrassingly parallel problem (all sorts of dependencies), and there isn't a lock as far as the eye can see. And it really does work in real deployments under heavy load; it's not one of those things that says it works when it really means "it ought to work in theory" or one of those open source projects that describe what they hope to accomplish as if they're already the leading library that does it... it really works. The code says what it does and it does what it says. It's really quite impressive.
Oh, and it clusters across multiple machines too. No big deal, right...?
That's why the really interesting work is being done in how to do multithreading without mutexes, semaphores, and all the other lock-based concurrency systems, which I'm surprised to see few people talking about in my skimming so far. Possibly because the old bucks aren't aware of these things, think they aren't fundamentally different, or possibly think they've been tried already.
The two most promising approaches I know of are message-passing based concurrency and Software Transactional Memory.
Message-passing based concurrency has been tried before, but to really reap the benefits it needs to be pervasive and easy. The current market leader in the field of message-based concurrency that gets things done in the real world (not just on paper) is Erlang. (I program professionally in Erlang.) I characterize it this way: Erlang doesn't make the issues of concurrency like deadlock go away, but it takes them from requiring superhuman intelligence to human intelligence.
As I like to say, a large enough quantitative change becomes a qualitative change. Yes, you can do everything in C that you do in Erlang, though you've got a lot of implementation in front of you and the resulting syntax will be awful. But by making the Right Thing easier in Erlang than the Wrong Thing, the experience is qualitatively different.
Software Transactional Memory is still in flux; the Haskell implementation has some relatively unique characteristics, but I have not yet decided how I feel about it. Erlang actually has an embryonic STM system called "Mnesia", which isn't quite the same as the hyper-pure on Haskell has, but on the other hand, we know it works. But it's something to keep an eye on; I'm experimenting with it in a personal project right now.
There are several promising lines of attack on the multicore problem, but none of them look like "semaphores and mutexes". Consensus seems to have developed that those are fundamentally flawed primitives to build software on. (Of course, message-passing is implemented with some locking, and STM can use things like that to be implemented, but saying that we are therefore building our systems on locks anyhow is exactly like claiming that since everything ends up as machine code, we should just program in machine code instead.)
We're getting there, incrementally. We can't jump to it all at once for the reasons cited by others.
If you want to see that world, be sure to support the Futurama movies, and pick up Stargate: The Ark of Truth, too. Right now, that's the most direct way to indicate interest in this.
Love Psychedelico, Help!
:)
Sung in pretty decent English, as Japanese singers go. It's not a flawless American accent... but then, neither is the original, no?
Plusses:
- One of the primary fathers of the FPS genre.
- Wolfenstien 3D
- Doom
- Quake 1
- Quake 2
- Quake 3
- Endless articles and commentary on the field
- A shitload of stuff I'm forgetting
Minusses:- "Thought multiplicative lighting was the way to go, rather than dealing with the performance hit of additive lighting in Quake 3."
Conclusion: Carmack sucks!I mean, seriously, what's your point? The man's not actually a God so we shouldn't listen to him? Is there somebody more experienced I should prefer to listen to? Is "n3tcat" the handle for somebody with thirty years experience in first-person shooter engines or something?