He already commented on where he thinks game graphics are going at this year's GDC -- in short, I believe he's saying that 10 years more improvement should get us technologically there, but that there are severe problems with how we create the more detailed and larger quantity of content necessary cheaply enough.
I think the big problem with swapping is how it interferes with interaction. So the problem is when something gets swapped out (or possibly paged in the first time, if the OS is lazily loading the executable) that's the front-line of interaction. Say, the data needed for showing an infrequently used menu.
We either need some way of expressing this in code in a way that's exposed to the OS so it can avoid paging it out (and don't say 'well, just lock the pages containing it', think about how GUIs code & data are arranged--i.e. OO), or, perhaps instead of LRU, paging should pay attention to when a page was last used relative to an interaction--if you use some menu to bring up a dialog that triggers a long, memory-intensive process, that menu may have been used longer-ago than stuff happening as a result, but it's going to be used again first--LRU is the wrong model. (Also consider the interactive parts of other apps that you have open windows for.) Maybe it should still get swapped out, but swapped back in when some memory is freed up--'this thing was used very soon after an interaction, so it should be in memory if possible'. I don't know if any OSes attempt to page-in from swap before something is requested. Aren't some processors these days trying to prefetch memory requests based on patterns of memory access (not just prefetching code memory)? Same sort of idea.
Tuning for interaction isn't new to OSes; the VMS operating system's process scheduler treated 'interactive' apps differently from 'batch' apps (where, if I remember correctly, an app was interactive if it paused for I/O before using up its timeslice). I dunno if Unixen or Windows do things like that.
Yes. Adjustment layers allow you to tweak an image 'under' the adjustment. And they allow non-destructive editing. Very powerful. For example, in this image I used a hue/saturation adjustment layer to desaturate the cardboard photo, allowing me to continually refine it for the exact look I wanted. Moreover, I could now drop any other photograph in, and automatically get the same cardboard-multiply followed by desaturation.
In combination with layer sets, adjustment layers are superpowerful. And it only took me a year of infrequent use of Photoshop before I discovered them! Sadly, at least in PS 6, there are no filter layers for applying filters non-destructively. I'd kill for those. I guess they'd be slow, though.
I'm Agnostic because it is the only intellectually honest thing that I can be.
It's all well and good to say 'science is all about making models and we never know if those models are true, so really we don't ever know anything', but saying 'we don't know anything' (being 'agnostic about everything') is essentially just a semantic game, in which you've defined the word 'belief' so strongly that you believe nothing, basically taking yourself out of the game since that's not how other people mean 'belief'.
Now, maybe you're not agnostic about everything, you're just saying it about god--hey, maybe god exists, I can't disprove it. But then, are you also agnostic about unicorns, fairys, leprechauns, and ghosts? If not, why not?
For me, atheism is the only intellectually honest thing I can be (although I admit it took me a few years as an agnostic to complete my theism-deprogramming). I have no reason to believe a god exists any more than I do to believe elves exist--despite plenty of literary and oral tradition for both. Occam's razor says chop chop, and so, goodbye god. (I usually summarize this argument as follows: I find it kind of surprising that theism is such the default that I have to name myself as a-theistic to describe my lack of belief in something there's no evidence for. After all, we don't need a word acornic to describe people who don't believe in unicorns.)
Of course, you can make the move that 'some sort of prime mover that built the universe and triggered the big bang exists', and call that god, even if that thing isn't omniscient, omnipotent, or anything like any of the gods traditionally worshipped by people. But then you've basically gone and redefined god, and what's the point. An example in that vein: unlike fairys and elves, there's pretty clear evidence that the Santa Claus of familiar tales doesn't exist--at least, he's never left me any presents or come down my chimney, and I assume you share that experience. But maybe Santa Claus really DOES exist--he just doesn't actually leave presents or eat milk and cookies left for people. And he doesn't live at the North Pole literally, where he would have been spotted; he lives in an underground base in Antarctica, in fact. But if you believed that, why is this guy anything to do with traditional "Santa Claus", other than maybe happening to have that name?
I am independent game developer burned out on the mainstream industry, and not that thrilled with the web downloadable publishers who are turning out much the same as the mainstream publishers, writ small. But there was plenty of creativity on display at the GDC if you looked for it.
At the IGDA awards, three games were given "Game Innovation Spotlights": the EyeToy, Viewtiful Joe, and WarioWare Inc. All three of these seem quite novel and worthy of the attention.
At the Experimental Gameplay Workshop, both indies and mainstream games were shown. On the indie front, this year's Indie Game Jam games (full disclosure: I co-run this event); Yohoho! Puzzle Pirates; and Zoesis' The Demon and the Princess. On the commercial front, the creator of Namco's Katamari Damashii spoke about and demoed the game ("Was it difficult to convince Namco to let you do this game?" "Of course." was even funnier with the long pause for translation between question and answer); we had presentations about WarioWare and about the explorations of time as a game mechanic (specifically in Prince of Persia, Max Payne 1 & 2, and Viewtiful Joe).
(There were a few more presentations about more academic "games": Ken Perlin's work on natural-language-programming for kids, "Haptic Battle Pong", and I forget what else, as I was developing a fever during the 3-hour EGW.)
The winner of the Indie Games Festival's web downloadable grand prize, Oasis, is a fairly original and creative game (full disclosure: I did contract work for Oasis' developers on a different project), and since this is announced at essentially the same ceremony as the IGDA awards it has a fairly significant cachet.
So I think the Reuters reporters just didn't go to the right events at the GDC.
The story itself has plenty of debatable claims. Are gamers, as the article claims, getting more conservative, or are publishers just getting extremely conservative and releasing more sequels and focusing their marketing dollars there? Hint: nobody debates the truth of the latter.
The thinking, if I am not mistaken, is based on the idea that a word in published print has gone through an editing process. The editor is then responsible for making sure that the words used in the final publication are valid and used accurately.
versus
...without the Star-Trek script that illustrates the use of the term "cloaking device", they cannot verify it and date it properly
A Star Trek script is neither a published work of literature nor something that has been vetted editorially for spelling or usage (it is generally edited for 'content' with an eye towards how it will be used).
If the Star Trek script is relevant, then a videotape should be even better. (Suppose the script said 'cloaker device' and on the set they decided 'cloaking device' sounded better? Who cares what the script said.)
This is actually an important question, because Usenet articles are actually published more than a Star Trek script.
Or maybe people don't use Ocaml due to flaws in the language and flaws in the implementation.
Here's an example of each:
- (language) lame support for imperative programming -- no equivalent to 'break' for loops, and even no equivalent to 'return' to allow you to return a value from the middle of a loop. Of course, imperative programming isn't the emphasis of OCaml, but it means that that "benefit" of having imperative features available when you need them isn't really quite as strong as you might like
- (implementation) useless compiler errors -- numerous mistakes will give you the unadorned response "Syntax error", and because the language is so lacking in redundancy, mistakes can be syntactically valid for a long way, causing the syntax error to show up 20 lines later; similarly, typecheck errors themselves can be hard to decipher, since the compiler doesn't show you where the inferred types are coming from, leaving you to track them down on your own
I base this opinion on having used Ocaml twice, working with several other programmers on IFCP entries, and from occasionaly pair programming with my officemate, who is probably the only game developer in the world using Ocaml.
Try the grand-daddy of first-person PC games, Ultima Underworld. You interactively swung your sword for every blow.
It was released a few months before Wolfenstein 3D, and is widely credited for spawning the first-person CRPG.
The lag problem is in no way unique to MMOGs; the description you give for Planetside is exactly how Valve has described Half-Life's multiplayer working.
Warren Spector did not work on Terra Nova in any capacity.
And, in fact, other than DX 1, his contributions to many games he's worked on have been much more on the management side of things, not the development side of things, so I wouldn't over-attribute the things you like in "his" games to him. (Half Life may have been influenced by System Shock, but influenced by Warren Spector's work on System Shock?!)
Prior Art is NOT the biggest problem, because prior art can be addressed in court. (It sucks, and costs money, but it can be addressed.)
Obviousness is the big problem. Inventions are supposed to novel (no prior art) and non-obvious. The granting of patents on obvious inventions is the big problem. As far as I can tell, this can't be addressed in court. (I guess because you bring in your expert witness to say it's obvious; they bring in theirs to say it's not. I don't actually know; I've just never heard of it happening.) Upon occasion, you do hear about the patent office deciding to reexamine a patent, but it's pretty rare.
95% of the time I read on Slashdot about something being patented and people run around starting to post "wasn't this prior art", the thing being patented is the obvious solution to the problem. Half the time, the patent is really just on the problem: "we have patented the idea of solving this problem", e.g. displaying if the other person in an IM session is actively typing.
(Well, and then the other big problem is the profit motive; those in favor of software patents--the PTO, patent lawyers, and current patent holders--all profit from them, even if the public doesn't.)
Access to fast machines probably made a big difference in the results, and this was a major change from past ICFP competitions (which were run by the judges under fixed time constraints).
The winner noted:
I used 17 Dual P4 1800Mhz computers. (Without the permission of the CS Dept... sorry guys no time to get permission..)
He also probably wrote the least amount of code of anyone. (A link to his source is posted in the same forum.)
To be fair, he also took an approach that my team wrote off as unachievable when it obviously wasn't unachievable--brute force search of the solution space. (Specfically, brute force search of the "whole" solution space reduced by restricting solutons to only change control settings on pixel crossings [and certain quantized speed changes].)
Meanwhile my team's solver got beat by some people who just interactively solved the courses with a human driver. Embarassing.
Here's a much clearer argument why the Shuttle is a bad idea, which actually gets into the history of the trade-offs (now largely irrelevant) underlying its design: Homer Hickam's op-ed.
Take a look at the Shuttle stack and what do you see? A fragile spaceplane sitting on the back of a huge propellant tank between two massive solid rocket boosters. The tank holds liquid oxygen and hydrogen and towers above the spaceplane. It is the foam off this tank that hit Columbia and knocked a hole in her wing. But why is there foam at all? Because without it, ice would form on the super-cooled tank and hit the spaceplane. But why would ice or foam hit it in the first place? Because of where the spaceplane sits. But why does it sit there? Because the Shuttle Main Engines (SME's) need to come back to Earth and therefore must be attached to the spaceplane to be returned. And why do the SME's need to be returned? So that they can be reused. And why do they have to be reused? Because, theoretically, it's cheaper to refurbish them than build new ones. Therefore, the spaceplane we think of as the Shuttle has to sit right in the middle of all the turmoil of launch because we once believed it would be cheaper to bring back those engines and rebuild them than to build new ones. That has not proved to be the case-far from it-but it has left us with a crew sitting in the most vulnerable position possible in terms of engineering design and safety.
Yeah, the quote is entirely bogus for its use of fun. Clearly what she is advocating is to bring games more in line with syndicated newspaper comic strips.
Yay.
Like many online cartoonists, many game developers (or at least many of the best) approach game development as a passion, not as a money-maximizing endeavor. Yes, money must be made to stay in business, but there's no need to target the utter most common denominator just to make enough money to be happen.
Publishers, on the other hand, just care about the bottom line. Especially the publisher this "leading game expert" works for.
Technology improvements leading to, for example, more realistic 3d graphics, drive budgets up because the costs of creating that content go up.
We can draw 1000X as many polygons as some time ago; somebody has to make those 1000X polygons. Yes, we can get some of them by building objects with NURBS; we can get some by instancing. But NURBS have a lot more editable parameters than the same number of polygons, and they only go so far--curvature is not detail, which is what people expect to see. Similarly, instancing (replicating the same object multiple times) is fairly obvious when you see it if it's overused. So we spend more artist time building out geometry.
Generally, it's not the programming costs that have increased--it's not the cost of the technology itself that's increased; it's the cost of using the technology.
And rather than replying to one of the several film industry comparisons elsewhere, I'll comment here:
It's true that games have become a big budget hits-driven industry; this was already well on its way five years ago. But there's a big difference between moviemaking and gamemaking. The big-budget movies and the low-budget movies still make their content the same way: they stick some human beings in front of a camera. Compare Robert Rodriguez' El Mariachi to Desperado visually--yeah, they have a different quality of acting, different quality of production value, but they are in many ways incredibly similar: both movies show the exact same sort of thing, despite one costing $7,000,000 and one costing $7,000.
The difference between big budget games and low budget games is radically different. If you attempted to recreate Doom 3 on 1/1000th the budget, using, say, the Quake 3 engine, you'd probably get laughably hideous results, or 1/10th of the total content. (El Mariachi is much longer than 1/10th the length of Desperado.)
Even PopCap-style games have professional art--a lot less, yes, but a lone programmer/game-designer can't get the same quality of art, and few artists are willing to do that work for free, for the experience, for the exposure, the way actors are willing to do for micro-budget movies.
You couldn't figure this out from the all-caps boldfacing and suchlike in the post itself? OUR AWESOME NEW LANGUAGE SOLVES ALL PROBLEMS SO YOU DON'T HAVE TO!
About 90% of the time that a story pisses me off with its irrelevance, poor summary, or just plain-wrongness, it was posted by timothy. Thus, it's not so much slashdot trustworthiness as timothy trustworthiness at issue.
Your mileage may vary; try keepin your own stats.
Sadly, you know just enough to shoot yourself in the foot.
As swanton noted, 1/4 times 1/4 does not equal 1/8. That's why your 4.1% is so much below your 12.5%--it should have been 6.25% in the first place.
If you keep a pair of twos and an 8, you might draw an 8 and get two pair. But if you just kept the pair of twos, you might draw a 9 first, and then there's a chance you'll draw another 9 in the next two cards. I know you're not looking more than one card ahead, but this is a place where that's a mistake; your odds of getting two pair are almost identical whether you keep the 8 or not. I think it is actually slightly better to keep a card you already have, since there are three more of those out there, whereas if you just keep the pair, the first card you draw might be another 8, and now there are only two left to make further pairs with; but this effect is much smaller than the effect you describe, and I'm still oversimplifying it.
The chance of drawing the 2^ and 3^ are not 1/52 * 1/52; the chance of drawing the 2^ then the 3^ is 1/52 * 1/51, and the chance of drawing the 3^ then the 2^ is 1/52 * 1/51. To put it unambiguously, the chance of drawing either the 2^ or 3^ is 1/26, and the chance of drawing the remaining is 1/51 (overall odds is 1/1326).
Since lawyers (or their associations and what not) have come out explicitly pro- software patents, I see no reason not to blame them. (There was a slashdot story a while back with their response to Bezos' suggestion that there needed to be patent reform, dissing the idea.)
When it comes to software patents, just follow the money. There have always been three vocal groups in the US in favor of software patents: the patent office, (patent) lawyers, and software patent holders: exactly the people who profit if software patents exist. Those people may argue that software patents are "for the public good", but nobody else does, and they all profit directly from there being software patents.
Slashdot Mirror
He already commented on where he thinks game graphics are going at this year's GDC -- in short, I believe he's saying that 10 years more improvement should get us technologically there, but that there are severe problems with how we create the more detailed and larger quantity of content necessary cheaply enough.
I say this as a liberal.
We either need some way of expressing this in code in a way that's exposed to the OS so it can avoid paging it out (and don't say 'well, just lock the pages containing it', think about how GUIs code & data are arranged--i.e. OO), or, perhaps instead of LRU, paging should pay attention to when a page was last used relative to an interaction--if you use some menu to bring up a dialog that triggers a long, memory-intensive process, that menu may have been used longer-ago than stuff happening as a result, but it's going to be used again first--LRU is the wrong model. (Also consider the interactive parts of other apps that you have open windows for.) Maybe it should still get swapped out, but swapped back in when some memory is freed up--'this thing was used very soon after an interaction, so it should be in memory if possible'. I don't know if any OSes attempt to page-in from swap before something is requested. Aren't some processors these days trying to prefetch memory requests based on patterns of memory access (not just prefetching code memory)? Same sort of idea.
Tuning for interaction isn't new to OSes; the VMS operating system's process scheduler treated 'interactive' apps differently from 'batch' apps (where, if I remember correctly, an app was interactive if it paused for I/O before using up its timeslice). I dunno if Unixen or Windows do things like that.
In combination with layer sets, adjustment layers are superpowerful. And it only took me a year of infrequent use of Photoshop before I discovered them! Sadly, at least in PS 6, there are no filter layers for applying filters non-destructively. I'd kill for those. I guess they'd be slow, though.
I don't know about MS, but why in the world would IBM do that? "IBM raked in $1.6 billion in intellectual property license fees" in 2000 (value of patents vs. copyrights unknown).
Of course, "raked in" might be overstating it a tad when they had gross revenues of $88 billion that year, but I bet it's basically sheer profit.
It's all well and good to say 'science is all about making models and we never know if those models are true, so really we don't ever know anything', but saying 'we don't know anything' (being 'agnostic about everything') is essentially just a semantic game, in which you've defined the word 'belief' so strongly that you believe nothing, basically taking yourself out of the game since that's not how other people mean 'belief'.
Now, maybe you're not agnostic about everything, you're just saying it about god--hey, maybe god exists, I can't disprove it. But then, are you also agnostic about unicorns, fairys, leprechauns, and ghosts? If not, why not?
For me, atheism is the only intellectually honest thing I can be (although I admit it took me a few years as an agnostic to complete my theism-deprogramming). I have no reason to believe a god exists any more than I do to believe elves exist--despite plenty of literary and oral tradition for both. Occam's razor says chop chop, and so, goodbye god. (I usually summarize this argument as follows: I find it kind of surprising that theism is such the default that I have to name myself as a-theistic to describe my lack of belief in something there's no evidence for. After all, we don't need a word acornic to describe people who don't believe in unicorns.)
Of course, you can make the move that 'some sort of prime mover that built the universe and triggered the big bang exists', and call that god, even if that thing isn't omniscient, omnipotent, or anything like any of the gods traditionally worshipped by people. But then you've basically gone and redefined god, and what's the point. An example in that vein: unlike fairys and elves, there's pretty clear evidence that the Santa Claus of familiar tales doesn't exist--at least, he's never left me any presents or come down my chimney, and I assume you share that experience. But maybe Santa Claus really DOES exist--he just doesn't actually leave presents or eat milk and cookies left for people. And he doesn't live at the North Pole literally, where he would have been spotted; he lives in an underground base in Antarctica, in fact. But if you believed that, why is this guy anything to do with traditional "Santa Claus", other than maybe happening to have that name?
At the IGDA awards, three games were given "Game Innovation Spotlights": the EyeToy, Viewtiful Joe, and WarioWare Inc. All three of these seem quite novel and worthy of the attention.
At the Experimental Gameplay Workshop, both indies and mainstream games were shown. On the indie front, this year's Indie Game Jam games (full disclosure: I co-run this event); Yohoho! Puzzle Pirates; and Zoesis' The Demon and the Princess. On the commercial front, the creator of Namco's Katamari Damashii spoke about and demoed the game ("Was it difficult to convince Namco to let you do this game?" "Of course." was even funnier with the long pause for translation between question and answer); we had presentations about WarioWare and about the explorations of time as a game mechanic (specifically in Prince of Persia, Max Payne 1 & 2, and Viewtiful Joe).
(There were a few more presentations about more academic "games": Ken Perlin's work on natural-language-programming for kids, "Haptic Battle Pong", and I forget what else, as I was developing a fever during the 3-hour EGW.)
The winner of the Indie Games Festival's web downloadable grand prize, Oasis, is a fairly original and creative game (full disclosure: I did contract work for Oasis' developers on a different project), and since this is announced at essentially the same ceremony as the IGDA awards it has a fairly significant cachet.
So I think the Reuters reporters just didn't go to the right events at the GDC.
The story itself has plenty of debatable claims. Are gamers, as the article claims, getting more conservative, or are publishers just getting extremely conservative and releasing more sequels and focusing their marketing dollars there? Hint: nobody debates the truth of the latter.
versus
A Star Trek script is neither a published work of literature nor something that has been vetted editorially for spelling or usage (it is generally edited for 'content' with an eye towards how it will be used).
If the Star Trek script is relevant, then a videotape should be even better. (Suppose the script said 'cloaker device' and on the set they decided 'cloaking device' sounded better? Who cares what the script said.)
This is actually an important question, because Usenet articles are actually published more than a Star Trek script.
Fast TCP To Increase Speed of File Transfers? and 8.6 GB Internet?
Here's an example of each:
- (language) lame support for imperative programming -- no equivalent to 'break' for loops, and even no equivalent to 'return' to allow you to return a value from the middle of a loop. Of course, imperative programming isn't the emphasis of OCaml, but it means that that "benefit" of having imperative features available when you need them isn't really quite as strong as you might like
- (implementation) useless compiler errors -- numerous mistakes will give you the unadorned response "Syntax error", and because the language is so lacking in redundancy, mistakes can be syntactically valid for a long way, causing the syntax error to show up 20 lines later; similarly, typecheck errors themselves can be hard to decipher, since the compiler doesn't show you where the inferred types are coming from, leaving you to track them down on your own
I base this opinion on having used Ocaml twice, working with several other programmers on IFCP entries, and from occasionaly pair programming with my officemate, who is probably the only game developer in the world using Ocaml.
Try the grand-daddy of first-person PC games, Ultima Underworld. You interactively swung your sword for every blow.
It was released a few months before Wolfenstein 3D, and is widely credited for spawning the first-person CRPG.
The lag problem is in no way unique to MMOGs; the description you give for Planetside is exactly how Valve has described Half-Life's multiplayer working.
And, in fact, other than DX 1, his contributions to many games he's worked on have been much more on the management side of things, not the development side of things, so I wouldn't over-attribute the things you like in "his" games to him. (Half Life may have been influenced by System Shock, but influenced by Warren Spector's work on System Shock?!)
To quote a very old news item from oldmanmurray.com:
The first sentence of a recent Deus Ex 2 preview on pc.ign.com:
(Full disclosure: I worked on Thief and Terra Nova.)
Prior Art is NOT the biggest problem, because prior art can be addressed in court. (It sucks, and costs money, but it can be addressed.)
Obviousness is the big problem. Inventions are supposed to novel (no prior art) and non-obvious. The granting of patents on obvious inventions is the big problem. As far as I can tell, this can't be addressed in court. (I guess because you bring in your expert witness to say it's obvious; they bring in theirs to say it's not. I don't actually know; I've just never heard of it happening.) Upon occasion, you do hear about the patent office deciding to reexamine a patent, but it's pretty rare.
95% of the time I read on Slashdot about something being patented and people run around starting to post "wasn't this prior art", the thing being patented is the obvious solution to the problem. Half the time, the patent is really just on the problem: "we have patented the idea of solving this problem", e.g. displaying if the other person in an IM session is actively typing.
(Well, and then the other big problem is the profit motive; those in favor of software patents--the PTO, patent lawyers, and current patent holders--all profit from them, even if the public doesn't.)
I expect the software to last longer than the hardware. They're just bits.
I don't actually have anything before TNT2 on here, but I think it's still interesting.
The winner noted:
(here)He also probably wrote the least amount of code of anyone. (A link to his source is posted in the same forum.)
To be fair, he also took an approach that my team wrote off as unachievable when it obviously wasn't unachievable--brute force search of the solution space. (Specfically, brute force search of the "whole" solution space reduced by restricting solutons to only change control settings on pixel crossings [and certain quantized speed changes].)
Meanwhile my team's solver got beat by some people who just interactively solved the courses with a human driver. Embarassing.
Yay.
Like many online cartoonists, many game developers (or at least many of the best) approach game development as a passion, not as a money-maximizing endeavor. Yes, money must be made to stay in business, but there's no need to target the utter most common denominator just to make enough money to be happen.
Publishers, on the other hand, just care about the bottom line. Especially the publisher this "leading game expert" works for.
We can draw 1000X as many polygons as some time ago; somebody has to make those 1000X polygons. Yes, we can get some of them by building objects with NURBS; we can get some by instancing. But NURBS have a lot more editable parameters than the same number of polygons, and they only go so far--curvature is not detail, which is what people expect to see. Similarly, instancing (replicating the same object multiple times) is fairly obvious when you see it if it's overused. So we spend more artist time building out geometry.
Generally, it's not the programming costs that have increased--it's not the cost of the technology itself that's increased; it's the cost of using the technology.
And rather than replying to one of the several film industry comparisons elsewhere, I'll comment here:
It's true that games have become a big budget hits-driven industry; this was already well on its way five years ago. But there's a big difference between moviemaking and gamemaking. The big-budget movies and the low-budget movies still make their content the same way: they stick some human beings in front of a camera. Compare Robert Rodriguez' El Mariachi to Desperado visually--yeah, they have a different quality of acting, different quality of production value, but they are in many ways incredibly similar: both movies show the exact same sort of thing, despite one costing $7,000,000 and one costing $7,000.
The difference between big budget games and low budget games is radically different. If you attempted to recreate Doom 3 on 1/1000th the budget, using, say, the Quake 3 engine, you'd probably get laughably hideous results, or 1/10th of the total content. (El Mariachi is much longer than 1/10th the length of Desperado.)
Even PopCap-style games have professional art--a lot less, yes, but a lone programmer/game-designer can't get the same quality of art, and few artists are willing to do that work for free, for the experience, for the exposure, the way actors are willing to do for micro-budget movies.
You couldn't figure this out from the all-caps boldfacing and suchlike in the post itself? OUR AWESOME NEW LANGUAGE SOLVES ALL PROBLEMS SO YOU DON'T HAVE TO!
The linked-to page is missing a closing tag, and so does not render in certain old browsers.
About 90% of the time that a story pisses me off with its irrelevance, poor summary, or just plain-wrongness, it was posted by timothy. Thus, it's not so much slashdot trustworthiness as timothy trustworthiness at issue. Your mileage may vary; try keepin your own stats.
As swanton noted, 1/4 times 1/4 does not equal 1/8. That's why your 4.1% is so much below your 12.5%--it should have been 6.25% in the first place.
If you keep a pair of twos and an 8, you might draw an 8 and get two pair. But if you just kept the pair of twos, you might draw a 9 first, and then there's a chance you'll draw another 9 in the next two cards. I know you're not looking more than one card ahead, but this is a place where that's a mistake; your odds of getting two pair are almost identical whether you keep the 8 or not. I think it is actually slightly better to keep a card you already have, since there are three more of those out there, whereas if you just keep the pair, the first card you draw might be another 8, and now there are only two left to make further pairs with; but this effect is much smaller than the effect you describe, and I'm still oversimplifying it.
The chance of drawing the 2^ and 3^ are not 1/52 * 1/52; the chance of drawing the 2^ then the 3^ is 1/52 * 1/51, and the chance of drawing the 3^ then the 2^ is 1/52 * 1/51. To put it unambiguously, the chance of drawing either the 2^ or 3^ is 1/26, and the chance of drawing the remaining is 1/51 (overall odds is 1/1326).
Been there, done that.
Since lawyers (or their associations and what not) have come out explicitly pro- software patents, I see no reason not to blame them. (There was a slashdot story a while back with their response to Bezos' suggestion that there needed to be patent reform, dissing the idea.) When it comes to software patents, just follow the money. There have always been three vocal groups in the US in favor of software patents: the patent office, (patent) lawyers, and software patent holders: exactly the people who profit if software patents exist. Those people may argue that software patents are "for the public good", but nobody else does, and they all profit directly from there being software patents.