it will just require faster cameras with better movement algorithms.
Depends, at least with traditional "gesture detection" that won't help. The problem with gesture detection is that you have to actually complete your gesture before the computer can start to analyzing your movement and figure out what you just did, which leads to the expected lag and can't be fixed with a faster camera or computer, since you still have to complete your gesture. The only way to fix this is if one can manage to move away from gestures to true 1:1 mapping where there aren't gestures, but just fluid movement that gets instantly mapped onto the computer (i.e. just like with a mouse you simply move the pointer, you don't do the move-pointer-to-the-left gesture).
If they get that solved tactile feedback is of course the next big hurdle and of course there is also the issue of lack of arm support, holding your arms out into thin air for a long time is not exactly easy and gets tiering very quickly. So I don't expect this to replace a mouse and keyboard anytime soon or ever, but as an additional way for input it might be useful for some cases.
The thing is, old games still provide tons of ideas that you haven't seen in 15 years. They might not be totally new, but they are certainly a lot more interesting then many of the stuff that goes into todays games, since what goes into todays games is mostly the same stuff that went into last years games. And of course also keep in mind that some of those 'old ideas' are older then the gamers that play todays games, so those things will be fresh and new for a newer generation of gamers.
And about the nostalgia thing, its totally different for me. I played X-Com just a few years ago for the first time, back when it was already a well over a decade old and yet it was nothing short of a mind blowing experience where nothing I have played more recently came close and I don't even like turn based games. With Fallout, which I currently play for the first time, it was a little different, since that game has a ton of bugs and issues, but still it was a pretty good experience so far and I had plenty of fun with it, more fun then with many of todays AAA titles. That of course doesn't mean that in the past everything was great, in fact most of the user interfaces of old games are truly terrible and I really wouldn't mind to see those polished up. But some of the core ideas of past games are nothing short of awesome and I would like to see them reused in more modern games. The whole complexity of weapon research, base building, UFO interception, inspection of crash sites, environmental destruction and stuff in XCom was pretty damn cool and I would like to see recycled. I wouldn't even mind if they change the whole thing around from turn-based to something like Full Spectrum Warrior or even something like Gears of War style as long as they keep the scale of the whole thing. Todays games are far to often focused on presenting something from a single point of view, instead of allowing you to see the bigger picture and acting in it.
The problem is that a "good game" is something different from a game I care to play. I fully agree that many of todays games are very solid and would be plenty of fun, however most of those games come from the 'been there, done that'-land and feature the same old gameplay that already got boring last year or the year before. I mean, just look at the weapons in a shooter genre, pistol, shotgun, machine gun and rocket launcher. I have already played with those weapons back then 15 years ago in Doom1, I don't care to repeat that experience over and over and over again. It stopped to be interesting long long ago, not because there is anything fundamentally bad with that game mechanic, but simply because it has been done to death before. Games today have become far to stale and far to tightly locked into their genres to be much of an excitement for me any more.
The reason why old games are still fun is not because they are necessarily a better kind of game, but for most part simply because they are something very different from the big titles that dominate todays market.
It is worse then that. A real simulation at least is interesting because it is complex and allows many different ways how things can interact together and most importantly because it gives you plenty of freedom to act, todays games however seldomly go in that direction. Instead most of them go more into the direction of a roller coaster ride, they are flashy and noise, but ultimately they are repetitive and pointless, because the player really doesn't have all that much to do in them. Its always the same: point at enemy, pull trigger, rinse and repeat. There aren't decision to be made, characters to interact with or stories worth to listen to.
Now I don't expect games to be full of deep meaning, but I want at least some descent player involvement, if that means learning boss patterns in a MegaMan or making moral choices in a deep storyline I don't care, at least both of those keep me involved. But the roller coaster rides that most of todays games provide, which are so easy that you never have to remember much of anything, are getting really annoying.
So, where exactly is the place for vi / emacs today and what benefits do they bring??
In practical terms: very little, if any.
On the more theoretical side of things: When using vi or Emacs you use one editor for all your text-work, so all you shortcuts and commands will work no matter if you are programming java, editing xorg.conf, sending mail or doing stuff over ssh on a computer that is thousand miles away.
However if you are already familiar with an IDE for the language you program in, I seriously doubt that switching to Emacs or vi would have any benefit, in fact there might be plenty of features that you will be missing after the switch.
Which "standard" issues required the console, if I may be so bold to ask?
Tweaking things in xorg.conf generally requires use of a text editor. On SuSE you have Sax to allow many of those modifications via GUI, but on Ubuntu you are stuck with auto-detection and a rather pathetic X configuration tool, which doesn't allow much tweaking of what the auto-detection got wrong.
One of the things that took me by far the longest to figure out was how to construct a while loop over the results of a "find -print0". Solution is this:
find -print0 | while read -d $'\0' -r filename; do
echo $filename; done
Of course in many cases one can also use 'find -exec {} \;', but sometimes a while loop is more handy and without the '-d $'\0'' trick it wouldn't be able to read the file list properly without screwing up on weird filenames.
I don't think the issue is really the 3D part, even so I agree that there should be much more 2D games these days, I think the real issue is simply the development process. Many games of the good old days where written by small teams and thus had much more freedom to experiment around with stuff. Today on the other side you have pretty huge teams, which makes doing something new pretty close to impossible, so everything gets based on tried and tested game concept. The indie fraction on the other side seems to have the problem of having no teams at all, a single person simply can't do as much as a small team of three or five people, thus many indie games end up a little lacking in one area or the other. And last not least, gaming trends seem to ruin it all. Back when Bridge Builder was first released it was awesome, today however we have at least half a dozen other games that are build around the exact same concept, it gets old. Physic engines in general seem to be overused a lot. And there is of course also stuff like the dual-joystick shooter, half the games on XBLA end up falling in that genre, nothing wrong with that genre by itself, but its really over done.
I think one of the core problems is simply that developers these days have grown up with video games, while people in the early in the video game era hadn't, so they took inspiration from books and movies, while todays developers take inspiration from video games themselves, so you end up with an endless cloning process instead of people trying completly new ideas.
This isn't a glitch nor a problem with the machines.
Yeah, the good old "blame the user" solution, its after all just democracy that is at stake...
Why is it even possible for the user to eject the card before stuff is done? Any half decent ATM doesn't allow that, it holds the card inside until everything is finished. Why doesn't the voting machine do the same? Seems to me to be a pretty clear case of a badly designed system.
How do you know if your vote is registered correctly or not?
You stand there and watch while they do the counting. The whole point of pen&paper is that the voter themselves can verify that the voting process happens correctly, everything that isn't pen&paper adds a layer of intransparency that makes it much harder or impossible for the voter to verify the voting process is going as advertised.
e-voting doesn't make fraud any more or less difficult. It just makes things less transparent, and probably makes fraud easier.
E-Voting doesn't only make fraud easier, it makes large scale fraud possible in the first place. With paper you will have a really though time manipulating more then a single ballot box, with E-Voting on the other side you can do large scale fraud pretty easily when you sit at the right spot.
The good thing about pen&paper is that it works even when you can't trust the government, it of course doesn't stop fraud in that case, but it makes it much easier to detect.
I've played games for over 30 years. I'm just tired of games that are now made so frustrating.
You got that all backwards. Games today are relatively long (6-15h for an action game) and pretty easy (reset points all over the place, saving), games in the past however they were very short (1-2h) and very hard (no reset points, limited continues, no saving at all). Also your console vs PC argument doesn't make much sense, in console games you have functions in well documented and well thought out places (i.e. much of the game design is actually build around the controller), on the PC on the other side they are spread all over the keyboard in often seemingly random places and with little or no thought about the actually gaming hardware used, since its different everywhere (game build for five button mouse doesn't play well with three button trackball). Also its much more easy to find a function on a 16 button gamepad then trying to hunt one down on a 102 key keyboard.
Now that of course doesn't mean that gaming today is great, there is tons of stuff wrong with it, but I'd pin that down on lack of innovation and design-by-committee, which makes most games look and feel pretty much the same (bulky space marine with rocket launchers have been around for at least 15 years, so they are rather boring now).
The real problem with the Linux GUIs isn't the window manager or where the buttons are, but that a lot of stuff is still only configurable via the console or just not very good in GUI form (X11 configuration is horribly limited in Ubuntu, until you go 'vi xorg.conf'). Its has gotten slowly better in that area, but still quite a bit away from Windows, where you can 'fix' most stuff by clicking some setup.exe.
Game prices are hardly a real problem. For one thing they have remained constant or even gone down for at least a decade or two, since inflation is something that doesn't happen with game prices for some reason (new PS3 game cost the same as a new SNES or an Amiga game back then). And secondly you know have all those 'classic' and 'platinum' rereleases and via eBay and Amazon access to a huge library of used games, so if you can wait a few month, you can easily get pretty much every game for 30EUR or less. Hardly something to complain about.
Where do you get an unlimited number of real cat pictures from? How do you categorize the pictures in the first place? What stops the spammer from just repeating the process? If you buy a database of cat pictures from a third party the spammer will just buy them too. I can see how KittenAuth might work if only used on a single site, say a cat owner community webpage, where pictures and categorisation is available, but as a general purpose mechanism it just seems highly impractical. It also is possible to brute force, unless you have lots of pictures to click at.
With 3D rendering you have the full control of what objects you want to display, how you want to pose them and what questions you want to ask about them. The storage requirements are tiny compared to a gigantic database of cat pictures and the flexibility you have is much greater. You don't have to work with static meshes, you could do it fully procedural if you want, Spore style. And you could also have a big backlog of 3d models to replace those that turned out to be easy to detect. If spammers have figured out how to detect cats, you just replace them with kitchen sinks.
If you would do that, you would need a database that lists the properties of objects. But if you have such a database, the spammer will simply create one too and match things against that.
What you need is information in a form that is easy for the server to create and encode, but hard for the client to do decode. The problem with simple logic question is that they are hard to create randomly, so even if you can exclude brute force, you are still following a pattern in the creation that the client can reverse engineer.
My solution for the issue would be 3d CAPTCHA, since they are easy to create randomly, but very hard to decode for a machine. And as an added bonus they are also extremely flexible, i.e. you could ask all kinds of logic questions about a scene, without ever getting easy hints that could be matched against a database.
The problem is that you cannot generate pictures of kittens automatically.
Of course you can, thats what we have 3d graphics for. The nice thing about 3d graphics is that you can randomly vary the pose, texture, background, camera angle and so on, so you can produce a pretty much infinite amount of 2d cat pictures. The nice thing about this is that the spammer only gets to see the final 2d render, not the 3d data used to generate it, that way you can easily generate the pictures, but the spammer will have a very hard time getting information out of them. And if cats aren't enough, you can throw a heapload of other 3d meshes into the mix. You can even make this extra hard in that you not only have to click on the picture with the cat, but the cat itself. The server knows where the cat is in the 2d picture, since he has the 3d data, the client on the other side has no easy way to figure that out, which makes brute forcing quite a bit harder. You can also have many variants of questions, like "click on the two cats that look the same" or "click on the cat that has the same texture like the carpet on which the dog lies" or whatever. And you can of course also throw the spammer off by having picture of the cat inside the scene where the cat itself is.
The problem with questions that have a limited answer set is that it would only be hard to answer correctly all the time, but it would be extremly easy to just brute force it and get it right every now and then. A bot doesn't have to be right all the time, but only once to come in. So you have to have questions that you can't just brute force your way around by simply trying A, B and C answers.
Be creative, make it hollow so that its cheap to print and leave a hole so you can fill it with something cheap to give it weight later on.
Re:Sorry, Loebner Has Done Nothing for AI
on
Loebner Talks AI
·
· Score: 2, Interesting
Yes, it tests for one particular form of AI, but that form would be extremely useful to have if achieved.
The problem I have with the test is that it isn't about creating an AI, but about creating something that behaves like human. Most AI, even if highly intelligent, will never behave anything like a human, simply because its something vastly different and build for very different tasks. Now one could of course try to teach that AI to fool a human, but then its simply a game of how good it can cheat, not something that tells you much about its intelligence.
I prefer things like the DARPA Grand Challenge where the goal isn't to create something that behaves like a human, but simply something that gets the given task done. That way you can slowly raise the bar instead of setting a goal where you don't even know if there is a point in chasing it. Turning test feels to much like a challenge to fly by sticking wings to your arms, you might be able to do that one day, but the aviation industry doesn't really care, jumbo jets are flying fine even without flapping their wings.
It is true that a lot of beginners OpenGL code won't work on OpenGL-ES, but the differences between the two isn't really all that big. The main difference is simply that in OpenGL-ES you have to use vertex arrays, while in OpenGL it is an optional thing. While this breaks many code examples its not hard to fix and code that uses vertex arrays will work on both OpenGL and OpenGL-ES.
they kept the ability to play old game boy games for years and years as they brought out new systems.
GBAmicro already lacked support for Gameboy and Gameboy Color titles, DS lacked support for those as well, the DS also lacked the GBA link port, making GBA multiplayer or Gamecube links impossible. So everything after the GBAsp was already crippled and its no surprise that they removed the port on the DSi, since it wastes quite a huge amount of space that likely was needed for the bigger screens, cameras and SD card slot.
Re:Any chance we can draw circles and boxes now
on
GIMP 2.6 Released
·
· Score: 1
Yes, those pictures are a little old, however even with current Gimp I am unable to do a perfect circle, they look a little better then before, but there are still way to many pixels in the wrong places.
Slashdot Mirror
it will just require faster cameras with better movement algorithms.
Depends, at least with traditional "gesture detection" that won't help. The problem with gesture detection is that you have to actually complete your gesture before the computer can start to analyzing your movement and figure out what you just did, which leads to the expected lag and can't be fixed with a faster camera or computer, since you still have to complete your gesture. The only way to fix this is if one can manage to move away from gestures to true 1:1 mapping where there aren't gestures, but just fluid movement that gets instantly mapped onto the computer (i.e. just like with a mouse you simply move the pointer, you don't do the move-pointer-to-the-left gesture).
If they get that solved tactile feedback is of course the next big hurdle and of course there is also the issue of lack of arm support, holding your arms out into thin air for a long time is not exactly easy and gets tiering very quickly. So I don't expect this to replace a mouse and keyboard anytime soon or ever, but as an additional way for input it might be useful for some cases.
Rehash old sure-bet ideas.
The thing is, old games still provide tons of ideas that you haven't seen in 15 years. They might not be totally new, but they are certainly a lot more interesting then many of the stuff that goes into todays games, since what goes into todays games is mostly the same stuff that went into last years games. And of course also keep in mind that some of those 'old ideas' are older then the gamers that play todays games, so those things will be fresh and new for a newer generation of gamers.
And about the nostalgia thing, its totally different for me. I played X-Com just a few years ago for the first time, back when it was already a well over a decade old and yet it was nothing short of a mind blowing experience where nothing I have played more recently came close and I don't even like turn based games. With Fallout, which I currently play for the first time, it was a little different, since that game has a ton of bugs and issues, but still it was a pretty good experience so far and I had plenty of fun with it, more fun then with many of todays AAA titles. That of course doesn't mean that in the past everything was great, in fact most of the user interfaces of old games are truly terrible and I really wouldn't mind to see those polished up. But some of the core ideas of past games are nothing short of awesome and I would like to see them reused in more modern games. The whole complexity of weapon research, base building, UFO interception, inspection of crash sites, environmental destruction and stuff in XCom was pretty damn cool and I would like to see recycled. I wouldn't even mind if they change the whole thing around from turn-based to something like Full Spectrum Warrior or even something like Gears of War style as long as they keep the scale of the whole thing. Todays games are far to often focused on presenting something from a single point of view, instead of allowing you to see the bigger picture and acting in it.
The problem is that a "good game" is something different from a game I care to play. I fully agree that many of todays games are very solid and would be plenty of fun, however most of those games come from the 'been there, done that'-land and feature the same old gameplay that already got boring last year or the year before. I mean, just look at the weapons in a shooter genre, pistol, shotgun, machine gun and rocket launcher. I have already played with those weapons back then 15 years ago in Doom1, I don't care to repeat that experience over and over and over again. It stopped to be interesting long long ago, not because there is anything fundamentally bad with that game mechanic, but simply because it has been done to death before. Games today have become far to stale and far to tightly locked into their genres to be much of an excitement for me any more.
The reason why old games are still fun is not because they are necessarily a better kind of game, but for most part simply because they are something very different from the big titles that dominate todays market.
It is worse then that. A real simulation at least is interesting because it is complex and allows many different ways how things can interact together and most importantly because it gives you plenty of freedom to act, todays games however seldomly go in that direction. Instead most of them go more into the direction of a roller coaster ride, they are flashy and noise, but ultimately they are repetitive and pointless, because the player really doesn't have all that much to do in them. Its always the same: point at enemy, pull trigger, rinse and repeat. There aren't decision to be made, characters to interact with or stories worth to listen to.
Now I don't expect games to be full of deep meaning, but I want at least some descent player involvement, if that means learning boss patterns in a MegaMan or making moral choices in a deep storyline I don't care, at least both of those keep me involved. But the roller coaster rides that most of todays games provide, which are so easy that you never have to remember much of anything, are getting really annoying.
So, where exactly is the place for vi / emacs today and what benefits do they bring??
In practical terms: very little, if any.
On the more theoretical side of things: When using vi or Emacs you use one editor for all your text-work, so all you shortcuts and commands will work no matter if you are programming java, editing xorg.conf, sending mail or doing stuff over ssh on a computer that is thousand miles away.
However if you are already familiar with an IDE for the language you program in, I seriously doubt that switching to Emacs or vi would have any benefit, in fact there might be plenty of features that you will be missing after the switch.
Which "standard" issues required the console, if I may be so bold to ask?
Tweaking things in xorg.conf generally requires use of a text editor. On SuSE you have Sax to allow many of those modifications via GUI, but on Ubuntu you are stuck with auto-detection and a rather pathetic X configuration tool, which doesn't allow much tweaking of what the auto-detection got wrong.
One of the things that took me by far the longest to figure out was how to construct a while loop over the results of a "find -print0". Solution is this:
find -print0 | while read -d $'\0' -r filename; do
echo $filename;
done
Of course in many cases one can also use 'find -exec {} \;', but sometimes a while loop is more handy and without the '-d $'\0'' trick it wouldn't be able to read the file list properly without screwing up on weird filenames.
Other useful things:
export HISTCONTROL=ignoredups
export HISTSIZE="100000"
export BASH_HISTSIZE="1000000"
So you have a long history to search for useful commands.
Solution to a non-existent problem? The gaming world seems to disagree, since Ogg is used in plenty of games.
I don't think the issue is really the 3D part, even so I agree that there should be much more 2D games these days, I think the real issue is simply the development process. Many games of the good old days where written by small teams and thus had much more freedom to experiment around with stuff. Today on the other side you have pretty huge teams, which makes doing something new pretty close to impossible, so everything gets based on tried and tested game concept. The indie fraction on the other side seems to have the problem of having no teams at all, a single person simply can't do as much as a small team of three or five people, thus many indie games end up a little lacking in one area or the other. And last not least, gaming trends seem to ruin it all. Back when Bridge Builder was first released it was awesome, today however we have at least half a dozen other games that are build around the exact same concept, it gets old. Physic engines in general seem to be overused a lot. And there is of course also stuff like the dual-joystick shooter, half the games on XBLA end up falling in that genre, nothing wrong with that genre by itself, but its really over done.
I think one of the core problems is simply that developers these days have grown up with video games, while people in the early in the video game era hadn't, so they took inspiration from books and movies, while todays developers take inspiration from video games themselves, so you end up with an endless cloning process instead of people trying completly new ideas.
This isn't a glitch nor a problem with the machines.
Yeah, the good old "blame the user" solution, its after all just democracy that is at stake...
Why is it even possible for the user to eject the card before stuff is done? Any half decent ATM doesn't allow that, it holds the card inside until everything is finished. Why doesn't the voting machine do the same? Seems to me to be a pretty clear case of a badly designed system.
How do you know if your vote is registered correctly or not?
You stand there and watch while they do the counting. The whole point of pen&paper is that the voter themselves can verify that the voting process happens correctly, everything that isn't pen&paper adds a layer of intransparency that makes it much harder or impossible for the voter to verify the voting process is going as advertised.
e-voting doesn't make fraud any more or less difficult. It just makes things less transparent, and probably makes fraud easier.
E-Voting doesn't only make fraud easier, it makes large scale fraud possible in the first place. With paper you will have a really though time manipulating more then a single ballot box, with E-Voting on the other side you can do large scale fraud pretty easily when you sit at the right spot.
The good thing about pen&paper is that it works even when you can't trust the government, it of course doesn't stop fraud in that case, but it makes it much easier to detect.
I've played games for over 30 years. I'm just tired of games that are now made so frustrating.
You got that all backwards. Games today are relatively long (6-15h for an action game) and pretty easy (reset points all over the place, saving), games in the past however they were very short (1-2h) and very hard (no reset points, limited continues, no saving at all). Also your console vs PC argument doesn't make much sense, in console games you have functions in well documented and well thought out places (i.e. much of the game design is actually build around the controller), on the PC on the other side they are spread all over the keyboard in often seemingly random places and with little or no thought about the actually gaming hardware used, since its different everywhere (game build for five button mouse doesn't play well with three button trackball). Also its much more easy to find a function on a 16 button gamepad then trying to hunt one down on a 102 key keyboard.
Now that of course doesn't mean that gaming today is great, there is tons of stuff wrong with it, but I'd pin that down on lack of innovation and design-by-committee, which makes most games look and feel pretty much the same (bulky space marine with rocket launchers have been around for at least 15 years, so they are rather boring now).
The real problem with the Linux GUIs isn't the window manager or where the buttons are, but that a lot of stuff is still only configurable via the console or just not very good in GUI form (X11 configuration is horribly limited in Ubuntu, until you go 'vi xorg.conf'). Its has gotten slowly better in that area, but still quite a bit away from Windows, where you can 'fix' most stuff by clicking some setup.exe.
Game prices are hardly a real problem. For one thing they have remained constant or even gone down for at least a decade or two, since inflation is something that doesn't happen with game prices for some reason (new PS3 game cost the same as a new SNES or an Amiga game back then). And secondly you know have all those 'classic' and 'platinum' rereleases and via eBay and Amazon access to a huge library of used games, so if you can wait a few month, you can easily get pretty much every game for 30EUR or less. Hardly something to complain about.
Quite right, there's no reason whatsoever why 98% of users shouldn't be behind NAT gateways.
Users should be behind a firewall, not behind NAT, beside breaking the Internet, not really doesn't help much.
Where do you get an unlimited number of real cat pictures from? How do you categorize the pictures in the first place? What stops the spammer from just repeating the process? If you buy a database of cat pictures from a third party the spammer will just buy them too. I can see how KittenAuth might work if only used on a single site, say a cat owner community webpage, where pictures and categorisation is available, but as a general purpose mechanism it just seems highly impractical. It also is possible to brute force, unless you have lots of pictures to click at.
With 3D rendering you have the full control of what objects you want to display, how you want to pose them and what questions you want to ask about them. The storage requirements are tiny compared to a gigantic database of cat pictures and the flexibility you have is much greater. You don't have to work with static meshes, you could do it fully procedural if you want, Spore style. And you could also have a big backlog of 3d models to replace those that turned out to be easy to detect. If spammers have figured out how to detect cats, you just replace them with kitchen sinks.
If you would do that, you would need a database that lists the properties of objects. But if you have such a database, the spammer will simply create one too and match things against that.
What you need is information in a form that is easy for the server to create and encode, but hard for the client to do decode. The problem with simple logic question is that they are hard to create randomly, so even if you can exclude brute force, you are still following a pattern in the creation that the client can reverse engineer.
My solution for the issue would be 3d CAPTCHA, since they are easy to create randomly, but very hard to decode for a machine. And as an added bonus they are also extremely flexible, i.e. you could ask all kinds of logic questions about a scene, without ever getting easy hints that could be matched against a database.
The problem is that you cannot generate pictures of kittens automatically.
Of course you can, thats what we have 3d graphics for. The nice thing about 3d graphics is that you can randomly vary the pose, texture, background, camera angle and so on, so you can produce a pretty much infinite amount of 2d cat pictures. The nice thing about this is that the spammer only gets to see the final 2d render, not the 3d data used to generate it, that way you can easily generate the pictures, but the spammer will have a very hard time getting information out of them. And if cats aren't enough, you can throw a heapload of other 3d meshes into the mix. You can even make this extra hard in that you not only have to click on the picture with the cat, but the cat itself. The server knows where the cat is in the 2d picture, since he has the 3d data, the client on the other side has no easy way to figure that out, which makes brute forcing quite a bit harder. You can also have many variants of questions, like "click on the two cats that look the same" or "click on the cat that has the same texture like the carpet on which the dog lies" or whatever. And you can of course also throw the spammer off by having picture of the cat inside the scene where the cat itself is.
The problem with questions that have a limited answer set is that it would only be hard to answer correctly all the time, but it would be extremly easy to just brute force it and get it right every now and then. A bot doesn't have to be right all the time, but only once to come in. So you have to have questions that you can't just brute force your way around by simply trying A, B and C answers.
Be creative, make it hollow so that its cheap to print and leave a hole so you can fill it with something cheap to give it weight later on.
Yes, it tests for one particular form of AI, but that form would be extremely useful to have if achieved.
The problem I have with the test is that it isn't about creating an AI, but about creating something that behaves like human. Most AI, even if highly intelligent, will never behave anything like a human, simply because its something vastly different and build for very different tasks. Now one could of course try to teach that AI to fool a human, but then its simply a game of how good it can cheat, not something that tells you much about its intelligence.
I prefer things like the DARPA Grand Challenge where the goal isn't to create something that behaves like a human, but simply something that gets the given task done. That way you can slowly raise the bar instead of setting a goal where you don't even know if there is a point in chasing it. Turning test feels to much like a challenge to fly by sticking wings to your arms, you might be able to do that one day, but the aviation industry doesn't really care, jumbo jets are flying fine even without flapping their wings.
It is true that a lot of beginners OpenGL code won't work on OpenGL-ES, but the differences between the two isn't really all that big. The main difference is simply that in OpenGL-ES you have to use vertex arrays, while in OpenGL it is an optional thing. While this breaks many code examples its not hard to fix and code that uses vertex arrays will work on both OpenGL and OpenGL-ES.
They did, the DSi has both a 3MP camera (on the outside) and a 0.3MP camera (on the hinge inside), later one is for video chat I assume.
they kept the ability to play old game boy games for years and years as they brought out new systems.
GBAmicro already lacked support for Gameboy and Gameboy Color titles, DS lacked support for those as well, the DS also lacked the GBA link port, making GBA multiplayer or Gamecube links impossible. So everything after the GBAsp was already crippled and its no surprise that they removed the port on the DSi, since it wastes quite a huge amount of space that likely was needed for the bigger screens, cameras and SD card slot.
Yes, those pictures are a little old, however even with current Gimp I am unable to do a perfect circle, they look a little better then before, but there are still way to many pixels in the wrong places.