You also have to factor in the fact that in most states the outcome is obvious. If the better electable candidate is getting twice the number of votes in the state that the worse electable candidate is getting, he can't lose that state due to protest votes. If enough people vote for a different candidate to make the better electable one lose, the third-party candidate was electable after all.
Likewise, if you know the state doesn't have enough voters for the better electable candidate to get him to win, you might as well vote for a third-party candidate, because you're not going to help to defeat the worse candidate.
So the only states where voting for a particular candidate will matter are close states. Other than that, the only thing that votes change are the popular vote count, and voting third-party doesn't take away support from an electable candidate there.
Or, at least, it helps demonstrate that something was reverse engineered, rather than simply copied. If I have no chance of following the wires and such, but I can still replicate the device, I must have treated it as a black box, and I thus escape copyright. It's easy to show your version to be a clean-room implementation if there's no way you can get yourself contaminated.
One reason to use a swap partition is so that your disk usage doesn't vary widely, depending on memory usage. Also, if your file system tends to get corrupted if the OS crashes while writing to it, it's a bad idea to have the OS write to a file system whenever memory is low. It's somewhat safer if you crash while only writing to a different partition.
Actually, using anything on the same disk is a generally poor idea. When you're low on memory, it's generally because something is putting a lot of stuff there. Where does this stuff come from? Generally, disk. So you're pulling stuff in from disk and writing other stuff out to disk. This creates a major performance bottleneck. Whenever possible, I use a swap disk (or even disk chain, if there's one not in use), which is clearly going to be a fixed size.
Do you, incidentally, know the current status of air traffic control computers? The system in use now runs only on hardware that doesn't exist anywhere else anymore and can't be replaced. There have been several attempts to replace it with a newer system, but none have worked. It's a reasonably difficult system to build the basic functionality for, and then there are a ton of features that the users would like, but the system can never crash, and has to be able to deal with various disasters without crippling anything.
It's precisely the sort of project that a company will tend to get bogged down on writing and then be unable to debug.
I don't think it's a terribly good candidate for an open-source project, though, since open-source is generally good when there are programmers who actually want a copy of the program. On the other hand, I don't see any reason to have it be closed-source. It's not like the company is going to have to deal with competitors stealing their code or have to track down unlicensed installations.
One thing I'd like about a wearable (which I like about my visor) is that I can walk away from my desk, be inaccessable, but still be able to write stuff. Technology has a habit of invading the technology, as well as the simple things. I'd really go for a chance to get away from my desk more without being reduced to pen and paper (or memo program).
I've got the desktop (fvwm with xterms, mainly), the web site (slashdot lite, for example), a newsgroup or two (but they're clearly a secret), and the keyboard (veteran of a cream soda, which made the keys stick, because they were sticky).
There's no reason all of the distributions will merge once they become functionally equivalent. At that point, you'll probably have different distributions for the same reason that you have Dell, Gateway, IBM, etc. making the machines: they come with slightly different stuff, the company's own tech support, various tweaks in style that don't matter for compatibility, and so on.
Any particular company will probably use all the same distribution, just to have a single vendor relationship, but there's no reason that multiple compatible distributions can't co-exist with different patrons.
Linux capabilities aren't nearly sufficient yet, but they could eventually become that complete. The main problem is that UNIX is just not designed for that sort of thing. Even just adding acls to a filesystem makes things not work the way people and existing program expect. Eventually, however, the issues may get worked out, and there'd be a real capabilities system.
Even breaking up root priviledges somewhat would be useful, though; make a stack-smashing attack only give you the little priviledge that the program that you found a hole in had. Beyond that, you could also restrict some usual user priviledges (exec comes to mind), to eliminate abilities that daemons don't really need. I.e., make it so bind can't call exec or do anything priviledged except open low-numbered sockets.
On the other hand, this adds a bunch more stuff that has to be configured correctly, as well as requiring some new way of setting all this.
I think one reason that some stuff isn't going into the official standard kernel is that there's no way to put code into the official kernel such that people who don't want it don't have to download it. It would be really helpful if you could run a configuration pass, and then download only those files that you were actually going to use. That way the kernel sources could get really big, containing all the patches and versions of stuff that are probably good ideas, without making it impractical to get and unpack.
There's no real reason there can't be different official memory managers for low memory and high memory situations, since there are clearly different issues. Of course, at this point, lots of people testing a single one is important.
There is starting to be support for capabilities in Linux (a major requirement for B and higher stuff). Of course, this is a major break from the UNIX model, where root has all the capabilities, and other people may be granted root-like power in restricted code, and everything else is done through the file system with simple ACLs (i.e. gorups).
It may be possible to make a B-level distribution, assuming that physical access is controlled, and programs are set up very carefully. But you probably wouldn't find it terribly useful, since nobody could become root, as that would seriously break the security model. You'd basically have to deal with not having a user-level capabilities system by lacking abilities.
Of course, you could probably get C2 by turning off all the services you don't actually want, removing the setuid bit from programs that shouldn't have it, restricting access to some other programs, and replacing the rest of the setuid programs with versions which are simple enough to verify their security.
Generally, many of these ratings aren't very helpful unless you're a government, because at the higher levels it's mostly concerned with making sure that your secret data can't go to untrusted places. If you're big enough that you actually talk to trusted places, this is helpful, but for most places, it means the computer is unusable.
For example... the machine can't let you cut and paste from a secret document to anything like, say, a web browser or ssh window. It can't let you accomplish this in several steps, either. It quickly becomes impossible to deal with having anything that can send information out to anything but verifiable secure and trusted sites. Not only do your directory listings not include secret files if you're not a trusted user, they don't even if you are, if you can copy out to something untrusted. It's actually easier on the user to have a separate machine for secret data, and it's all silly unless you're also searching your employees for secret files at the door.
Of course, in the business world, you generally don't deal with secret data on this level. Security is aimed at preventing access from users who shouldn't get it, not preventing spies from getting information out. B3 or C2 is about where it's worth getting, and beyond that what you're interested in is an entirely different scale of security.
It would probably be in the "B" range (access control to data, separation between public and private keys, except it doesn't keep the necessary logs), but, since it's not a full system, and only a single application, it doesn't get a rating. As a component it could probably be helpful in doing the required authentication, but that's not enough to get a rating (it would also need to provide logs for this, probably).
Angband is actually developed in a bazaar style. There are a lot of variants, created when people think things should be different. Patches move between them as the variant maintainers see and like what other people are doing. People sometimes create generic patches, which are based on one variant but could easily be put into different ones.
It actually seems to work quite well, for what they're doing. Part of it is that there's a large artistic component-- determining what stuff should be in the game, and people have a reason to fork the code to suit their tastes. Part of it is that there are still maintainers, so there is a single person in charge of determining whether a patch is applied to a file; but the patch may go to many different people, or circulate as an unofficial patch.
Linux (for example) does seem to follow this model if you include the large collection of unofficial patches which circulate. It's just that what you tend to see is the cathedral version of the program on the official site.
It seems like everyone these days is trying to get their site to appear on all searches. This mystifies me, because if the page isn't titled something that looks like what I'm looking for, I'm not going to look at the page, and I'm probably going to be annoyed when a site keeps coming up on searches for other things, and I wouldn't go there even if I *was* interested.
It makes sense to try to get your site to appear first on a search that actually fits your site. But if the only sites fighting for the top spot are informative pages on the topic I'm actually interested in, I don't think that's really a problem, and it's not really cheating.
It may be that I have a tendancy to look for a particular web page (e.g., the official screen home page), and I don't want other pages on the topic (e.g., the GNU page about screen, which is non-gnu), but cheating never gets my eyes and certainly never gets me to explore a site, and definitely won't get me to advertizers' sites.
It would probably be helpful if search engines would give instructions on how to make your site come up on searches that are looking for it and not on searches which aren't. Of course, having the search engine be better at determining what the site is about without special help and making it easier to tell the engine what you're looking for would be even more helpful.
(incidentally, I've found an official screen ftp site, but no web site; it may very well not have one)
A lot of temporary GPL violations and mistaken impressions of such seem to come from the fact that there's not a standard way to find the required source distribution when you see a product with GPLed code in it. The source fails to appear because there's not someone at the company making sure it's up to date, or the link isn't right, or the web page is unreadable or just hard to navigate. Even if people *offer* the source, it's often hard to actually *get* it. It would be nice if the FSF came up with at least a suggested method for distributing the source.
The other problem I see a lot is sites where you follow a link, and you're not sure what you're getting. "Download Linux source" might give you the source to Linux (under GPL) or the source to a Linux version of a program (proprietary). Sometimes you follow a link expecting to get the program, and you get a generic page telling you where to find mirrors or something. Then other times you're trying to find information about a program, and you get the program itself instead.
MS Research doesn't generally make products, but makes the hidden bits of stuff that MS then incorporates into products. For instance, MS Research figured out how to search help files based on what you seemed to be trying to do in the application; MS made the paperclip annoying.
If anything, this is the version that's standards compliant that MS will start from and extend.
I was a TA for a compiler class last year (think really big programs that need to be graded), and we decided we'd only officially support one version of the compiler and library and interpreter (we used Java). This was essentially so that we wouldn't have to deal with incompatabilities between environments, and, in particular, the situation where the code works for the student but we can't get it to work.
It's pretty much impossible to avoid this sort of thing-- otherwise, you make the TAs find a different type of machine to grade some of the projects, which just makes a process that is already very hard and slow even worse.
The choice of official platform is, of course, very important. If some of the students don't have access to the official platform, it's going to be very difficult for them to take the class; while they may not actually have any problems, they can't tell if their program will work for the TAs. We chose a version present on most of the public workstations, because we knew that any student concerned about compatibility issues would be able to try their code on the platform we were using.
Of course, when I took the class, I used the pizza compiler which came with the version of Kaffe I was using, and it worked fine. My group used a bunch of different compilers and make sure they all agreed.
On the other hand, C++ has some specification issues; the language specification actually doesn't say what happens in many situations, and different compilers treat it differently. (xevil, written in C++, actually works differently depending on the compiler used to build it) So be sure that you don't do anything dangerous like that, and make sure you're only doing things actually specified by the language.
Re:Proving ADD Doesn't Exist in These Kids
on
Video Games and ADD
·
· Score: 2
ADD is generally an inability to concentrate when you want to-- you can concentrate sometimes (often better than most people), but not on what you choose to focus on.
This tends to mean that people with ADD have more good ideas, since they shift easily and hyperfocus, but their idea tend to be less useful to them, because they tend to be about something other than what they're intending to think about.
Learning to control their focus when they want to can be very helpful for this reason; in some situations it's better not to have ADD, while in others it's beneficial. Being able to switch under your own control is really helpful.
Just because it can be controlled without drugs doesn't mean it's not a neurochemical disorder, either. A large portion of what the brain does is produce drugs for itself, so being able to control it mentally may involve producing the right neurotransmitters to get a more normal mix.
At least one common form of dyslexia, caused in part, ironically, by some video games, makes it nearly impossible to read, but can be fixed by covering parts of the text that aren't being read. Still, the problem manifests itself in ways that can be measured objectively (the person is trained to respond to somewhere next to where they're looking, i.e., the place stuff will happen next in a side-scrolling video game).
Somebody's really got to enter a program in purely functional C. Or, at least, C without functions with side effects (only modifying temporary variables). It should be possible, although clearly harder than your average C style. Probably be pretty hard to make it readable, but it would have to get style points for making it contest-relevant.
Unfortuantely, anything I could consider doing in 72 hours in purely functional C would be really easy for just about anyone with a more sensible environment.
I think the most important point he makes in this article is that Napster gives people a chance to develop taste individually; they can just get a bunch of things to try, and listen to music they actually like. There's not the advertising that gets done in the traditional channels telling people what they should like. In fact, there are parody songs and mixes which are actually reasonably well known (e.g., the Eminem+other stuff mixes) which probably haven't even got spare-time ads.
It's a very different world where people only have to invest the time of listening to try something new. There's much more pressure to have the music actually be good, because that's all the exposure it'll get; only word of mouth is a fast enough advertisement when you listen to whatever you happen to want to at the moment. Of course, there still needs to be an incentive for people to make their music do well, but that'll certainly come eventually-- whether tour sales, nicely packaged CDs, or direct fan support.
IIRC, they originally got this added to the bill *after* it was passed, which tends to indicate that congress actually wanted not to have this in there (otherwise they would have gotten actual congressmen to insert it). Once it become obvious how it got there and the government was taking notice of them, the RIAA probably felt that congress was likely to repeal that provision, and possibly do something nasty to them in retribution, and so they should try to get rid of the evidence.
After all, having congress know you've defrauded them and stand by your actions is probably not a good way to ensure your future existence.
It seems to me that what we've seen so far of C# has been language features and syntax. What we haven't seen is the main library, which is presumably the part that ties to whole thing to Windows.
We ought to be able to pretty simply modify a Java compiler to compile C# code that uses the standard Java libraries instead of MS ones and produces slightly stylized Java bytecode. Of the language features that seem important, none of them seem like anything a compiler couldn't handle and get the standard Java libraries to support. If we hacked a bytecode interpreter sufficiently, we could probably even get the efficiency gains that C# is supposed to give.
It would be really amusing if there was a working C# for linux available before there was one for Windows. Ignoring, of course, all the Windows-specific parts that will probably take a while to implement anyway.
I just tried it and found that it worked quite well, especially considering that I wasn't yet used to it at all.
The thing that I like better about graffitti, though, is that I can do it reasonably accurately without looking at all towards the device, so I can take notes while continuing to watch the speaker, or enjoy the view while writing. Does your aim get sufficiently good that you don't need to look to type?
Much of that can be configured (at least with X) by either changing or configuring the window manager, or by running apps that have those features.
I'd like to see interface plugins that work at a more application-internal level. For instance, the application could specify the menu items, but some configurations might put some of them additionally in a tool bar, and some might organize them into menus differently (is "Stop Loading" View, Go, Edit, or something else? Are those even different menus? Are "Cut" and "Paste" in menus at all, or just done with the mouse?).
File management generally fits in this category; the application should just say it wants a file name and what sort of guidelines there should be (should it exist already? Be a certain type of file? Have a certain extension?), and the plugin widget would handle getting the user's choice.
Having used Linux almost exclusively for years, I have found that I managed to unlearn some of the other GUI elements. There was one time I sat down at either a Mac or a Windows box and couldn't figure out why there were icons for programs that weren't running and none for programs that were running. The concept of having the filesystem visible in windows and in the background was just one I had entirely forgotten about. I've had times when I couldn't understand that it's okay that you can't iconify windows on Macs.
So there really is quite a bit of variation in GUIs, enough to significantly confuse people who aren't used to the particular style they are using.
Slashdot Mirror
You also have to factor in the fact that in most states the outcome is obvious. If the better electable candidate is getting twice the number of votes in the state that the worse electable candidate is getting, he can't lose that state due to protest votes. If enough people vote for a different candidate to make the better electable one lose, the third-party candidate was electable after all.
Likewise, if you know the state doesn't have enough voters for the better electable candidate to get him to win, you might as well vote for a third-party candidate, because you're not going to help to defeat the worse candidate.
So the only states where voting for a particular candidate will matter are close states. Other than that, the only thing that votes change are the popular vote count, and voting third-party doesn't take away support from an electable candidate there.
Or, at least, it helps demonstrate that something was reverse engineered, rather than simply copied. If I have no chance of following the wires and such, but I can still replicate the device, I must have treated it as a black box, and I thus escape copyright. It's easy to show your version to be a clean-room implementation if there's no way you can get yourself contaminated.
One reason to use a swap partition is so that your disk usage doesn't vary widely, depending on memory usage. Also, if your file system tends to get corrupted if the OS crashes while writing to it, it's a bad idea to have the OS write to a file system whenever memory is low. It's somewhat safer if you crash while only writing to a different partition.
Actually, using anything on the same disk is a generally poor idea. When you're low on memory, it's generally because something is putting a lot of stuff there. Where does this stuff come from? Generally, disk. So you're pulling stuff in from disk and writing other stuff out to disk. This creates a major performance bottleneck. Whenever possible, I use a swap disk (or even disk chain, if there's one not in use), which is clearly going to be a fixed size.
Do you, incidentally, know the current status of air traffic control computers? The system in use now runs only on hardware that doesn't exist anywhere else anymore and can't be replaced. There have been several attempts to replace it with a newer system, but none have worked. It's a reasonably difficult system to build the basic functionality for, and then there are a ton of features that the users would like, but the system can never crash, and has to be able to deal with various disasters without crippling anything.
It's precisely the sort of project that a company will tend to get bogged down on writing and then be unable to debug.
I don't think it's a terribly good candidate for an open-source project, though, since open-source is generally good when there are programmers who actually want a copy of the program. On the other hand, I don't see any reason to have it be closed-source. It's not like the company is going to have to deal with competitors stealing their code or have to track down unlicensed installations.
One thing I'd like about a wearable (which I like about my visor) is that I can walk away from my desk, be inaccessable, but still be able to write stuff. Technology has a habit of invading the technology, as well as the simple things. I'd really go for a chance to get away from my desk more without being reduced to pen and paper (or memo program).
I've got the desktop (fvwm with xterms, mainly), the web site (slashdot lite, for example), a newsgroup or two (but they're clearly a secret), and the keyboard (veteran of a cream soda, which made the keys stick, because they were sticky).
Can't help you with the others...
There's no reason all of the distributions will merge once they become functionally equivalent. At that point, you'll probably have different distributions for the same reason that you have Dell, Gateway, IBM, etc. making the machines: they come with slightly different stuff, the company's own tech support, various tweaks in style that don't matter for compatibility, and so on.
Any particular company will probably use all the same distribution, just to have a single vendor relationship, but there's no reason that multiple compatible distributions can't co-exist with different patrons.
Linux capabilities aren't nearly sufficient yet, but they could eventually become that complete. The main problem is that UNIX is just not designed for that sort of thing. Even just adding acls to a filesystem makes things not work the way people and existing program expect. Eventually, however, the issues may get worked out, and there'd be a real capabilities system.
Even breaking up root priviledges somewhat would be useful, though; make a stack-smashing attack only give you the little priviledge that the program that you found a hole in had. Beyond that, you could also restrict some usual user priviledges (exec comes to mind), to eliminate abilities that daemons don't really need. I.e., make it so bind can't call exec or do anything priviledged except open low-numbered sockets.
On the other hand, this adds a bunch more stuff that has to be configured correctly, as well as requiring some new way of setting all this.
I think one reason that some stuff isn't going into the official standard kernel is that there's no way to put code into the official kernel such that people who don't want it don't have to download it. It would be really helpful if you could run a configuration pass, and then download only those files that you were actually going to use. That way the kernel sources could get really big, containing all the patches and versions of stuff that are probably good ideas, without making it impractical to get and unpack.
There's no real reason there can't be different official memory managers for low memory and high memory situations, since there are clearly different issues. Of course, at this point, lots of people testing a single one is important.
There is starting to be support for capabilities in Linux (a major requirement for B and higher stuff). Of course, this is a major break from the UNIX model, where root has all the capabilities, and other people may be granted root-like power in restricted code, and everything else is done through the file system with simple ACLs (i.e. gorups).
It may be possible to make a B-level distribution, assuming that physical access is controlled, and programs are set up very carefully. But you probably wouldn't find it terribly useful, since nobody could become root, as that would seriously break the security model. You'd basically have to deal with not having a user-level capabilities system by lacking abilities.
Of course, you could probably get C2 by turning off all the services you don't actually want, removing the setuid bit from programs that shouldn't have it, restricting access to some other programs, and replacing the rest of the setuid programs with versions which are simple enough to verify their security.
Generally, many of these ratings aren't very helpful unless you're a government, because at the higher levels it's mostly concerned with making sure that your secret data can't go to untrusted places. If you're big enough that you actually talk to trusted places, this is helpful, but for most places, it means the computer is unusable.
For example... the machine can't let you cut and paste from a secret document to anything like, say, a web browser or ssh window. It can't let you accomplish this in several steps, either. It quickly becomes impossible to deal with having anything that can send information out to anything but verifiable secure and trusted sites. Not only do your directory listings not include secret files if you're not a trusted user, they don't even if you are, if you can copy out to something untrusted. It's actually easier on the user to have a separate machine for secret data, and it's all silly unless you're also searching your employees for secret files at the door.
Of course, in the business world, you generally don't deal with secret data on this level. Security is aimed at preventing access from users who shouldn't get it, not preventing spies from getting information out. B3 or C2 is about where it's worth getting, and beyond that what you're interested in is an entirely different scale of security.
"Pretty Good", of course.
It would probably be in the "B" range (access control to data, separation between public and private keys, except it doesn't keep the necessary logs), but, since it's not a full system, and only a single application, it doesn't get a rating. As a component it could probably be helpful in doing the required authentication, but that's not enough to get a rating (it would also need to provide logs for this, probably).
Angband is actually developed in a bazaar style. There are a lot of variants, created when people think things should be different. Patches move between them as the variant maintainers see and like what other people are doing. People sometimes create generic patches, which are based on one variant but could easily be put into different ones.
It actually seems to work quite well, for what they're doing. Part of it is that there's a large artistic component-- determining what stuff should be in the game, and people have a reason to fork the code to suit their tastes. Part of it is that there are still maintainers, so there is a single person in charge of determining whether a patch is applied to a file; but the patch may go to many different people, or circulate as an unofficial patch.
Linux (for example) does seem to follow this model if you include the large collection of unofficial patches which circulate. It's just that what you tend to see is the cathedral version of the program on the official site.
It seems like everyone these days is trying to get their site to appear on all searches. This mystifies me, because if the page isn't titled something that looks like what I'm looking for, I'm not going to look at the page, and I'm probably going to be annoyed when a site keeps coming up on searches for other things, and I wouldn't go there even if I *was* interested.
It makes sense to try to get your site to appear first on a search that actually fits your site. But if the only sites fighting for the top spot are informative pages on the topic I'm actually interested in, I don't think that's really a problem, and it's not really cheating.
It may be that I have a tendancy to look for a particular web page (e.g., the official screen home page), and I don't want other pages on the topic (e.g., the GNU page about screen, which is non-gnu), but cheating never gets my eyes and certainly never gets me to explore a site, and definitely won't get me to advertizers' sites.
It would probably be helpful if search engines would give instructions on how to make your site come up on searches that are looking for it and not on searches which aren't. Of course, having the search engine be better at determining what the site is about without special help and making it easier to tell the engine what you're looking for would be even more helpful.
(incidentally, I've found an official screen ftp site, but no web site; it may very well not have one)
A lot of temporary GPL violations and mistaken impressions of such seem to come from the fact that there's not a standard way to find the required source distribution when you see a product with GPLed code in it. The source fails to appear because there's not someone at the company making sure it's up to date, or the link isn't right, or the web page is unreadable or just hard to navigate. Even if people *offer* the source, it's often hard to actually *get* it. It would be nice if the FSF came up with at least a suggested method for distributing the source.
The other problem I see a lot is sites where you follow a link, and you're not sure what you're getting. "Download Linux source" might give you the source to Linux (under GPL) or the source to a Linux version of a program (proprietary). Sometimes you follow a link expecting to get the program, and you get a generic page telling you where to find mirrors or something. Then other times you're trying to find information about a program, and you get the program itself instead.
MS Research doesn't generally make products, but makes the hidden bits of stuff that MS then incorporates into products. For instance, MS Research figured out how to search help files based on what you seemed to be trying to do in the application; MS made the paperclip annoying.
If anything, this is the version that's standards compliant that MS will start from and extend.
I was a TA for a compiler class last year (think really big programs that need to be graded), and we decided we'd only officially support one version of the compiler and library and interpreter (we used Java). This was essentially so that we wouldn't have to deal with incompatabilities between environments, and, in particular, the situation where the code works for the student but we can't get it to work.
It's pretty much impossible to avoid this sort of thing-- otherwise, you make the TAs find a different type of machine to grade some of the projects, which just makes a process that is already very hard and slow even worse.
The choice of official platform is, of course, very important. If some of the students don't have access to the official platform, it's going to be very difficult for them to take the class; while they may not actually have any problems, they can't tell if their program will work for the TAs. We chose a version present on most of the public workstations, because we knew that any student concerned about compatibility issues would be able to try their code on the platform we were using.
Of course, when I took the class, I used the pizza compiler which came with the version of Kaffe I was using, and it worked fine. My group used a bunch of different compilers and make sure they all agreed.
On the other hand, C++ has some specification issues; the language specification actually doesn't say what happens in many situations, and different compilers treat it differently. (xevil, written in C++, actually works differently depending on the compiler used to build it) So be sure that you don't do anything dangerous like that, and make sure you're only doing things actually specified by the language.
ADD is generally an inability to concentrate when you want to-- you can concentrate sometimes (often better than most people), but not on what you choose to focus on.
This tends to mean that people with ADD have more good ideas, since they shift easily and hyperfocus, but their idea tend to be less useful to them, because they tend to be about something other than what they're intending to think about.
Learning to control their focus when they want to can be very helpful for this reason; in some situations it's better not to have ADD, while in others it's beneficial. Being able to switch under your own control is really helpful.
Just because it can be controlled without drugs doesn't mean it's not a neurochemical disorder, either. A large portion of what the brain does is produce drugs for itself, so being able to control it mentally may involve producing the right neurotransmitters to get a more normal mix.
At least one common form of dyslexia, caused in part, ironically, by some video games, makes it nearly impossible to read, but can be fixed by covering parts of the text that aren't being read. Still, the problem manifests itself in ways that can be measured objectively (the person is trained to respond to somewhere next to where they're looking, i.e., the place stuff will happen next in a side-scrolling video game).
Somebody's really got to enter a program in purely functional C. Or, at least, C without functions with side effects (only modifying temporary variables). It should be possible, although clearly harder than your average C style. Probably be pretty hard to make it readable, but it would have to get style points for making it contest-relevant.
Unfortuantely, anything I could consider doing in 72 hours in purely functional C would be really easy for just about anyone with a more sensible environment.
I think the most important point he makes in this article is that Napster gives people a chance to develop taste individually; they can just get a bunch of things to try, and listen to music they actually like. There's not the advertising that gets done in the traditional channels telling people what they should like. In fact, there are parody songs and mixes which are actually reasonably well known (e.g., the Eminem+other stuff mixes) which probably haven't even got spare-time ads.
It's a very different world where people only have to invest the time of listening to try something new. There's much more pressure to have the music actually be good, because that's all the exposure it'll get; only word of mouth is a fast enough advertisement when you listen to whatever you happen to want to at the moment. Of course, there still needs to be an incentive for people to make their music do well, but that'll certainly come eventually-- whether tour sales, nicely packaged CDs, or direct fan support.
www.danni.com is running "Apache/1.3.12 (Unix)",
www.cindymargolis.com is running "Microsoft-IIS/4.0".
Just found it funny to notice...
IIRC, they originally got this added to the bill *after* it was passed, which tends to indicate that congress actually wanted not to have this in there (otherwise they would have gotten actual congressmen to insert it). Once it become obvious how it got there and the government was taking notice of them, the RIAA probably felt that congress was likely to repeal that provision, and possibly do something nasty to them in retribution, and so they should try to get rid of the evidence.
After all, having congress know you've defrauded them and stand by your actions is probably not a good way to ensure your future existence.
It seems to me that what we've seen so far of C# has been language features and syntax. What we haven't seen is the main library, which is presumably the part that ties to whole thing to Windows.
We ought to be able to pretty simply modify a Java compiler to compile C# code that uses the standard Java libraries instead of MS ones and produces slightly stylized Java bytecode. Of the language features that seem important, none of them seem like anything a compiler couldn't handle and get the standard Java libraries to support. If we hacked a bytecode interpreter sufficiently, we could probably even get the efficiency gains that C# is supposed to give.
It would be really amusing if there was a working C# for linux available before there was one for Windows. Ignoring, of course, all the Windows-specific parts that will probably take a while to implement anyway.
I just tried it and found that it worked quite well, especially considering that I wasn't yet used to it at all.
The thing that I like better about graffitti, though, is that I can do it reasonably accurately without looking at all towards the device, so I can take notes while continuing to watch the speaker, or enjoy the view while writing. Does your aim get sufficiently good that you don't need to look to type?
Much of that can be configured (at least with X) by either changing or configuring the window manager, or by running apps that have those features.
I'd like to see interface plugins that work at a more application-internal level. For instance, the application could specify the menu items, but some configurations might put some of them additionally in a tool bar, and some might organize them into menus differently (is "Stop Loading" View, Go, Edit, or something else? Are those even different menus? Are "Cut" and "Paste" in menus at all, or just done with the mouse?).
File management generally fits in this category; the application should just say it wants a file name and what sort of guidelines there should be (should it exist already? Be a certain type of file? Have a certain extension?), and the plugin widget would handle getting the user's choice.
Having used Linux almost exclusively for years, I have found that I managed to unlearn some of the other GUI elements. There was one time I sat down at either a Mac or a Windows box and couldn't figure out why there were icons for programs that weren't running and none for programs that were running. The concept of having the filesystem visible in windows and in the background was just one I had entirely forgotten about. I've had times when I couldn't understand that it's okay that you can't iconify windows on Macs.
So there really is quite a bit of variation in GUIs, enough to significantly confuse people who aren't used to the particular style they are using.