It seems like search engines are generally going to lag slightly in censoring things behind how they index the web, so people searching for subversive information will tend to get only the most recent bunch of sites to come up, but they'll have an easier time getting information than if they didn't have search engines or if they only had search engines that included only sites previously approved by the government.
The real concern should be that search engines will keep records of searches and turn them over to the government. Google seems unwilling to do this (while other companies seem quite willing). And, of course, government-sponsored or locally-based engines will be more likely to turn over their data than foreign ones.
If this senator really wanted to help the Chinese people rather than the Chinese government, he'd propose banning keeping search records or revealing this information to outside parties.
It seems to me like most OSS projects reach a state where almost all of the code is written by people employed to do so. This is because it becomes advantageous for businesses to make sure that the projects don't get abandoned, and developers who are doing it as a full-time job just spend more time writing code for the project than other people do. What probably matters more is whether everybody is employed by the same business. The ideal situation is something like Linux, where lots of companies which use Linux or make equipment that Linux could run on employ developers, or the original Apache, where the developers were employed in positions which would require using the software.
The issue is when you listen to "Parabol" followed by "We Are The Champions" in a mix, and there's no silence at the end of the first track or beginning of the second, and you get a loud pop instead, because the waveforms don't line up across the change. The worst case would be if you had two tracks that were originally together, but one of them had gotten the sign reversed; you'd get a pop twice as loud as the music. The real purpose of having silence at the edges of audio is that all silence is the same (since they filter out DC bias), so the jump from one silent section to another silent section doesn't cause a pop.
So the tricky thing that the iPod doesn't handle is figuring out that the current track runs to the end, that the next track runs from the beginning, and, most importantly, that these two particular tracks fit together. It's not impossible to do automatically, but it's a bit computationally expensive, and is a somewhat different operation than the other ones going on.
Of course, there doesn't have to be half a second there; 50 ms would probably be sufficient, but that would give something that sounds like a skip, rather than a moment of silence, which could be even more annoying.
The original question, however, said they were considering standardizing on a single language, which is a dumb idea. It sounds to me like they're conflating the good idea of having standards with the bad idea of having a standard that doesn't respond to requirements.
Actually, the blurb on Slashdot talks about ODF, but the actual article almost exclusively talks about OpenOffice. The IBM statement is only about ODF, but Suter also says that his earlier email about OpenOffice was entirely non-committal; IBM's not holding anything back for copyright ownership reasons. Furthermore, the OpenOffice project lead quoted in both articles actually says that he doesn't think Sun should spin off OpenOffice to a foundation. He says that, if IBM wanted them to (which is not the case), maybe that would be a good idea. IBM's statement is probably directly mostly at the situation where Sun decided they didn't want to employ 80% of the OpenOffice developers any more, and were spinning it off to a foundation for that reason; in this case, IBM would want to talk to them, probably to work something out where the project membership is maintained by IBM hiring the developers.
I'm a bit mystified that Andy Updegrove, when writing the blurb, failed to write it to cover the overall subject material, after covering it accurately in the linked article. I wouldn't be surprised if he'd written it as a full article, and the slashdot editors cut it down for slashback to a portion that wasn't a good summary.
First of all, nobody knows what the GPLv3 will say about DRM. What's been released is only a draft, and the purpose of releasing a draft was to get feedback on it. Enough people are unhappy with exactly what the draft says about DRM and how it is phrased that the final version will almost certainly be different in some way (or the GPLv3 will be seen as not having taken into account community needs, and quickly become irrelevant).
As far as I can tell, nothing in the GPLv3 would prevent Bluehat from releasing a signature for a binary without giving the signing key. (I doubt that a court would even find that the hash in a signature is a copyrightable work, let alone a derived work, so the license on the binary doesn't matter to what they can do with the signature.) Certainly nothing would prevent Mactel from producing a motherboard that would only run binaries signed by Bluehat; they're not dealing with GPL code at all (unless they're checking the signature with GPG in the BIOS, I guess). On the other hand, the GPLv3 would probably prohibit distributing a complete system consisting of the Mactel motherboard, the Bluehat kernel, and the Bluehat signature, without the Bluehat private key.
Your claim: So GPLV3 effectively prevents digital signatures from being used to determine if a binary may be from a source the user trusts!
is wrong, however. It prevent signatures from being used to determine if a binary is from a source the vendor trusts. The user can refuse the GPL and ignore it entirely, and be legally fine with fair use rights to use of the binaries. The user can't distribute the resulting system with Bluehat's public key built in, but that's a good thing; the recipient should be allowed to decide that Bluehat is an evil organization and only Boydriva is trustworthy.
You should know it's Mozilla-based because the name ends with "bird". Of course, they'll probably have to rename it to "Songfox", which should clarify things further. (For that matter, foxes are cuter than birds.)
Google might benefit from being a desirable ISP, but they'd lose big from being an undesirable ISP, and blocking access to sites would make them undesirable with the early-adopter and high-tech crowd that they generally depend on to popularize their stuff. I'd think it would be much more likely that they'd explicitly market an ISP service as being not tiered and not blocking anything. Of course, on their network, their servers would naturally be one hop away from their users, so they'd obviously be faster than anything else.
The real benefit isn't the durability, it's that the total size of the thing is less than the size of the minimum area needed to provide visual feedback for interaction to the user. I wouldn't be too surprised if the shuffle got a tiny screen with the current track info scrolling across, or if it got even smaller, but it appeals almost exclusively to a market that wouldn't be interested in a nano, even if the nano had infinite capacity and were indestructable and free.
Realistically, a rogue state with a nuke wanting to hit the US would probably use black market connections to get it on a cargo ship in a US harbor. Smuggling something into the US by conventional means is far more reliable than an untested long-range missile, assuming you're only sending a few at most.
Lacking Romero, id has released a bunch of uninspired games that were entertaining. With Carmack's engine-writing capabilities, they end up with games where you can do all sorts of neat things, but have no reason to do them. Except that they're neat things to do, even without any in-game motivation. With Romero's game-writing capabilities, Ion Storm's released a bunch of inspired games make pointless by technical limitations. This is a worse failure mode for a game.
Re:Obesity comes from a simple condition...
on
Obesity Contagious?
·
· Score: 1
I'm personally mostly focused on the "eat well" part. I think there's too much focus on "eat less", which ignores the fact that a lot of people are malnourished eating as much as they do, simply because they aren't getting any of the nutrients that don't contribute to flavor, because their diet is largely synthetic (either chemically or by breeding), and the producers don't bother with stuff you don't taste. So they get too many calories and are still hungry, because they still haven't had a complete meal.
There's also a tricky balance on the calorie side: if you eat less, your blood sugar drops, you have less energy, and you use less. The basic thermodynamic point doesn't take into account the fact that a lot of the output side is resting metabolism, which is not under conscious control, and is partially based on routinely feeling full. You can't really model the body like a passive system, because it's affected by internal feedback. And, really, it shouldn't take willpower to avoid getting fat; eating good food should make you feel full after a reasonable amount, and should make you sufficiently active to burn off that much food.
In any case, I was noticing a lot of posts that were unrealistic about the off the wall, but that most of the posts which were accurate were missing the point of the original article, which was essentially that we didn't evolve to be infected with adenoviruses, and if our natural impulses are making us unhealthy, there's a reasonable chance that something's interfering with them.
Maybe I give you a session where your shipping address is my house. You buy some stuff from the site, don't pay much attention, and have it shipped to me. Or I give tons of people shopping carts containing something I sell through Amazon. Some of the people don't pay attention and accidentally include my item in their next purchase. Or I create a session with a site but don't log in, and I give it to them. They use it, find they aren't logged in, log in, and I'm also logged in as them (since I'm using the same session).
There was some news story where the mainstream sources mostly had the wording of some critical quotation wrong in various ways (which is actually generally true of mainstream news quotations, since they come from reporters quickly writing something down when it's said, not recordings; they usually get the right meaning, but rarely the right words). Surprisingly, Wikinews almost alone had the quotation exactly right (i.e., perfectly matching the available audio recording of the event). But the weather map that day was a picture of some guy's butt, a mistake that none of the other media sources made.
Re:Obesity comes from a simple condition...
on
Obesity Contagious?
·
· Score: 4, Informative
If you eat less than your body requires, by definition, you die. Being able to eat less and lose weight instead of starving while fat depends on the ability to get blood sugar out of fat, which is compromised in people with a number of conditions. It also depends on having a low-calorie source of non-energy nutrients, which is often expensive. It also depends on being able to maintain a reasonable blood sugar level without more energy being taken out as fat.
For people without a medical condition that causes obesity, it is possible to take in fewer calories and run off of fat instead. But there are a number of medical conditions which can interfere with this process, which depends on a non-trivial cascade of signals between different organs (something has to detect that your blood sugar is low; it has to release a hormone in response; the fat cells have to respond to this hormone; they have to produce sugar from fat; the fat cells have to stop pulling sugar out of the bloodstream and storing the energy). This research found that some people are obese because of a particular virus. Of course, most of the people they looked at probably just eat too much, but not everybody.
The big databases seem to do better with tons of reads and writes. They also tend to have extensions that let you do things to make certain types of complicated queries a lot faster (such as "for each month, give me the number of rows with a date column in that month and the total of the value column, also the total number of rows, and the total of the value column. But only include rows where the owner column is this value.")
The reason that they're releasing free versions is for developers whose software is used by big companies. If you were working on software for some big company and the database companies didn't offer free versions, you couldn't provide a solution using an expensive database, because you wouldn't be able to test it without paying a lot of money that you won't want to. Then the big company doesn't have a reason to buy the expensive database, because their applications aren't tested with it and probably won't run any faster. When I was working for a company that made a web application intended to be deployed internally by big companies, we used half a dozen free copies of Oracle, and probably wouldn't have used Oracle if these hadn't been available. Of course, these free copies of Oracle were running on workstations and only dealing with example data, not real customer data, so, from Oracle's point of view, the weren't actually doing any database work.
So what benefit do you see to the Chinese people from not having any access to Google at all? It's not like the Chinese government would collapse without the ability to search the web, or like the Chinese people would rise up and overthrow the government if only they didn't have search engines. Chances are that dissidents would benefit more from a censored search engine than the government would, just because miscellaneous information is more useful for trying to do things than trying to prevent them.
I don't think this is really a staffer job. The staffers generally deal with writing replies to questions the pol has told the staff answers to. The point is that congressional offices receive a whole lot of similar letters, and having the politican actually write each response by hand isn't efficient and wouldn't actually improve the responses over having staffers write them. But blogs are probably generally written by the politicians (and probably editted by their speachwriters or something), because the politician only has to write the post once. For that matter, it makes sense for Kerry to write the blog himself, and then the staffers can read the blog and write replies to paper mail. The thing about blogs is that bloggers aren't expected to reply individually to every comment; they tend to write a next post that covers what people are generally asking, and they respond individually (but publicly) only to particularly interesting and unusual comments. Probably Kerry has staffers looking through comments on his blog for comments that would be good for him to handle, but otherwise, the blog medium serves much of the communicative purpose of staffers.
He hasn't really looked at the GPLv3 yet, which is just as well, because it will probably change in various ways before it becomes official. I wouldn't put too much stock in what he's saying at this point either being well-informed or long-term relevant. Once the GPLv3 has been revised and finally released, he'll presumably look at what it actually means, and make a decision about what he wants to do with his code.
As it's presently written, you're not allowed to distribute a signed binary, something that checks the signature, and only the public key. I expect that it will be rewritten to permit this so long as the recipient is able to change what public key is checked against.
Reading his message, it seems to me that he isn't against prohibiting works which are derived from the kernel and include public keys (he says private keys, but he must mean public keys, since you can easily generate the public key for a private key that's included in something) which are used to check that the whole thing is signed by a private key that isn't included. I suspect that this will turn out to be the effect the GPLv3 has when it's revised (because otherwise, no distribution that uses binary packages can use the GPLv3 and be secure).
Well, git actually tries to be sufficiently customizable that you can use it however you want, and it's just that the first set of git scripts people would actually use were done by Linus for his own use. Then they were followed by Jeff Garzik doing his own scripts (he maintains a ton of little trees, rather than one big one), and more people doing scripts for the use of individual developers, users who want to help with debugging, and now people with entirely different sorts of projects. It's following the UNIX way, based on the idea that, if you give people all the parts, they'll be able to build a better selection of things than you could build for them. You don't have to solve all problems, you just have to make sure that you don't rule out solving any, and someone will solve each problem when it comes up.
They're actually significantly more productive using git than they were using BitKeeper. To some extent, this is because more people are comfortable using git, so there's more uniformity of process. To some extent, this is because git is faster for some critical processes. To some extent, this is because people have tools for git tuned for their own use (because they can). To some extent, this is because people continue to work on the maintainability of the kernel, so productivity improves over time, tools aside.
As far as I can tell, the switch took a lot of Linus's attention, so nothing got done on putting changes in for a month, but development continued approximately as before, and then there was a period where Linus was applying patches blazingly fast, because they'd been developed and tested while he was doing git (and he designed git so he could apply and commit patches faster than 1/second).
For that matter, who other than Apple sells a machine you can run Linux on that has the whole computer in a flat panel display? Just because you're a Linux user doesn't mean you're not willing to pay more for hardware so that you can have more space on your desk. Apple's stuff is also quiet, and I bet the portion of Linux users who sleep in a room with a running computer is higher than that for other operating systems.
He specifically avoided making any requirements on documents submitted to his department. The policy explicitly states that you can use whatever you want, and they'll do their best to deal with it. His mandate included making recommendations for what people in general use, and he decided that the correct thing to do was to tell people to continue using the software and formats they were already using. The policy only applies to the documents the executive department produces, and says that they should be in a standard format, not whatever the person preparing them feels like.
If you really want the choice of what format documents are in, why don't you demand that Slashdot post their stories in DOC format, too? Slashdot is forcing you to use HTML right now.
For that matter, the normal thing will be for them to use ODF documents internally, but send out PDFs, since they aren't trying to make the documents they publish easy to edit. So use you be complaining that Slashdot's back end doesn't store their stories in DOC.
The issue in this case is that the general networking maintainer doesn't get excited about wireless, and until a few weeks ago, there wasn't a wifi maintainer. So there's actually a lot of support for wifi written, but it hasn't gotten into the kernel, because none of the drivers are quite up to the maintainability standards of the kernel, and nobody'd been organizing fixing this.
However, there's now a new maintainer for wifi, and he's been arranging discussion on getting all of this cleaned up and into the kernel, so it should change soon.
Essentially, if the company claims to its shareholders that it can do something IP-related, and they are wrong, that's now securities fraud. Think of it this way: if a company doesn't have enough licenses for their commercial software, and the BSA knocks down their door and takes their money, their shareholders can claim that they were led by the company to believe that it wasn't going to get hit by this sort of thing, and so the SEC steps in and punishes the company for misleading the investors. Similarly, if the company claims that it can have a particular business model, and tells investors that, and it turns out that the GPL prohibits acting like that, not only do the copyright holders of the code the company is using get to make the company shape up, but the company's investors (as represented by the SEC) get to punish the company.
Of course, simply using GPL software doesn't require that you accept the GPL, since when people distribute software under the GPL, they actually give you ownership of copies of the software, which you can make use of yourself without a license at all. So most businesses have no need to accept the GPL and aren't at risk of messing up.
Also, the owners of copyrights on GPL code are generally pretty nice about it. They tend to demand from violators things that aren't really that big a deal; most often, the company has to release a few minor changes, and has to arrange linking such that the GPL-covered parts can be replaced by users who don't have source to the proprietary parts (i.e., the proprietary parts are often either scripts, or not linked, or the OSS parts are LGPL). Much of the time, the only requirement is that the company admit to its customers that they didn't actually write any software of their own. But the investors, on the other hand, have invested in a proprietary software company with a particular business model, and may be inordinately unhappy to discover that the company isn't what they thought. So the SOX settlement may actually be much more damaging to the company than the GPL settlement.
Slashdot Mirror
It seems like search engines are generally going to lag slightly in censoring things behind how they index the web, so people searching for subversive information will tend to get only the most recent bunch of sites to come up, but they'll have an easier time getting information than if they didn't have search engines or if they only had search engines that included only sites previously approved by the government.
The real concern should be that search engines will keep records of searches and turn them over to the government. Google seems unwilling to do this (while other companies seem quite willing). And, of course, government-sponsored or locally-based engines will be more likely to turn over their data than foreign ones.
If this senator really wanted to help the Chinese people rather than the Chinese government, he'd propose banning keeping search records or revealing this information to outside parties.
It seems to me like most OSS projects reach a state where almost all of the code is written by people employed to do so. This is because it becomes advantageous for businesses to make sure that the projects don't get abandoned, and developers who are doing it as a full-time job just spend more time writing code for the project than other people do. What probably matters more is whether everybody is employed by the same business. The ideal situation is something like Linux, where lots of companies which use Linux or make equipment that Linux could run on employ developers, or the original Apache, where the developers were employed in positions which would require using the software.
The issue is when you listen to "Parabol" followed by "We Are The Champions" in a mix, and there's no silence at the end of the first track or beginning of the second, and you get a loud pop instead, because the waveforms don't line up across the change. The worst case would be if you had two tracks that were originally together, but one of them had gotten the sign reversed; you'd get a pop twice as loud as the music. The real purpose of having silence at the edges of audio is that all silence is the same (since they filter out DC bias), so the jump from one silent section to another silent section doesn't cause a pop.
So the tricky thing that the iPod doesn't handle is figuring out that the current track runs to the end, that the next track runs from the beginning, and, most importantly, that these two particular tracks fit together. It's not impossible to do automatically, but it's a bit computationally expensive, and is a somewhat different operation than the other ones going on.
Of course, there doesn't have to be half a second there; 50 ms would probably be sufficient, but that would give something that sounds like a skip, rather than a moment of silence, which could be even more annoying.
The original question, however, said they were considering standardizing on a single language, which is a dumb idea. It sounds to me like they're conflating the good idea of having standards with the bad idea of having a standard that doesn't respond to requirements.
Actually, the blurb on Slashdot talks about ODF, but the actual article almost exclusively talks about OpenOffice. The IBM statement is only about ODF, but Suter also says that his earlier email about OpenOffice was entirely non-committal; IBM's not holding anything back for copyright ownership reasons. Furthermore, the OpenOffice project lead quoted in both articles actually says that he doesn't think Sun should spin off OpenOffice to a foundation. He says that, if IBM wanted them to (which is not the case), maybe that would be a good idea. IBM's statement is probably directly mostly at the situation where Sun decided they didn't want to employ 80% of the OpenOffice developers any more, and were spinning it off to a foundation for that reason; in this case, IBM would want to talk to them, probably to work something out where the project membership is maintained by IBM hiring the developers.
I'm a bit mystified that Andy Updegrove, when writing the blurb, failed to write it to cover the overall subject material, after covering it accurately in the linked article. I wouldn't be surprised if he'd written it as a full article, and the slashdot editors cut it down for slashback to a portion that wasn't a good summary.
First of all, nobody knows what the GPLv3 will say about DRM. What's been released is only a draft, and the purpose of releasing a draft was to get feedback on it. Enough people are unhappy with exactly what the draft says about DRM and how it is phrased that the final version will almost certainly be different in some way (or the GPLv3 will be seen as not having taken into account community needs, and quickly become irrelevant).
As far as I can tell, nothing in the GPLv3 would prevent Bluehat from releasing a signature for a binary without giving the signing key. (I doubt that a court would even find that the hash in a signature is a copyrightable work, let alone a derived work, so the license on the binary doesn't matter to what they can do with the signature.) Certainly nothing would prevent Mactel from producing a motherboard that would only run binaries signed by Bluehat; they're not dealing with GPL code at all (unless they're checking the signature with GPG in the BIOS, I guess). On the other hand, the GPLv3 would probably prohibit distributing a complete system consisting of the Mactel motherboard, the Bluehat kernel, and the Bluehat signature, without the Bluehat private key.
Your claim:
So GPLV3 effectively prevents digital signatures from being used to determine if a binary may be from a source the user trusts!
is wrong, however. It prevent signatures from being used to determine if a binary is from a source the vendor trusts. The user can refuse the GPL and ignore it entirely, and be legally fine with fair use rights to use of the binaries. The user can't distribute the resulting system with Bluehat's public key built in, but that's a good thing; the recipient should be allowed to decide that Bluehat is an evil organization and only Boydriva is trustworthy.
You should know it's Mozilla-based because the name ends with "bird". Of course, they'll probably have to rename it to "Songfox", which should clarify things further. (For that matter, foxes are cuter than birds.)
Google might benefit from being a desirable ISP, but they'd lose big from being an undesirable ISP, and blocking access to sites would make them undesirable with the early-adopter and high-tech crowd that they generally depend on to popularize their stuff. I'd think it would be much more likely that they'd explicitly market an ISP service as being not tiered and not blocking anything. Of course, on their network, their servers would naturally be one hop away from their users, so they'd obviously be faster than anything else.
The real benefit isn't the durability, it's that the total size of the thing is less than the size of the minimum area needed to provide visual feedback for interaction to the user. I wouldn't be too surprised if the shuffle got a tiny screen with the current track info scrolling across, or if it got even smaller, but it appeals almost exclusively to a market that wouldn't be interested in a nano, even if the nano had infinite capacity and were indestructable and free.
Realistically, a rogue state with a nuke wanting to hit the US would probably use black market connections to get it on a cargo ship in a US harbor. Smuggling something into the US by conventional means is far more reliable than an untested long-range missile, assuming you're only sending a few at most.
Lacking Romero, id has released a bunch of uninspired games that were entertaining. With Carmack's engine-writing capabilities, they end up with games where you can do all sorts of neat things, but have no reason to do them. Except that they're neat things to do, even without any in-game motivation. With Romero's game-writing capabilities, Ion Storm's released a bunch of inspired games make pointless by technical limitations. This is a worse failure mode for a game.
I'm personally mostly focused on the "eat well" part. I think there's too much focus on "eat less", which ignores the fact that a lot of people are malnourished eating as much as they do, simply because they aren't getting any of the nutrients that don't contribute to flavor, because their diet is largely synthetic (either chemically or by breeding), and the producers don't bother with stuff you don't taste. So they get too many calories and are still hungry, because they still haven't had a complete meal.
There's also a tricky balance on the calorie side: if you eat less, your blood sugar drops, you have less energy, and you use less. The basic thermodynamic point doesn't take into account the fact that a lot of the output side is resting metabolism, which is not under conscious control, and is partially based on routinely feeling full. You can't really model the body like a passive system, because it's affected by internal feedback. And, really, it shouldn't take willpower to avoid getting fat; eating good food should make you feel full after a reasonable amount, and should make you sufficiently active to burn off that much food.
In any case, I was noticing a lot of posts that were unrealistic about the off the wall, but that most of the posts which were accurate were missing the point of the original article, which was essentially that we didn't evolve to be infected with adenoviruses, and if our natural impulses are making us unhealthy, there's a reasonable chance that something's interfering with them.
Maybe I give you a session where your shipping address is my house. You buy some stuff from the site, don't pay much attention, and have it shipped to me. Or I give tons of people shopping carts containing something I sell through Amazon. Some of the people don't pay attention and accidentally include my item in their next purchase. Or I create a session with a site but don't log in, and I give it to them. They use it, find they aren't logged in, log in, and I'm also logged in as them (since I'm using the same session).
There was some news story where the mainstream sources mostly had the wording of some critical quotation wrong in various ways (which is actually generally true of mainstream news quotations, since they come from reporters quickly writing something down when it's said, not recordings; they usually get the right meaning, but rarely the right words). Surprisingly, Wikinews almost alone had the quotation exactly right (i.e., perfectly matching the available audio recording of the event). But the weather map that day was a picture of some guy's butt, a mistake that none of the other media sources made.
If you eat less than your body requires, by definition, you die. Being able to eat less and lose weight instead of starving while fat depends on the ability to get blood sugar out of fat, which is compromised in people with a number of conditions. It also depends on having a low-calorie source of non-energy nutrients, which is often expensive. It also depends on being able to maintain a reasonable blood sugar level without more energy being taken out as fat.
For people without a medical condition that causes obesity, it is possible to take in fewer calories and run off of fat instead. But there are a number of medical conditions which can interfere with this process, which depends on a non-trivial cascade of signals between different organs (something has to detect that your blood sugar is low; it has to release a hormone in response; the fat cells have to respond to this hormone; they have to produce sugar from fat; the fat cells have to stop pulling sugar out of the bloodstream and storing the energy). This research found that some people are obese because of a particular virus. Of course, most of the people they looked at probably just eat too much, but not everybody.
The big databases seem to do better with tons of reads and writes. They also tend to have extensions that let you do things to make certain types of complicated queries a lot faster (such as "for each month, give me the number of rows with a date column in that month and the total of the value column, also the total number of rows, and the total of the value column. But only include rows where the owner column is this value.")
The reason that they're releasing free versions is for developers whose software is used by big companies. If you were working on software for some big company and the database companies didn't offer free versions, you couldn't provide a solution using an expensive database, because you wouldn't be able to test it without paying a lot of money that you won't want to. Then the big company doesn't have a reason to buy the expensive database, because their applications aren't tested with it and probably won't run any faster. When I was working for a company that made a web application intended to be deployed internally by big companies, we used half a dozen free copies of Oracle, and probably wouldn't have used Oracle if these hadn't been available. Of course, these free copies of Oracle were running on workstations and only dealing with example data, not real customer data, so, from Oracle's point of view, the weren't actually doing any database work.
So what benefit do you see to the Chinese people from not having any access to Google at all? It's not like the Chinese government would collapse without the ability to search the web, or like the Chinese people would rise up and overthrow the government if only they didn't have search engines. Chances are that dissidents would benefit more from a censored search engine than the government would, just because miscellaneous information is more useful for trying to do things than trying to prevent them.
I don't think this is really a staffer job. The staffers generally deal with writing replies to questions the pol has told the staff answers to. The point is that congressional offices receive a whole lot of similar letters, and having the politican actually write each response by hand isn't efficient and wouldn't actually improve the responses over having staffers write them. But blogs are probably generally written by the politicians (and probably editted by their speachwriters or something), because the politician only has to write the post once. For that matter, it makes sense for Kerry to write the blog himself, and then the staffers can read the blog and write replies to paper mail. The thing about blogs is that bloggers aren't expected to reply individually to every comment; they tend to write a next post that covers what people are generally asking, and they respond individually (but publicly) only to particularly interesting and unusual comments. Probably Kerry has staffers looking through comments on his blog for comments that would be good for him to handle, but otherwise, the blog medium serves much of the communicative purpose of staffers.
He hasn't really looked at the GPLv3 yet, which is just as well, because it will probably change in various ways before it becomes official. I wouldn't put too much stock in what he's saying at this point either being well-informed or long-term relevant. Once the GPLv3 has been revised and finally released, he'll presumably look at what it actually means, and make a decision about what he wants to do with his code.
As it's presently written, you're not allowed to distribute a signed binary, something that checks the signature, and only the public key. I expect that it will be rewritten to permit this so long as the recipient is able to change what public key is checked against.
Reading his message, it seems to me that he isn't against prohibiting works which are derived from the kernel and include public keys (he says private keys, but he must mean public keys, since you can easily generate the public key for a private key that's included in something) which are used to check that the whole thing is signed by a private key that isn't included. I suspect that this will turn out to be the effect the GPLv3 has when it's revised (because otherwise, no distribution that uses binary packages can use the GPLv3 and be secure).
Well, git actually tries to be sufficiently customizable that you can use it however you want, and it's just that the first set of git scripts people would actually use were done by Linus for his own use. Then they were followed by Jeff Garzik doing his own scripts (he maintains a ton of little trees, rather than one big one), and more people doing scripts for the use of individual developers, users who want to help with debugging, and now people with entirely different sorts of projects. It's following the UNIX way, based on the idea that, if you give people all the parts, they'll be able to build a better selection of things than you could build for them. You don't have to solve all problems, you just have to make sure that you don't rule out solving any, and someone will solve each problem when it comes up.
They're actually significantly more productive using git than they were using BitKeeper. To some extent, this is because more people are comfortable using git, so there's more uniformity of process. To some extent, this is because git is faster for some critical processes. To some extent, this is because people have tools for git tuned for their own use (because they can). To some extent, this is because people continue to work on the maintainability of the kernel, so productivity improves over time, tools aside.
As far as I can tell, the switch took a lot of Linus's attention, so nothing got done on putting changes in for a month, but development continued approximately as before, and then there was a period where Linus was applying patches blazingly fast, because they'd been developed and tested while he was doing git (and he designed git so he could apply and commit patches faster than 1/second).
For that matter, who other than Apple sells a machine you can run Linux on that has the whole computer in a flat panel display? Just because you're a Linux user doesn't mean you're not willing to pay more for hardware so that you can have more space on your desk. Apple's stuff is also quiet, and I bet the portion of Linux users who sleep in a room with a running computer is higher than that for other operating systems.
He specifically avoided making any requirements on documents submitted to his department. The policy explicitly states that you can use whatever you want, and they'll do their best to deal with it. His mandate included making recommendations for what people in general use, and he decided that the correct thing to do was to tell people to continue using the software and formats they were already using. The policy only applies to the documents the executive department produces, and says that they should be in a standard format, not whatever the person preparing them feels like.
If you really want the choice of what format documents are in, why don't you demand that Slashdot post their stories in DOC format, too? Slashdot is forcing you to use HTML right now.
For that matter, the normal thing will be for them to use ODF documents internally, but send out PDFs, since they aren't trying to make the documents they publish easy to edit. So use you be complaining that Slashdot's back end doesn't store their stories in DOC.
The issue in this case is that the general networking maintainer doesn't get excited about wireless, and until a few weeks ago, there wasn't a wifi maintainer. So there's actually a lot of support for wifi written, but it hasn't gotten into the kernel, because none of the drivers are quite up to the maintainability standards of the kernel, and nobody'd been organizing fixing this.
However, there's now a new maintainer for wifi, and he's been arranging discussion on getting all of this cleaned up and into the kernel, so it should change soon.
Essentially, if the company claims to its shareholders that it can do something IP-related, and they are wrong, that's now securities fraud. Think of it this way: if a company doesn't have enough licenses for their commercial software, and the BSA knocks down their door and takes their money, their shareholders can claim that they were led by the company to believe that it wasn't going to get hit by this sort of thing, and so the SEC steps in and punishes the company for misleading the investors. Similarly, if the company claims that it can have a particular business model, and tells investors that, and it turns out that the GPL prohibits acting like that, not only do the copyright holders of the code the company is using get to make the company shape up, but the company's investors (as represented by the SEC) get to punish the company.
Of course, simply using GPL software doesn't require that you accept the GPL, since when people distribute software under the GPL, they actually give you ownership of copies of the software, which you can make use of yourself without a license at all. So most businesses have no need to accept the GPL and aren't at risk of messing up.
Also, the owners of copyrights on GPL code are generally pretty nice about it. They tend to demand from violators things that aren't really that big a deal; most often, the company has to release a few minor changes, and has to arrange linking such that the GPL-covered parts can be replaced by users who don't have source to the proprietary parts (i.e., the proprietary parts are often either scripts, or not linked, or the OSS parts are LGPL). Much of the time, the only requirement is that the company admit to its customers that they didn't actually write any software of their own. But the investors, on the other hand, have invested in a proprietary software company with a particular business model, and may be inordinately unhappy to discover that the company isn't what they thought. So the SOX settlement may actually be much more damaging to the company than the GPL settlement.