As I recall, when he won the initial case, the prosecutors were saying that they were definitely going to appeal, because he'd admitted to doing what they said he did, and they had thought it was not illegal. In order to have the law interpretted for them, they would have to bring the case to a higher court to get an explanation, so that they would know who to prosecute in the future.
IIRC, he is in less danger in the appeal than he was in the initial case, and there's not expected to be any new arguments or evidence, which means that his trial will be much easier.
First off, what they're talking about is whether to regulate companies which provide access to the telephone network via IP as telephone companies. Chances are that they will, since it's no less a use of the phone network if the call starts halfway there.
On the other hand, people in populous areas (with good network connectivity) may start making more pure VoIP calls and not using the phone network as much. This would lead to the phone network getting out of balance. On the other hand, at that point the government could step in and regulate the data lines, at which point the people in rural areas would get cheap internet, and everybody would just move to IP, which is probably a better design these days anyway.
In order to have a useable site, the user must be able to identify the right next step. The user can probably evaluate 5-9 choices at once, but the user may make multiple choices (looking into smaller areas of the same page) before clicking, assuming that the user gets feedback from the text that the section is appropriate. Of course, the user can go through a long list if each item is clearly either worth trying or not worth trying.
After each click, the user has to get feedback that the click was correct. The second page must look more promising than the first, the third more promising than the second, and so on. Otherwise the user goes back and eventually gives up. It's pretty hard to keep your site encouraging for many clicks.
The pathological case of a site which always responds to the first click by showing an identical page, and to the second click by answering the user's question will never be used successfully, despite being only two clicks. A site which provides information on a piece of computer hardware by model number with one click per character of the model number could probably keep the user clicking for dozens of clicks (JUSTer, JUSTer Active, JUSTer Active Sx-xxx, JUSTer Active SP-xxx,...), because the user feels closer with each click.
These aspects are actually best seen in games and puzzles, where the intent is that it be difficult to succeed, but it simultaneously has to keep the player from giving up. Therefore, the player has to either keep getting farther or keep feeling more skillful.
That's why everyone who's checking on their vote has to be anonymous, and has to therefore get all of the votes. For example, they could publish them all in a book, which you could go to the library to look at, search through, and find their vote, all without revealing which vote was theirs. In any case, it's necessary to publish the complete collection of votes together, or the adversary simply reports your vote correctly to everyone who asks, but actually registers theit choice; if the whole set is published, other people can count up the totals and demonstrate that the official registry doesn't match the real counts.
Otherwise, people who voted a certain way start dying mysteriously, and people don't vote that way any more.
No, according to Linus, after 2.6.0-test11, it's up to Andrew Morton, who hasn't said anything on the subject, to release 2.6.0 (or whatever he wants to make his first version).
In the case of zlib, the library is sufficiently small that some other packages distribute a version of it as part of their own sources, instead of expecting the system to contain it. Furthermore, some packages contain modified versions of zlib to account for different conditions (for example, the linux kernel contained a version of zlib restricted to do only some operations, to keep the kernel from getting too big).
When the bug was found, some other packages had to patch the versions of zlib they contained, but the critical thing to note is that the ones that included zlib could just apply the patch to the older API version they contained, or to the trimmed version, or whatever they had. This meant that people could apply the patch without breaking half of their software, which depended on the particular API they were using.
2.6 isn't 100% userspace-compatible with 2.4; there are a number of utilities which need to be upgraded to deal with 2.6, and a few cases where 2.4 stuff isn't supported at all. So I wouldn't expect all 2.4 installations to be able to go to 2.6 when the time comes. For that matter, 2.4 still has the better ACPI support, and probably still will when 2.6.0 comes out.
As for when 2.6.0 will be out, Linus is turning that over to Andrew Morton, and we really have no idea what his style of stable kernel releases will be like. I'd actually expect to next see a relatively long 2.6.0-rc series before 2.6.0; maybe even a 2.6.0-pre series before that, depending on what he thinks of the seriousness of the remaining "should-fix" and "must-fix" lists and the reported bugs.
Given that SCO has beaten Microsoft in a lawsuit in the past (or, at least, gotten them to settle), I wouldn't be surprised if SCO went after Microsoft again as soon as Microsoft stopped paying them off.
I presume when you say to publish the votes, you mean that you produce a book of ballots that you counted, where each person can recognize their own, but there's no way to identify which was cast by someone else. (If you just published the ballots, using the kinds of ballots used where I've voted, people could count them, and they'd be anonymous, but it would be damn hard to tell yours apart from other people's)
You could have a system where each ballot has a different random symbol on it and is given out with the symbol covered by the original wrapper. The voter marks the candidates, takes a copy of the symbol, covers the whole thing with the wrapper again, and puts it in the box. At the end of the election, the ballots are counted and published. Then each voter can look through the book for the ballot with the right symbol to make sure it is there, and count the totals. If you don't show your symbol to anyone, nobody can tell who you voted for. For that matter, the polls could actually give the voter the opportunity, in cases where they could not read a ballot, to come forward with the copy of their symbol; the voter could sacrifice their anonymity to clarify their vote, if they wanted to (assuming, of course, that the receipts with the symbol also got, say, your name printed on it by way of carbon paper when you signed to get the ballot).
Re:The main issue with XML is performance
on
Effective XML
·
· Score: 1
Care for some column headings for that record? Once you include the column headings, the flat version is 2/3 the size of the XML version. Sure, you don't need another set of column headings for each record, but, since you're talking about a single record, you can't just ignore the column headings.
Furthermore, XML isn't well suited for the case where you have a large number of records with the same column structure. It's not really even worthwhile for all different columns if you don't have nested records. But once you have a complicated structure, XML becomes much more manageable than defining separate tables with ids and references.
Once you get to something like an Apache config file, doing it as a flat file doesn't save you significant space, and makes the whole thing totally unusable.
The thing that I think was a success about the internet in relation to the outage is that there was no loss of connectivity between Boston and California. I actually found out that a huge chunk of the middle of the country had no power from a site on the other side of the outage from me.
As for the people who lost power, I suspect that they were largely more concerned about other things that their communications. When the neighborhood down the street from me lost power, the banks and stores closed. I don't think they closed due to communications issues; they weren't even handling depositing checks (which involves postal mail and takes overnight to clear anyway).
I suspect the reason that so many networks went offline was really that, while the internet is becoming critical for many businesses, it is critical only for things they do while they have power (which is also, obviously, critical). If businesses come to be able to handle power outages (which they might if they become sufficiently common for people to care), they will likely protect their networks against power failure at the same time. The main issue is the upstreams who might not prepare for outages, not realizing that the customers are actually hoping to work through the blackout.
Personally, I don't bother to put hubs and routers on UPS, because I doubt that it would help much. I would if the machines weren't self-sufficient (having been in an outage with a bunch of machines using NFS for things like the shutdown scripts). But, honestly, if the power's out, I'm not going to bother trying to work.
One vote doesn't matter all that much. But you can do better than one vote if you can convince somebody else to vote for your position. If you sit down in a local coffeeshop and convince a couple of people, your own vote is practically insignificant. If you manage to convince your favorite local newspaper, you're starting to sway a number of votes. Same for a union or an organization which would otherwise by neutral on the issue.
The candidates are largely identical, because they all want to be exactly the candidate that people want to vote for. All of the candidates who wouldn't compromise their position to get votes are too far from center to be electable. On the other hand, by maintaining a position convincingly, you can help to make the candidates think that center is closer to your point of view than they would otherwise. In this way, voicing your opinion to your candidate (and convincing other people to agree with you and do so), and responding to polls is more important than the actual vote.
If you want to be cynical, it's quite plausible that the 2000 election came down to the fact that voters were confused by the ballot and were refused help by poll workers. At least in my area, polls seem to be run by volunteers. Think of the impact you could have by volunteering for this. (To be a bit less cynical, the polls which were refusing to help people were evidentally understaffed. Just having more people show up to help might have given a more accurate result, even without those workers intending to affect the result in one direction or another.)
I personally think that there's too much emphasis placed on getting concerned individuals to vote, and too little on getting concerned individuals to get unconcerned individuals concerned. For that matter, our electoral system means that votes in elections which are close have a much larger effect than votes in elections which are not. My vote in the next presidential election won't count for anything, because, no matter what I do, my electoral vote will go to the Democrat. On the other hand, I could affect the outcome by convincing someone in a swing state. (In fact, I make a point of never voting for the candidate who will win if there's someone I prefer who won't win, although I do vote for a major candidate in a close race if there is one).
The reason that devfs is not yet deprecated in favor of udev yet is that udev depends on the kernel using the new driver model for everything that gets a device. Of course, that's not udev work per se, but it means that, for example, input devices (IIRC) don't yet work quite right with it.
The kernel having explicit knowledge of what it's doing in a uniform format is a new feature in 2.6, and it's not completely universal yet. Once that all works correctly, udev should work perfectly, and it is a better design than devfs, because it puts device naming in userspace, but device numbering comes from the kernel, and the kernel tells userspace what each device actually is. This is how the division of labor is supposed to be: the kernel has internal information, which it maintains, and an API, which it defines, but userspace can use that API to specify policy.
The name might, instead, indicate how he got enough time away from his family life to actually release test10.
Since, as you must know, he has a wife and kids. And he is sufficiently absent-minded to forget about the kids, but the wife might be more difficult. (According to a Wired? interview recently)
It might actually be easier to do singing than normal speech, because singing replaces intonation, tempo, and some of stress, all of which otherwise have to be determined from a syntactic and semantic analysis of the text in order to really sound right. There have been people who have learned to sing songs in languages they didn't know at all, while I have yet to hear of someone giving a lecture (as convincingly) in a language they didn't know.
You're missing the ever-popular "boy/girl doesn't love girl/boy", but that's about it. Except, of course, for those rare songs which avoid being cliche entirely by being (ironically) just wrong.
But even if the style isn't new, the melody isn't new, and the instrumentation isn't new, the combination can be new, and it takes skill to make the combination work (how do you play Beethoven's Ninth... on three electric guitars?). It might actually take 5 years with a given fashion to go through all of the melodies, at which point there's a new fashion.
It's certainly possible that classical music got all of the interesting underlying melodies. But the importance of style and instrumentation shouldn't be ignored (let alone lyrics). Interestingly, most currently available recordings of classical music are done by taking the underlying melody and playing it in the most boring way possible, which is rather different from how the scores were originally played. It's a bit like the works of Shakespear being read by a computer in a totally flat voice, except where the script actually specifies that the character whispers or shouts. Popular music is played (and was always played) with a substantial amount of interpretation by the performers, which forms the style.
Furthermore, I think that the style, and, especially, working out how the style and the underlying melody can be resolved, is as significant an application of artistic talent as writing the melody in the first place. It's like translating poetry; it's easy to do a direct translation, but making it actually work as poetry in the target language is at least as hard as writing the original (since you're constrained not only to write a good poem, but you have to also make it match another work in all of the ways that are important, while using entirely different grammer and vocabulary).
In short, even if classical music tried every melody, the existance of new styles and instrumentations means that there will be new complete works.
Between valgrind and ddd, I haven't had any bugs which I couldn't identify in a half hour, and only ones where there is something that the tools are bad at take more than a few minutes. Of course, that doesn't include programs that run fine but produce results which are simply not what I wanted and actually coming up with a way to fix the bug.
Valgrind, in particular, is really promising; the core architecture is now being used to identify potential race conditions (cases where different threads access the same memory without having the same lock), as well as identifying memory errors (including leaks).
What about the new enemies that you get when you kill all of your current enemies? Both the people who liked your current enemies (but not enough to join them before), and the people who are afraid of your ability to kill people that effectively.
The perfect weapon, in my opinion, would demoralize all of your enemies. War is never won by killing people; it's won by demoralizing the people who remain. You've won a war exactly when the enemy agrees that you've won. It's traditional to convince the enemy by killing people, but that's not really necessary, and doesn't necessarily help. (For example, we would have won Vietnam had we quit before we got too close to China; further success made the enemy unwilling to give up)
RSI is actually caused by using your hands while under too much stress. Really bad ergonomics (keyboard on desk while you sit on floor, e.g.) will make your hands hurt, but you can't keep it up long enough to damage yourself long term. Moderately bad ergonomics (standard desk situation) will make it require a certain amount of stress in your life. Really good ergonomics (special keyboard, etc.) might save you if you have to work under a lot of stress.
There are actually trends in stress-related problems. It used to be that people would faint. In the eighties, they had heart attacks. In the nineties (and still now, to a lesser extent), it was RSI. It's an interesting reflection on humanity that people can hear about a common disease and actually reproduce it in themselves, to the point of actually dying of it. But, if you're under stress, something's going to give out, and chances are that it'll be what you're worried about giving out.
So I'm not too sympathetic to RSI sufferers, but it's because I know that twenty years ago, they'd get heart attacks instead. RSI may suck, but at least it won't kill you.
A computer will be intelligent when it is capable of learning things which are entirely new, by applying its knowledge, trying things, general reasoning, and developing new representations. Chess is fine, go is fine, but I'll be impressed when a chess program goes on the internet and learns to play go.
There have been a series of problems that people have posed with the idea that they couldn't be solved without giving the program the ability to figure out the problem. Each time, however, people figure out how to solve the problem without giving the computer general intelligence. But the problems weren't definitions of intelligence ("Intelligence is what is necessary to play chess"), but rather people thought (incorrectly) that the problems couldn't be solved without intelligence.
On the other hand, there is a cluster of problems which have not been solved, despite many attempts, which are not intended to be difficult (which can be overcome eventually with clever coding and fast hardware), but rather which explicitly require undirected learning and the ability to generalize to new applications.
Of course, many of the problems originally posed as measures of intelligence were, in fact, problems that people wanted solutions to, because they were difficult for people. So the effort put into solving these problems without trying to use general intelligence is actually worthwhile.
Considering that this virus was synthesized from scratch, it's probably not something very effective. It's a long way from building a virus that works at all to building a virus that targets a particular organism or overcomes natural defenses.
Just because it's man-made doesn't make it more advanced than naturally-occurring viruses. It's been possible for a long time to build viruses from collected stocks, and these are generally much more frightening. What will be scary is when we have some clue as to how to design proteins, and could construct a virus with specific properties. Until then, we're not likely to create anything that doesn't arise in nature.
(Genetically modified foods are a slightly different issue; just because it might arise in nature doesn't mean we'd eat it if it did. Also, most of the organisms involved are much more resistant to mutation and genetic mixing than viruses and bacteria)
It makes sense that he didn't want to start the third movie with Saruman and finish a major plot in the first 7 minutes of the movie. It's somewhat unfortunate, however, that he didn't think of it last year and put it at the end of the second movie, and have Saruman done with at that point.
He'd already moved the beginning of the second book to the end of the first movie to avoid having a character die in the first scene of the second movie, so it's not unprecedented.
What I'd actually like to see is the whole series done as a single movie, without the changes made to make the breaks work. (One thing that hard about making a book into a movie is that people read books in chapters, and put the book down in between, but watch movies in theaters all together. On a DVD, however, you can watch a few scenes and then go do something else)
It scales perfectly well. I voted in an election with five races, one of which allowed voting for up to four candidates (alderman-at-large). In even years, there are many more spaces, and it still works fine; we have two columns/side of a legal-sized piece of heavy paper, large type, and fill in bars to indicate your choices. It's then read by an optical scanner. This year took a column and a half. IIRC, November 2000 took three and a half or so.
The next town over has a race (city council) where you rank a dozen candidates, and then use paper as well.
The real issue is voltages. Every type of chemical cell has a particular voltage it produces. It's easy to get a multiple of that voltage, but very hard to get any other voltage. Most of the new technologies produce voltages that don't match standard batteries, which means that, unless you have a custom device or a device designed for a range of voltages, you can't use anything new.
Actually, there is one place where battery technology has seen incredible advanced, and that is power tools. Ten years ago, a cordless ("powerless") drill was basically a toy, because it didn't produce enough torque, and the batteries would run out quickly. These days, people just don't use drills with cords, because the battery-powered ones are just as good, and cords are inconvenient. Of course, these use battery packs in the 10-15V range, which is sufficiently wide to handle a lot of variation, and is achieved with several cells, so the number of cells can be varied to change the voltage.
One AA (or AAA) battery is 1.5V; a single lithium cell is ~3V, and lithium ion and polymer are 3.6V. So you can make a newfangled rechargable which is perfect for devices that take 7 AAA batteries, and you can get pair-of-AA-battery lithium disposables, that's about it for new batteries of the sort that you don't get for Christmas.
Slashdot Mirror
As I recall, when he won the initial case, the prosecutors were saying that they were definitely going to appeal, because he'd admitted to doing what they said he did, and they had thought it was not illegal. In order to have the law interpretted for them, they would have to bring the case to a higher court to get an explanation, so that they would know who to prosecute in the future.
IIRC, he is in less danger in the appeal than he was in the initial case, and there's not expected to be any new arguments or evidence, which means that his trial will be much easier.
First off, what they're talking about is whether to regulate companies which provide access to the telephone network via IP as telephone companies. Chances are that they will, since it's no less a use of the phone network if the call starts halfway there.
On the other hand, people in populous areas (with good network connectivity) may start making more pure VoIP calls and not using the phone network as much. This would lead to the phone network getting out of balance. On the other hand, at that point the government could step in and regulate the data lines, at which point the people in rural areas would get cheap internet, and everybody would just move to IP, which is probably a better design these days anyway.
In order to have a useable site, the user must be able to identify the right next step. The user can probably evaluate 5-9 choices at once, but the user may make multiple choices (looking into smaller areas of the same page) before clicking, assuming that the user gets feedback from the text that the section is appropriate. Of course, the user can go through a long list if each item is clearly either worth trying or not worth trying.
...), because the user feels closer with each click.
After each click, the user has to get feedback that the click was correct. The second page must look more promising than the first, the third more promising than the second, and so on. Otherwise the user goes back and eventually gives up. It's pretty hard to keep your site encouraging for many clicks.
The pathological case of a site which always responds to the first click by showing an identical page, and to the second click by answering the user's question will never be used successfully, despite being only two clicks. A site which provides information on a piece of computer hardware by model number with one click per character of the model number could probably keep the user clicking for dozens of clicks (JUSTer, JUSTer Active, JUSTer Active Sx-xxx, JUSTer Active SP-xxx,
These aspects are actually best seen in games and puzzles, where the intent is that it be difficult to succeed, but it simultaneously has to keep the player from giving up. Therefore, the player has to either keep getting farther or keep feeling more skillful.
That's why everyone who's checking on their vote has to be anonymous, and has to therefore get all of the votes. For example, they could publish them all in a book, which you could go to the library to look at, search through, and find their vote, all without revealing which vote was theirs. In any case, it's necessary to publish the complete collection of votes together, or the adversary simply reports your vote correctly to everyone who asks, but actually registers theit choice; if the whole set is published, other people can count up the totals and demonstrate that the official registry doesn't match the real counts.
Otherwise, people who voted a certain way start dying mysteriously, and people don't vote that way any more.
No, according to Linus, after 2.6.0-test11, it's up to Andrew Morton, who hasn't said anything on the subject, to release 2.6.0 (or whatever he wants to make his first version).
In the case of zlib, the library is sufficiently small that some other packages distribute a version of it as part of their own sources, instead of expecting the system to contain it. Furthermore, some packages contain modified versions of zlib to account for different conditions (for example, the linux kernel contained a version of zlib restricted to do only some operations, to keep the kernel from getting too big).
When the bug was found, some other packages had to patch the versions of zlib they contained, but the critical thing to note is that the ones that included zlib could just apply the patch to the older API version they contained, or to the trimmed version, or whatever they had. This meant that people could apply the patch without breaking half of their software, which depended on the particular API they were using.
2.6 isn't 100% userspace-compatible with 2.4; there are a number of utilities which need to be upgraded to deal with 2.6, and a few cases where 2.4 stuff isn't supported at all. So I wouldn't expect all 2.4 installations to be able to go to 2.6 when the time comes. For that matter, 2.4 still has the better ACPI support, and probably still will when 2.6.0 comes out.
As for when 2.6.0 will be out, Linus is turning that over to Andrew Morton, and we really have no idea what his style of stable kernel releases will be like. I'd actually expect to next see a relatively long 2.6.0-rc series before 2.6.0; maybe even a 2.6.0-pre series before that, depending on what he thinks of the seriousness of the remaining "should-fix" and "must-fix" lists and the reported bugs.
Given that SCO has beaten Microsoft in a lawsuit in the past (or, at least, gotten them to settle), I wouldn't be surprised if SCO went after Microsoft again as soon as Microsoft stopped paying them off.
I presume when you say to publish the votes, you mean that you produce a book of ballots that you counted, where each person can recognize their own, but there's no way to identify which was cast by someone else. (If you just published the ballots, using the kinds of ballots used where I've voted, people could count them, and they'd be anonymous, but it would be damn hard to tell yours apart from other people's)
You could have a system where each ballot has a different random symbol on it and is given out with the symbol covered by the original wrapper. The voter marks the candidates, takes a copy of the symbol, covers the whole thing with the wrapper again, and puts it in the box. At the end of the election, the ballots are counted and published. Then each voter can look through the book for the ballot with the right symbol to make sure it is there, and count the totals. If you don't show your symbol to anyone, nobody can tell who you voted for. For that matter, the polls could actually give the voter the opportunity, in cases where they could not read a ballot, to come forward with the copy of their symbol; the voter could sacrifice their anonymity to clarify their vote, if they wanted to (assuming, of course, that the receipts with the symbol also got, say, your name printed on it by way of carbon paper when you signed to get the ballot).
Care for some column headings for that record? Once you include the column headings, the flat version is 2/3 the size of the XML version. Sure, you don't need another set of column headings for each record, but, since you're talking about a single record, you can't just ignore the column headings.
Furthermore, XML isn't well suited for the case where you have a large number of records with the same column structure. It's not really even worthwhile for all different columns if you don't have nested records. But once you have a complicated structure, XML becomes much more manageable than defining separate tables with ids and references.
Once you get to something like an Apache config file, doing it as a flat file doesn't save you significant space, and makes the whole thing totally unusable.
The thing that I think was a success about the internet in relation to the outage is that there was no loss of connectivity between Boston and California. I actually found out that a huge chunk of the middle of the country had no power from a site on the other side of the outage from me.
As for the people who lost power, I suspect that they were largely more concerned about other things that their communications. When the neighborhood down the street from me lost power, the banks and stores closed. I don't think they closed due to communications issues; they weren't even handling depositing checks (which involves postal mail and takes overnight to clear anyway).
I suspect the reason that so many networks went offline was really that, while the internet is becoming critical for many businesses, it is critical only for things they do while they have power (which is also, obviously, critical). If businesses come to be able to handle power outages (which they might if they become sufficiently common for people to care), they will likely protect their networks against power failure at the same time. The main issue is the upstreams who might not prepare for outages, not realizing that the customers are actually hoping to work through the blackout.
Personally, I don't bother to put hubs and routers on UPS, because I doubt that it would help much. I would if the machines weren't self-sufficient (having been in an outage with a bunch of machines using NFS for things like the shutdown scripts). But, honestly, if the power's out, I'm not going to bother trying to work.
One vote doesn't matter all that much. But you can do better than one vote if you can convince somebody else to vote for your position. If you sit down in a local coffeeshop and convince a couple of people, your own vote is practically insignificant. If you manage to convince your favorite local newspaper, you're starting to sway a number of votes. Same for a union or an organization which would otherwise by neutral on the issue.
The candidates are largely identical, because they all want to be exactly the candidate that people want to vote for. All of the candidates who wouldn't compromise their position to get votes are too far from center to be electable. On the other hand, by maintaining a position convincingly, you can help to make the candidates think that center is closer to your point of view than they would otherwise. In this way, voicing your opinion to your candidate (and convincing other people to agree with you and do so), and responding to polls is more important than the actual vote.
If you want to be cynical, it's quite plausible that the 2000 election came down to the fact that voters were confused by the ballot and were refused help by poll workers. At least in my area, polls seem to be run by volunteers. Think of the impact you could have by volunteering for this. (To be a bit less cynical, the polls which were refusing to help people were evidentally understaffed. Just having more people show up to help might have given a more accurate result, even without those workers intending to affect the result in one direction or another.)
I personally think that there's too much emphasis placed on getting concerned individuals to vote, and too little on getting concerned individuals to get unconcerned individuals concerned. For that matter, our electoral system means that votes in elections which are close have a much larger effect than votes in elections which are not. My vote in the next presidential election won't count for anything, because, no matter what I do, my electoral vote will go to the Democrat. On the other hand, I could affect the outcome by convincing someone in a swing state. (In fact, I make a point of never voting for the candidate who will win if there's someone I prefer who won't win, although I do vote for a major candidate in a close race if there is one).
The reason that devfs is not yet deprecated in favor of udev yet is that udev depends on the kernel using the new driver model for everything that gets a device. Of course, that's not udev work per se, but it means that, for example, input devices (IIRC) don't yet work quite right with it.
The kernel having explicit knowledge of what it's doing in a uniform format is a new feature in 2.6, and it's not completely universal yet. Once that all works correctly, udev should work perfectly, and it is a better design than devfs, because it puts device naming in userspace, but device numbering comes from the kernel, and the kernel tells userspace what each device actually is. This is how the division of labor is supposed to be: the kernel has internal information, which it maintains, and an API, which it defines, but userspace can use that API to specify policy.
The name might, instead, indicate how he got enough time away from his family life to actually release test10.
Since, as you must know, he has a wife and kids. And he is sufficiently absent-minded to forget about the kids, but the wife might be more difficult. (According to a Wired? interview recently)
It might actually be easier to do singing than normal speech, because singing replaces intonation, tempo, and some of stress, all of which otherwise have to be determined from a syntactic and semantic analysis of the text in order to really sound right. There have been people who have learned to sing songs in languages they didn't know at all, while I have yet to hear of someone giving a lecture (as convincingly) in a language they didn't know.
You're missing the ever-popular "boy/girl doesn't love girl/boy", but that's about it. Except, of course, for those rare songs which avoid being cliche entirely by being (ironically) just wrong.
But even if the style isn't new, the melody isn't new, and the instrumentation isn't new, the combination can be new, and it takes skill to make the combination work (how do you play Beethoven's Ninth... on three electric guitars?). It might actually take 5 years with a given fashion to go through all of the melodies, at which point there's a new fashion.
It's certainly possible that classical music got all of the interesting underlying melodies. But the importance of style and instrumentation shouldn't be ignored (let alone lyrics). Interestingly, most currently available recordings of classical music are done by taking the underlying melody and playing it in the most boring way possible, which is rather different from how the scores were originally played. It's a bit like the works of Shakespear being read by a computer in a totally flat voice, except where the script actually specifies that the character whispers or shouts. Popular music is played (and was always played) with a substantial amount of interpretation by the performers, which forms the style.
Furthermore, I think that the style, and, especially, working out how the style and the underlying melody can be resolved, is as significant an application of artistic talent as writing the melody in the first place. It's like translating poetry; it's easy to do a direct translation, but making it actually work as poetry in the target language is at least as hard as writing the original (since you're constrained not only to write a good poem, but you have to also make it match another work in all of the ways that are important, while using entirely different grammer and vocabulary).
In short, even if classical music tried every melody, the existance of new styles and instrumentations means that there will be new complete works.
Between valgrind and ddd, I haven't had any bugs which I couldn't identify in a half hour, and only ones where there is something that the tools are bad at take more than a few minutes. Of course, that doesn't include programs that run fine but produce results which are simply not what I wanted and actually coming up with a way to fix the bug.
Valgrind, in particular, is really promising; the core architecture is now being used to identify potential race conditions (cases where different threads access the same memory without having the same lock), as well as identifying memory errors (including leaks).
What about the new enemies that you get when you kill all of your current enemies? Both the people who liked your current enemies (but not enough to join them before), and the people who are afraid of your ability to kill people that effectively.
The perfect weapon, in my opinion, would demoralize all of your enemies. War is never won by killing people; it's won by demoralizing the people who remain. You've won a war exactly when the enemy agrees that you've won. It's traditional to convince the enemy by killing people, but that's not really necessary, and doesn't necessarily help. (For example, we would have won Vietnam had we quit before we got too close to China; further success made the enemy unwilling to give up)
RSI is actually caused by using your hands while under too much stress. Really bad ergonomics (keyboard on desk while you sit on floor, e.g.) will make your hands hurt, but you can't keep it up long enough to damage yourself long term. Moderately bad ergonomics (standard desk situation) will make it require a certain amount of stress in your life. Really good ergonomics (special keyboard, etc.) might save you if you have to work under a lot of stress.
There are actually trends in stress-related problems. It used to be that people would faint. In the eighties, they had heart attacks. In the nineties (and still now, to a lesser extent), it was RSI. It's an interesting reflection on humanity that people can hear about a common disease and actually reproduce it in themselves, to the point of actually dying of it. But, if you're under stress, something's going to give out, and chances are that it'll be what you're worried about giving out.
So I'm not too sympathetic to RSI sufferers, but it's because I know that twenty years ago, they'd get heart attacks instead. RSI may suck, but at least it won't kill you.
A computer will be intelligent when it is capable of learning things which are entirely new, by applying its knowledge, trying things, general reasoning, and developing new representations. Chess is fine, go is fine, but I'll be impressed when a chess program goes on the internet and learns to play go.
There have been a series of problems that people have posed with the idea that they couldn't be solved without giving the program the ability to figure out the problem. Each time, however, people figure out how to solve the problem without giving the computer general intelligence. But the problems weren't definitions of intelligence ("Intelligence is what is necessary to play chess"), but rather people thought (incorrectly) that the problems couldn't be solved without intelligence.
On the other hand, there is a cluster of problems which have not been solved, despite many attempts, which are not intended to be difficult (which can be overcome eventually with clever coding and fast hardware), but rather which explicitly require undirected learning and the ability to generalize to new applications.
Of course, many of the problems originally posed as measures of intelligence were, in fact, problems that people wanted solutions to, because they were difficult for people. So the effort put into solving these problems without trying to use general intelligence is actually worthwhile.
Considering that this virus was synthesized from scratch, it's probably not something very effective. It's a long way from building a virus that works at all to building a virus that targets a particular organism or overcomes natural defenses.
Just because it's man-made doesn't make it more advanced than naturally-occurring viruses. It's been possible for a long time to build viruses from collected stocks, and these are generally much more frightening. What will be scary is when we have some clue as to how to design proteins, and could construct a virus with specific properties. Until then, we're not likely to create anything that doesn't arise in nature.
(Genetically modified foods are a slightly different issue; just because it might arise in nature doesn't mean we'd eat it if it did. Also, most of the organisms involved are much more resistant to mutation and genetic mixing than viruses and bacteria)
It makes sense that he didn't want to start the third movie with Saruman and finish a major plot in the first 7 minutes of the movie. It's somewhat unfortunate, however, that he didn't think of it last year and put it at the end of the second movie, and have Saruman done with at that point.
He'd already moved the beginning of the second book to the end of the first movie to avoid having a character die in the first scene of the second movie, so it's not unprecedented.
What I'd actually like to see is the whole series done as a single movie, without the changes made to make the breaks work. (One thing that hard about making a book into a movie is that people read books in chapters, and put the book down in between, but watch movies in theaters all together. On a DVD, however, you can watch a few scenes and then go do something else)
It scales perfectly well. I voted in an election with five races, one of which allowed voting for up to four candidates (alderman-at-large). In even years, there are many more spaces, and it still works fine; we have two columns/side of a legal-sized piece of heavy paper, large type, and fill in bars to indicate your choices. It's then read by an optical scanner. This year took a column and a half. IIRC, November 2000 took three and a half or so.
The next town over has a race (city council) where you rank a dozen candidates, and then use paper as well.
The real issue is voltages. Every type of chemical cell has a particular voltage it produces. It's easy to get a multiple of that voltage, but very hard to get any other voltage. Most of the new technologies produce voltages that don't match standard batteries, which means that, unless you have a custom device or a device designed for a range of voltages, you can't use anything new.
Actually, there is one place where battery technology has seen incredible advanced, and that is power tools. Ten years ago, a cordless ("powerless") drill was basically a toy, because it didn't produce enough torque, and the batteries would run out quickly. These days, people just don't use drills with cords, because the battery-powered ones are just as good, and cords are inconvenient. Of course, these use battery packs in the 10-15V range, which is sufficiently wide to handle a lot of variation, and is achieved with several cells, so the number of cells can be varied to change the voltage.
One AA (or AAA) battery is 1.5V; a single lithium cell is ~3V, and lithium ion and polymer are 3.6V. So you can make a newfangled rechargable which is perfect for devices that take 7 AAA batteries, and you can get pair-of-AA-battery lithium disposables, that's about it for new batteries of the sort that you don't get for Christmas.