Whether you get an allergic reaction depends (in my experience, at least), on the total quantity of allergens you're being exposed to, compared to the quantity you're usually exposed to. So constant exposure will improve your tolerance, but extra exposure at a particular time will make you sick. And if you're exposed to something during the night but not during the day, you can get a reaction every night.
It makes a lot of sense, actually. It's a feedback system for responding to unusual situations while adapting to different normal conditions, sort of like just about everything in nature.
This is a new games with the models made from the original, not an exact copy of the original. If you want the original (or either of the similar games), you can get them on DVD; the menu-nagivation mechanism in DVD players is actually just right for these games.
At some point, the MPAA will realize that these things actually serve to promote the movies. People will go to see the movie (provided it didn't suck, which they'd probably have found out from reviews anyway) to see it with quality that isn't terrible. Consider how many people buy DVDs of movies they have on VHS for the difference in quality there; now consider the difference in quality between a camcorder and a movie theater.
The MPAA has some clever people; it seems like they could figure this out. Or they could ask the RIAA about it; they've been paying ClearChannel tons of money for decades to distribute low-quality versions of music before it is widely available. Maybe they're afraid the pirates will start charging them millions of dollars to pirate their movies?
A.D. 1989: The United States invades Panama to capture renowned "hacker" Manual Noriega, who is suspected of writing the DeCSS utility.
After the CIA keeps advising the president to "Remove The F'ing Manuel", I assume?
A.D. 1941: Outcomes of critical World War II battles are held up for years due to German allegations that the British illegally acquired trade secrets. Four years later, all of their money tied up in legal bills, Germany files for bankruptcy. Documents are eventually declassified that prove they were right all along.
What better way to defeat crappy patent law than to cause MicroSoft to turn against it? Like it or not, large companies have substantial influence, and act primarily in their own short-term interest. Companies won't lobby against a law until they feel that it would adversely affect them, and patent law hasn't really been a problem for many big companies yet.
People will keep buying music regardless. Their car cd-players, and home cd-players will play them, and if they don't, their new players will play their old CD's.
So people will keep buying their old CDs if new ones don't work? But they already have their old CDs.
It doesn't make sense to try to sell CDs that will only work with old players, and the ability to read rewritten CD-Rs is only going to become more common. In a few years if this continues, it will be impossible to buy a CD and a player that will play it. Nobody will buy CDs, because you can't listen to music by buying CDs. I somehow doubt that the average consumer will think to buy old players on ebay to play their new discs.
The problem is really that this doesn't prevent copying. CSS, similarly, doesn't prevent copying; it prevents unauthorized reading, which is somewhat similar, at least. Macrovision DVDs work on suitable recent players. This, however, prevents reading on recent players. Except, of course, that you can still copy the music off of the CD. So, if you want to listen to the CD you buy, you can copy it and listen to the copy.
For a while, I've been suspecting that you don't have the right to listen to music you buy. I'm glad to see that I was right about the RIAA's position. At least they're trying to stop this practice with technology instead of laws...
(Just think how much cheaper CDs would be if they didn't actually have to have any music on them and nobody could tell...)
None of these are stripped, because I've got plenty of disk space. It goes down by a large factor if you strip them (but then you don't get such interesting crash dumps). While the absolute sizes are a bit excessive, the proportion is what matters here.
For almost all applications, you won't be writing code that cares whether something is, in context, "heavy" or not without caring what the scale is. There's nothing that will treat an object that's heavy for an ant to carry the same as an object that's heavy for a planet. Once you've fixed the set of things that might be done depending on the weight, you'll know what the scale is, and it won't help to use a "heaviness" value instead of the actual measurement.
But even if you wanted to have something work across different scales, what you want is to divide the measured weight by your reference weight and take the log. This would give -24 for a 1kg planet and 5 for a 1kg ant load (I think, just guessing reference weights). If you cared, you could then divide by the normal variation. This is still more useful than values between 0 and 1, still not bounded, and takes into account the standards for heaviness. (I also notice that "not heavy at all" and "very heavy" aren't numbers between 0 and 1, either)
Fuzzy logic generally involved mapping from one system into a 0-1 range, doing something oversimplified, and then mapping into another system. Each of the mappings is very sound mathematics (although you generally want to do it with functions of continuous variables and a pair of fourier transforms). The "logic" part, however, is not particularly justified; "and" and "or" aren't really that much like min and max. And without anything justified happening at the intermediate layer, you'd do better to go straight from input to output. Going straight from input to output is far more common than using fuzzy logic; it's signal processing.
Sure, but that would still be true even if nobody had broken this key. The fact remains that the breaking of this key doesn't give any advantage toward breaking keys in the future.
More precisely, they broke a single key, not "weak elliptic curve cryptography". Breaking another 109-bit key (one that had been used to encrypt something, for example) would take another 4 years.
But density isn't mass. You make a design worse, not better, by adding more patterns than are needed to cover the volume of the problem.
The ideal design will do everything that needs to be done in a principled way using a pattern (either from the book or elsewhere), and nothing that doesn't need to be done.
Furthermore, you have to make sure that the right patterns are being used. Each pattern is good for one situation, and as bad as no design for any situation that doesn't fit. A lot of patterns may indicate that the engineer sat down with the book and went through the index putting in as many as possible (and some may be being used to deal with the poor fit of others of them).
The problem with fuzzy logic is that if it would be useful, something else would be more useful. Sure, you can have a fuzzy truth value for whether something is hot or not, but you'd do better with a temperature measurement instead of a value between 0 and 1. There's very little that's useful when converted into a bounded linear range.
Of course, fuzzy logic is only supposed to replace boolean values. But computer science has not failed to notice the limitations of boolean values; booleans are only used to control things where partial values wouldn't be useful.
Size of emacs: 4763354 Size of libqt-mt: 10796068 Size of libc.a: 24187284
A statically compiled emacs may be pretty big, but it's also 98% libc. Emacs has had a reputation for being really big for ten years, but it's really only big for ten years ago.
There is always "included" MicroSoft software. But, since it's not itemized, it would change absolutely nothing if they made Windows free and moved the cost into the bundled applications. Nobody ships just Windows without any other MS software (and MS could easily prevent OEMs from doing that), so MS can avoid restrictions on OS pricing and such by doing the accounting differently.
Would they do this? It depends on what their marketting department feels will drum up business. So long as they get their money from everyone, they don't care about the details. It's certainly possible that they'll offer to simplify licensing by making Windows free-as-in-beer and copy-but-don't-reverse-engineer, knowing that everybody who gets Windows free will then spend the money they save on Office, and people who run Office on Linux will lose their price advantage.
Generally, if you're pleading guilty, you'd be presumed to know who accused you, since you presumably know who you attacked. In fact, you're only supposed to plead guilty if you actually committed the crime (if you didn't do it but think you'd be convicted anyway, you plead no contest), so your guilty plea is unlikely to be accepted unless you at least know the identity of the victem. On the other hand, you don't necessarily get to face your accuser, which would reveal the identity of the accuser to other people who aren't presumed to already know (such as the jury and spectators) and potentially be hard on the victem (who might prefer to think of the hacker as a criminal rather than some scruffy kid).
One guy is testing the mouse in a lecture hall, using VNC to export the display to the overhead projector. Meanwhile, a second guy is stealing his laptop and fleeing the building.
They're talking about Vice City, not the existing ones in which you have to be violent if you want to get anywhere. The next installment evidentally will let you advance through peaceful missions as well as violence. I've actually always thought that the series would do well to support players who didn't want to hurt people intentionally; of course, it didn't help that I played GTA3 with a controller that tended to pull ever so slightly to the left, making it hard to fight on foot but not affecting running around or driving.
I think I'd actually have a really good time with an arc where all of the missions are driving places, following cars, and generally trying to survive while the police and the groups you're not working for try to stop you. There could be a game mode where this is the only reasonable arc because people get really pissed off at you whenever you kill people.
There isn't a killer new feature this time. It's really the large number of major features since 2.X started. They didn't change all that much since 2.4, but there's very little that hasn't changed since 2.0, and it doesn't make sense to never change the major number just because you improve things at a steady rate.
Or you could say that the number of minor version increases exponentially with respect to the major number, and, since the major number changed after 1.2, it should clearly change after 2.(2^2).
The fundamental problem is the concept of "opening" a file. Having an operation that's easy for the user to invoke, but that could do basically anything, is a really bad idea. (Yesterday, I tried to open a door and I ended up opening a restaurant. Today, I tried to open my wallet and opened a wound on my leg. Then I tried to close my wallet, but closed my bank account instead. Anyway...) It was a bad idea on the Mac, it was a bad idea on Windows, and it's a bad idea on Linux.
What KMail wants to let you do is "view" a file. You view.sh files with a text editor. You view.jpgs with GIMP. You don't view Windows executables. Programs that view files are safe to use (unless there's a bug in the viewer).
You may, at some point, want to execute a file. You do this with exec(). You don't do this with a viewer.
If you insist on acting on files without any concern for the operation you're going to do on the file, I'd suggest using "rm", which will work on any file, regardless of type, and will cause relatively little damage in the long run.
This isn't really about making people happy. It's about providing them with useful software. If what they want to do is extend it and sell it to people without source, well, they paid their taxes (supposedly) for that right. If what they want to do is extend it and GPL it, they paid for that, too. If they just want to use it, that's also fine.
The GPL is a way of protecting your work from reuse by people who don't extend the same rights to you. It is an important way of getting people who are writing software for personal reasons to release it to the public. In the case of work contracted by the government, that isn't necessary, because the programmer is paid for the work. The government also doesn't want to retain access to derived works; they've generally made the software sufficient for their purposes, and their mandate is to make their software available to everybody.
The stage boundaries are already essentially status polling, because the stages are, themselves, pipelined, with different lengths depending on the complexity of the particular unit. Due to all of the out-of-order and parallel execution, it would be a pain to keep track outside of the unit what to expect when; instead, the information just goes through with the data. This does, however, avoid the problem of the clock being too fast for some stages.
The real issue is clock distribution, because, unlike power and ground, you have to distribute clock in wires (rather than plates held at different voltages), and you have to move a lot of current around all over the place, which makes the inductive issues more complicated to simulate and work out: you've put a strangely-shaped antenna broadcasting a tone with harmonics at exactly the frequencies you care about in the middle of your chip.
Ah, but this time, the release management isn't going to be done by Linus, but by (IIRC) Marcello. In order to make this fact clear to people who don't follow the development process, he's going to skip right from 3.0-rc(NN) to 3.0.12, at which point it will be stable and go into mostly maintenence mode, with 3.1 starting up.
Not to mention all of the other features that have been added between 2.0 and 2.4, all of which should contribute to the decision to change the major number. It would actually be really interesting to see how much of the code hasn't been changed during the 2.x series.
In order to replace X, you'll need something that will emulate X for legacy applications, but also support a cleaner interface. That actually shouldn't be as hard as you might think: you just have a new core rendering engine that is sufficient to support X with all of the extensions you want to make, and then you just add a second interface to it. Since X doesn't specify the internals of how the server works, the fact that you've changed the engine doesn't matter.
X is bloated mostly by features that used to be important but that nobody cares about any more. There's a lot of complexity in handling colors in color tables, but people only really use DirectColor these days, and dealing with the color code is just a hassle. X has a good mechanism for adding extensions, but there's no way to remove the steps that are no longer useful, and there's no way to get features which have become significant into the core protocol.
Reportedly, that error message is traditional, and used to be accurate. You'd get that if the printer had jammed in such a way that there was paper pressed on one side against a spinning part, generating heat and paper dust. By the time you got to the printer, it would probably have burst into flames. Of course, the printer could have broken in a less catastrophic way, but people don't tend to complain when their computer tells them their huge printer is on fire and it turns out it's merely broken. These days, of course, printers rarely burst into flames, but if there's something mysteriously wrong with the printer that's not one of the standard problems, who knows? (The message tends to come up if the kernel doesn't understand the printer status quite right)
Slashdot Mirror
Whether you get an allergic reaction depends (in my experience, at least), on the total quantity of allergens you're being exposed to, compared to the quantity you're usually exposed to. So constant exposure will improve your tolerance, but extra exposure at a particular time will make you sick. And if you're exposed to something during the night but not during the day, you can get a reaction every night.
It makes a lot of sense, actually. It's a feedback system for responding to unusual situations while adapting to different normal conditions, sort of like just about everything in nature.
This is a new games with the models made from the original, not an exact copy of the original. If you want the original (or either of the similar games), you can get them on DVD; the menu-nagivation mechanism in DVD players is actually just right for these games.
At some point, the MPAA will realize that these things actually serve to promote the movies. People will go to see the movie (provided it didn't suck, which they'd probably have found out from reviews anyway) to see it with quality that isn't terrible. Consider how many people buy DVDs of movies they have on VHS for the difference in quality there; now consider the difference in quality between a camcorder and a movie theater.
The MPAA has some clever people; it seems like they could figure this out. Or they could ask the RIAA about it; they've been paying ClearChannel tons of money for decades to distribute low-quality versions of music before it is widely available. Maybe they're afraid the pirates will start charging them millions of dollars to pirate their movies?
A.D. 1989: The United States invades Panama to capture renowned "hacker" Manual Noriega, who is suspected of writing the DeCSS utility.
After the CIA keeps advising the president to "Remove The F'ing Manuel", I assume?
A.D. 1941: Outcomes of critical World War II battles are held up for years due to German allegations that the British illegally acquired trade secrets. Four years later, all of their money tied up in legal bills, Germany files for bankruptcy. Documents are eventually declassified that prove they were right all along.
What better way to defeat crappy patent law than to cause MicroSoft to turn against it? Like it or not, large companies have substantial influence, and act primarily in their own short-term interest. Companies won't lobby against a law until they feel that it would adversely affect them, and patent law hasn't really been a problem for many big companies yet.
People will keep buying music regardless. Their car cd-players, and home cd-players will play them, and if they don't, their new players will play their old CD's.
So people will keep buying their old CDs if new ones don't work? But they already have their old CDs.
It doesn't make sense to try to sell CDs that will only work with old players, and the ability to read rewritten CD-Rs is only going to become more common. In a few years if this continues, it will be impossible to buy a CD and a player that will play it. Nobody will buy CDs, because you can't listen to music by buying CDs. I somehow doubt that the average consumer will think to buy old players on ebay to play their new discs.
The problem is really that this doesn't prevent copying. CSS, similarly, doesn't prevent copying; it prevents unauthorized reading, which is somewhat similar, at least. Macrovision DVDs work on suitable recent players. This, however, prevents reading on recent players. Except, of course, that you can still copy the music off of the CD. So, if you want to listen to the CD you buy, you can copy it and listen to the copy.
For a while, I've been suspecting that you don't have the right to listen to music you buy. I'm glad to see that I was right about the RIAA's position. At least they're trying to stop this practice with technology instead of laws...
(Just think how much cheaper CDs would be if they didn't actually have to have any music on them and nobody could tell...)
None of these are stripped, because I've got plenty of disk space. It goes down by a large factor if you strip them (but then you don't get such interesting crash dumps). While the absolute sizes are a bit excessive, the proportion is what matters here.
For almost all applications, you won't be writing code that cares whether something is, in context, "heavy" or not without caring what the scale is. There's nothing that will treat an object that's heavy for an ant to carry the same as an object that's heavy for a planet. Once you've fixed the set of things that might be done depending on the weight, you'll know what the scale is, and it won't help to use a "heaviness" value instead of the actual measurement.
But even if you wanted to have something work across different scales, what you want is to divide the measured weight by your reference weight and take the log. This would give -24 for a 1kg planet and 5 for a 1kg ant load (I think, just guessing reference weights). If you cared, you could then divide by the normal variation. This is still more useful than values between 0 and 1, still not bounded, and takes into account the standards for heaviness. (I also notice that "not heavy at all" and "very heavy" aren't numbers between 0 and 1, either)
Fuzzy logic generally involved mapping from one system into a 0-1 range, doing something oversimplified, and then mapping into another system. Each of the mappings is very sound mathematics (although you generally want to do it with functions of continuous variables and a pair of fourier transforms). The "logic" part, however, is not particularly justified; "and" and "or" aren't really that much like min and max. And without anything justified happening at the intermediate layer, you'd do better to go straight from input to output. Going straight from input to output is far more common than using fuzzy logic; it's signal processing.
Sure, but that would still be true even if nobody had broken this key. The fact remains that the breaking of this key doesn't give any advantage toward breaking keys in the future.
More precisely, they broke a single key, not "weak elliptic curve cryptography". Breaking another 109-bit key (one that had been used to encrypt something, for example) would take another 4 years.
But density isn't mass. You make a design worse, not better, by adding more patterns than are needed to cover the volume of the problem.
The ideal design will do everything that needs to be done in a principled way using a pattern (either from the book or elsewhere), and nothing that doesn't need to be done.
Furthermore, you have to make sure that the right patterns are being used. Each pattern is good for one situation, and as bad as no design for any situation that doesn't fit. A lot of patterns may indicate that the engineer sat down with the book and went through the index putting in as many as possible (and some may be being used to deal with the poor fit of others of them).
The problem with fuzzy logic is that if it would be useful, something else would be more useful. Sure, you can have a fuzzy truth value for whether something is hot or not, but you'd do better with a temperature measurement instead of a value between 0 and 1. There's very little that's useful when converted into a bounded linear range.
Of course, fuzzy logic is only supposed to replace boolean values. But computer science has not failed to notice the limitations of boolean values; booleans are only used to control things where partial values wouldn't be useful.
Size of emacs: 4763354
Size of libqt-mt: 10796068
Size of libc.a: 24187284
A statically compiled emacs may be pretty big, but it's also 98% libc. Emacs has had a reputation for being really big for ten years, but it's really only big for ten years ago.
(file sizes are from my local builds)
There is always "included" MicroSoft software. But, since it's not itemized, it would change absolutely nothing if they made Windows free and moved the cost into the bundled applications. Nobody ships just Windows without any other MS software (and MS could easily prevent OEMs from doing that), so MS can avoid restrictions on OS pricing and such by doing the accounting differently.
Would they do this? It depends on what their marketting department feels will drum up business. So long as they get their money from everyone, they don't care about the details. It's certainly possible that they'll offer to simplify licensing by making Windows free-as-in-beer and copy-but-don't-reverse-engineer, knowing that everybody who gets Windows free will then spend the money they save on Office, and people who run Office on Linux will lose their price advantage.
Generally, if you're pleading guilty, you'd be presumed to know who accused you, since you presumably know who you attacked. In fact, you're only supposed to plead guilty if you actually committed the crime (if you didn't do it but think you'd be convicted anyway, you plead no contest), so your guilty plea is unlikely to be accepted unless you at least know the identity of the victem. On the other hand, you don't necessarily get to face your accuser, which would reveal the identity of the accuser to other people who aren't presumed to already know (such as the jury and spectators) and potentially be hard on the victem (who might prefer to think of the hacker as a criminal rather than some scruffy kid).
One guy is testing the mouse in a lecture hall, using VNC to export the display to the overhead projector. Meanwhile, a second guy is stealing his laptop and fleeing the building.
They're talking about Vice City, not the existing ones in which you have to be violent if you want to get anywhere. The next installment evidentally will let you advance through peaceful missions as well as violence. I've actually always thought that the series would do well to support players who didn't want to hurt people intentionally; of course, it didn't help that I played GTA3 with a controller that tended to pull ever so slightly to the left, making it hard to fight on foot but not affecting running around or driving.
I think I'd actually have a really good time with an arc where all of the missions are driving places, following cars, and generally trying to survive while the police and the groups you're not working for try to stop you. There could be a game mode where this is the only reasonable arc because people get really pissed off at you whenever you kill people.
There isn't a killer new feature this time. It's really the large number of major features since 2.X started. They didn't change all that much since 2.4, but there's very little that hasn't changed since 2.0, and it doesn't make sense to never change the major number just because you improve things at a steady rate.
Or you could say that the number of minor version increases exponentially with respect to the major number, and, since the major number changed after 1.2, it should clearly change after 2.(2^2).
The fundamental problem is the concept of "opening" a file. Having an operation that's easy for the user to invoke, but that could do basically anything, is a really bad idea. (Yesterday, I tried to open a door and I ended up opening a restaurant. Today, I tried to open my wallet and opened a wound on my leg. Then I tried to close my wallet, but closed my bank account instead. Anyway...) It was a bad idea on the Mac, it was a bad idea on Windows, and it's a bad idea on Linux.
.sh files with a text editor. You view .jpgs with GIMP. You don't view Windows executables. Programs that view files are safe to use (unless there's a bug in the viewer).
What KMail wants to let you do is "view" a file. You view
You may, at some point, want to execute a file. You do this with exec(). You don't do this with a viewer.
If you insist on acting on files without any concern for the operation you're going to do on the file, I'd suggest using "rm", which will work on any file, regardless of type, and will cause relatively little damage in the long run.
This isn't really about making people happy. It's about providing them with useful software. If what they want to do is extend it and sell it to people without source, well, they paid their taxes (supposedly) for that right. If what they want to do is extend it and GPL it, they paid for that, too. If they just want to use it, that's also fine.
The GPL is a way of protecting your work from reuse by people who don't extend the same rights to you. It is an important way of getting people who are writing software for personal reasons to release it to the public. In the case of work contracted by the government, that isn't necessary, because the programmer is paid for the work. The government also doesn't want to retain access to derived works; they've generally made the software sufficient for their purposes, and their mandate is to make their software available to everybody.
The stage boundaries are already essentially status polling, because the stages are, themselves, pipelined, with different lengths depending on the complexity of the particular unit. Due to all of the out-of-order and parallel execution, it would be a pain to keep track outside of the unit what to expect when; instead, the information just goes through with the data. This does, however, avoid the problem of the clock being too fast for some stages.
The real issue is clock distribution, because, unlike power and ground, you have to distribute clock in wires (rather than plates held at different voltages), and you have to move a lot of current around all over the place, which makes the inductive issues more complicated to simulate and work out: you've put a strangely-shaped antenna broadcasting a tone with harmonics at exactly the frequencies you care about in the middle of your chip.
Ah, but this time, the release management isn't going to be done by Linus, but by (IIRC) Marcello. In order to make this fact clear to people who don't follow the development process, he's going to skip right from 3.0-rc(NN) to 3.0.12, at which point it will be stable and go into mostly maintenence mode, with 3.1 starting up.
Not to mention all of the other features that have been added between 2.0 and 2.4, all of which should contribute to the decision to change the major number. It would actually be really interesting to see how much of the code hasn't been changed during the 2.x series.
In order to replace X, you'll need something that will emulate X for legacy applications, but also support a cleaner interface. That actually shouldn't be as hard as you might think: you just have a new core rendering engine that is sufficient to support X with all of the extensions you want to make, and then you just add a second interface to it. Since X doesn't specify the internals of how the server works, the fact that you've changed the engine doesn't matter.
X is bloated mostly by features that used to be important but that nobody cares about any more. There's a lot of complexity in handling colors in color tables, but people only really use DirectColor these days, and dealing with the color code is just a hassle. X has a good mechanism for adding extensions, but there's no way to remove the steps that are no longer useful, and there's no way to get features which have become significant into the core protocol.
Reportedly, that error message is traditional, and used to be accurate. You'd get that if the printer had jammed in such a way that there was paper pressed on one side against a spinning part, generating heat and paper dust. By the time you got to the printer, it would probably have burst into flames. Of course, the printer could have broken in a less catastrophic way, but people don't tend to complain when their computer tells them their huge printer is on fire and it turns out it's merely broken. These days, of course, printers rarely burst into flames, but if there's something mysteriously wrong with the printer that's not one of the standard problems, who knows? (The message tends to come up if the kernel doesn't understand the printer status quite right)
See this linux kernel post.