The iPod is closely associated with iTunes, which makes getting music legally more convenient than downloading. Also, iPod owners are likely to have more money to spend on music than owners of less expensive mp3 players.
Sequence constructors: Your suggestion doesn't work, for reasons which should be obvious. First of all, numbers in curly braces are ambiguous; they could be either an array initializer or an initializer for any struct with that many members. And second, arrays don't carry around their size.
Template concepts: Templates were the one thing most of the compilers got horribly wrong the first time around. What makes you think this time will be better? Hence the need for caution.
The auto keyword: Yeah, it's overdue. But it's a nice thing to have.
Garbage collection: Are you joking? It's anything *but* "the easiest thing to do". It's easy for Java, Perl and the like because they run their programs inside of interpreters; they don't have to handle the case where there not only is no interpreter, the hardware can't even support one (no threads, no memory management unit, no operating system support; all of these are problems for embedded programmers, all of them make garbage collection impossible, and all of them can be gracefully handled by C++.)
Bias towards systems programming: And just what *else* are systems programmers supposed to code with? Just because Linux is written in C doesn't mean that C++ can abandon systems programmers. If C++ were to make a change that made it unsuitable for systems programming, the computer world *would* come crashing down. Or the new version wouldn't be adopted by anyone.
Properties: It's not just syntactic sugar, it's a trap which hinders binary compatibility.
Continuations: Unsuitable for systems programming, because they not only require garbage collection, they require garbage collection *performed on stack frames*. Try maintaining binary compatibility with a feature like that!
Closures: Your example is no good, because window.close() should already be defined as a library function.
Runtime substitution of vtable method pointers: Sounds like an ugly hack which, if given language support, would encourage many people to write extremely bad code.
AMD has yet to prove the Athlon64 is more than a fluke (which is not to say I think it is a fluke, but AMD fanbois have a penchant for pretending AMD can do no wrong).
Funny, I thought the Athlon64 was proving that the original Athlon wasn't a fluke. I think two highly superior architectures in a row is proof enough.
Routers running out of RAM is an IPv4-specific problem, too. With IPv6 the IP address space should be almost completely uniform, so that even a core router can figure out which way a packet goes from only the first few bits of the destination address.
That just means that your 900MHz phone broadcasts at higher power. The difference between 900MHz and 1.2GHz is small enough that differences in power are more important.
No, forgetting to lock your door is different because you have to do it so often that it's easy to forget. Leaving an access point open is like buying a house and not taking down the "Open House" sign the realtor put up. You should expect visitors until you take it down (configure a password), but you only have to do it once.
SCO does not currently own the UNIX trademark; UNIX a registered trademark of The Open Group. See Google's first hit for "Unix trademark". The Open Group is a "vendor- and technology-neutral consortium".
A competent technical writer can take a well-written software specification and have a draft user manual ready by the time the code is compilable.
Really? Most of the advice I've seen has said that the code should be compilable from day one and be compilable all the time. And "well-written software specification" isn't something one normally sees in the real world.
That's because AltaVista has too many syllables, and is actually two words. "Google" replaces "search for" in a sentence, and has the same number of syllables but a more specific meaning, so it catches on.
The floppy drive power connector is a smaller version of the hard drive power connectors you already have a bunch of, so you should be able to get a splitter cable with the right connector.
IIRC, the Microsoftian way to do this would be to add make a CD-R of the install CD with the drivers included. I don't know how you would go about this, though.
OTOH, there are reasons Windows-based games are so far ahead besides simple market share--graphics interfaces are one of those "funny shaped blocks" in Windows that is very well suited to its task.
This is demonstrably false, because Windows games don't use the funny shaped blocks Windows provides, because they (a) don't mix well with 3D rendering, and (b) usually don't match the aesthetic the game designer wants.
They require LOTS of runtime support, they just bundle it in the executable. OCaml is "easy to integrate with C code" in the sense that an OCaml program can use a C library, NOT the other way around. If every library fired up its own garbage collector and loaded its own language runtime, your average large program would immediately run out of memory.
Remember the GDI image handling vulnerability awhile back? The one where Microsoft released a tool to scan for programs that kept private copies of the vulnerable DLL, but which tended to miss things? That's a very common situation in the Windows world; there's no dependency management, so every program bundles every dependency. And if a common dependency happens to be exploitable, then there's no easy way to collect all the different copies.
They've mostly taken care of programs overwriting eachothers' copies with broken or back-leveled versions, at least. That's an improvement. But Microsoft's package management (letting installers do whatever they want) is still broken.
Because the advertising networks are in no way trustworthy, and they insert their crap all over the place. A site is not safe to view with IE unless its server is well secured, it has no advertisements and it uses a cryptographic signature to prove its identity. And don't forget that mistyping a URL will usually lead you to a shady domain-name squatter's page, as will following an outdated link. In other words, you can use IE to visit Windows Update (which you should set as your home page), and that's pretty much it.
The Doom 3 box says it requires 384MB of memory, which you have a third of, so of *course* it wasn't playable. For the console version they downscaled all the textures (which take tons of memory) so that it would fit, plus the stripped-down OS. That's in addition to turning the detail settings way down. If you had a reasonable amount of memory, Doom 3 would look much better on your computer than on the Xbox.
And Doom 3 is not and never was CPU bound on the PC; its reputation for high system requirements is solely due to its demands on the graphics card, and the Xbox wasn't particularly lacking in that department.
I saw this earlier in the print edition, and it's not really what it sounds like. The question to which 15% said yes was whether you'd ever changed the procedure, methodology, or results of an experiment in response to pressure from a funding source. Well, changing the results would be very, very bad, but they actually asked a separate question on that one and only 0.3% (a statistically insignificant number) said yes. Changing methodology is not necessarily illegitimate; if your funding source says "give me X precision", or "measure Y too while you're at it", then the procedure's going to change to reflect that. It doesn't mean there's bias, it means the question was asked incorrectly.
It's not illegal to download something if it's authorized, directly or indirectly, by the author. If your friend agrees to let you put the file on KaZaA, that makes it legal. If he doesn't, that makes what *you're* doing illegal. And also liable for any damages stemming from the RIAA's downloading of that file, so any money your friend won would ultimately come out of your pocket.
The law is generally not conducive to silly schemes such as this one. Usually the only ways to make money off the courts are for someone to actually wrong you, to bully a smaller entity into settling, or to attempt fraud (which is unlikely to work).
Yes, it matters who has standing, but it doesn't matter whether you knew who had standing when you committed the tort, which is what I said. Once it comes to court, proving ownership - and therefore standing - is easy. The article said that because the GPL doesn't spell out in perfectly clear terms who the owner is, no one would be able to prosecute, which is rediculous.
This isn't a real problem. The basic issue is that only the copyright holder has standing to litigate copyright violations. But it's never really ambiguous who the copyright holder is. The FSF recommends that free software developers assign their copyrights to the FSF, so that they can deal with violations. Many individual projects require all contributors to assign their copyrights to a consortium, to the project leader, or something similar. There are some projects with copyright held jointly by many developers, but there's almost always someone who you can point to and say "this person/organization holds copyright over the majority of the code". And even if it's not immediately obvious from the license who the copyright holder is, that doesn't matter in court; not knowing who has standing to prosecute is no defense.
Also, notice that "Richard Stallman, the founder of the Free Software Foundation and the author of the GPL, was unable to comment in time for this article." A brief interview with RMS would surely have cleared this up as a non-issue.
Your math is completely wrong. First of all, a system only draws its peak amount of power while it's booting (typically about 5 seconds while the hard drives spin up). Second, the power supply wattage ratings are inflated because they add up the max wattages on each of the rails, when in reality you'll only come close to max on one. Third, cooling a room with one computer sure as hell doesn't take 580W, either. Finally, there's no way you'd ever need a power supply like this unless you had a whole lot of hard drives at once. Overall, your number is off by at least a factor of 10 - and that's not even considering the fact that you assumed the computer would be on 24/7 and in California.
Slashdot Mirror
*Ahem*
You are hereby forbidden to use the English language in a pedantic and patronising manner ever again.
That should be:
You are hereby forbidden from using the English language in a pedantic and patronising manner ever again.
It's hard to download security patches on a slow connection. So you'll be less secure with dial-up.
FAT is commonly used on floppies and memory cards, so they had to support it. It has nothing to do with retaining pieces of Win9x.
The iPod is closely associated with iTunes, which makes getting music legally more convenient than downloading. Also, iPod owners are likely to have more money to spend on music than owners of less expensive mp3 players.
Sequence constructors: Your suggestion doesn't work, for reasons which should be obvious. First of all, numbers in curly braces are ambiguous; they could be either an array initializer or an initializer for any struct with that many members. And second, arrays don't carry around their size.
Template concepts: Templates were the one thing most of the compilers got horribly wrong the first time around. What makes you think this time will be better? Hence the need for caution.
The auto keyword: Yeah, it's overdue. But it's a nice thing to have.
Garbage collection: Are you joking? It's anything *but* "the easiest thing to do". It's easy for Java, Perl and the like because they run their programs inside of interpreters; they don't have to handle the case where there not only is no interpreter, the hardware can't even support one (no threads, no memory management unit, no operating system support; all of these are problems for embedded programmers, all of them make garbage collection impossible, and all of them can be gracefully handled by C++.)
Bias towards systems programming: And just what *else* are systems programmers supposed to code with? Just because Linux is written in C doesn't mean that C++ can abandon systems programmers. If C++ were to make a change that made it unsuitable for systems programming, the computer world *would* come crashing down. Or the new version wouldn't be adopted by anyone.
Properties: It's not just syntactic sugar, it's a trap which hinders binary compatibility.
Continuations: Unsuitable for systems programming, because they not only require garbage collection, they require garbage collection *performed on stack frames*. Try maintaining binary compatibility with a feature like that!
Closures: Your example is no good, because window.close() should already be defined as a library function.
Runtime substitution of vtable method pointers: Sounds like an ugly hack which, if given language support, would encourage many people to write extremely bad code.
Funny, I thought the Athlon64 was proving that the original Athlon wasn't a fluke. I think two highly superior architectures in a row is proof enough.
Routers running out of RAM is an IPv4-specific problem, too. With IPv6 the IP address space should be almost completely uniform, so that even a core router can figure out which way a packet goes from only the first few bits of the destination address.
That just means that your 900MHz phone broadcasts at higher power. The difference between 900MHz and 1.2GHz is small enough that differences in power are more important.
No, forgetting to lock your door is different because you have to do it so often that it's easy to forget. Leaving an access point open is like buying a house and not taking down the "Open House" sign the realtor put up. You should expect visitors until you take it down (configure a password), but you only have to do it once.
SCO does not currently own the UNIX trademark; UNIX a registered trademark of The Open Group. See Google's first hit for "Unix trademark". The Open Group is a "vendor- and technology-neutral consortium".
That's because AltaVista has too many syllables, and is actually two words. "Google" replaces "search for" in a sentence, and has the same number of syllables but a more specific meaning, so it catches on.
They may not "finish", but they won't max out the pipe either; games typically use a fixed amount of bandwidth and leave the rest alone.
The floppy drive power connector is a smaller version of the hard drive power connectors you already have a bunch of, so you should be able to get a splitter cable with the right connector.
IIRC, the Microsoftian way to do this would be to add make a CD-R of the install CD with the drivers included. I don't know how you would go about this, though.
They require LOTS of runtime support, they just bundle it in the executable. OCaml is "easy to integrate with C code" in the sense that an OCaml program can use a C library, NOT the other way around. If every library fired up its own garbage collector and loaded its own language runtime, your average large program would immediately run out of memory.
Remember the GDI image handling vulnerability awhile back? The one where Microsoft released a tool to scan for programs that kept private copies of the vulnerable DLL, but which tended to miss things? That's a very common situation in the Windows world; there's no dependency management, so every program bundles every dependency. And if a common dependency happens to be exploitable, then there's no easy way to collect all the different copies.
They've mostly taken care of programs overwriting eachothers' copies with broken or back-leveled versions, at least. That's an improvement. But Microsoft's package management (letting installers do whatever they want) is still broken.
Because the advertising networks are in no way trustworthy, and they insert their crap all over the place. A site is not safe to view with IE unless its server is well secured, it has no advertisements and it uses a cryptographic signature to prove its identity. And don't forget that mistyping a URL will usually lead you to a shady domain-name squatter's page, as will following an outdated link. In other words, you can use IE to visit Windows Update (which you should set as your home page), and that's pretty much it.
The Doom 3 box says it requires 384MB of memory, which you have a third of, so of *course* it wasn't playable. For the console version they downscaled all the textures (which take tons of memory) so that it would fit, plus the stripped-down OS. That's in addition to turning the detail settings way down. If you had a reasonable amount of memory, Doom 3 would look much better on your computer than on the Xbox.
And Doom 3 is not and never was CPU bound on the PC; its reputation for high system requirements is solely due to its demands on the graphics card, and the Xbox wasn't particularly lacking in that department.
Not always. What many zombies do is connect to an IRC channel, and wait for commands there. That goes through firewalls no problem.
I saw this earlier in the print edition, and it's not really what it sounds like. The question to which 15% said yes was whether you'd ever changed the procedure, methodology, or results of an experiment in response to pressure from a funding source. Well, changing the results would be very, very bad, but they actually asked a separate question on that one and only 0.3% (a statistically insignificant number) said yes. Changing methodology is not necessarily illegitimate; if your funding source says "give me X precision", or "measure Y too while you're at it", then the procedure's going to change to reflect that. It doesn't mean there's bias, it means the question was asked incorrectly.
It's not illegal to download something if it's authorized, directly or indirectly, by the author. If your friend agrees to let you put the file on KaZaA, that makes it legal. If he doesn't, that makes what *you're* doing illegal. And also liable for any damages stemming from the RIAA's downloading of that file, so any money your friend won would ultimately come out of your pocket.
The law is generally not conducive to silly schemes such as this one. Usually the only ways to make money off the courts are for someone to actually wrong you, to bully a smaller entity into settling, or to attempt fraud (which is unlikely to work).
Yes, it matters who has standing, but it doesn't matter whether you knew who had standing when you committed the tort, which is what I said. Once it comes to court, proving ownership - and therefore standing - is easy. The article said that because the GPL doesn't spell out in perfectly clear terms who the owner is, no one would be able to prosecute, which is rediculous.
This isn't a real problem. The basic issue is that only the copyright holder has standing to litigate copyright violations. But it's never really ambiguous who the copyright holder is. The FSF recommends that free software developers assign their copyrights to the FSF, so that they can deal with violations. Many individual projects require all contributors to assign their copyrights to a consortium, to the project leader, or something similar. There are some projects with copyright held jointly by many developers, but there's almost always someone who you can point to and say "this person/organization holds copyright over the majority of the code". And even if it's not immediately obvious from the license who the copyright holder is, that doesn't matter in court; not knowing who has standing to prosecute is no defense.
Also, notice that "Richard Stallman, the founder of the Free Software Foundation and the author of the GPL, was unable to comment in time for this article." A brief interview with RMS would surely have cleared this up as a non-issue.
Your math is completely wrong. First of all, a system only draws its peak amount of power while it's booting (typically about 5 seconds while the hard drives spin up). Second, the power supply wattage ratings are inflated because they add up the max wattages on each of the rails, when in reality you'll only come close to max on one. Third, cooling a room with one computer sure as hell doesn't take 580W, either. Finally, there's no way you'd ever need a power supply like this unless you had a whole lot of hard drives at once. Overall, your number is off by at least a factor of 10 - and that's not even considering the fact that you assumed the computer would be on 24/7 and in California.