A zombie process is dead already, there's no point in killing it. It only exists to provide a return code to its parent. Processes whose parent has exited are reparented to init, which immediately checks and discards the return code. Therefore, zombie processes always have parents which still exist, and those parents may be interested in the return code from their child. If you were allowed to kill the zombie child process, then what do you propose waitpid() return to the parent when it's called?
Re:Slashdot as Upgrade Offerer
on
Ubuntu 6.10 is Out
·
· Score: 2, Informative
I read on an Ubuntu wiki that Dapper, being their current long-term-support version, does not automatically suggest upgrading to Edgy, which is the 'bleeding edge' version. Presumably Edgy will notify users when Feisty is ready.
You don't have to enter anything twice. Everything is a single transaction which refers to two accounts. So as soon as you enter something saying "$200 from Checking into Expenses" there is automatically a "$200 into Expenses from Checking". It's just two ways of looking at the same transaction.
Indeed. I work on an MMOG which will be released in about two years, and we've considered and rejected using DX10 because there won't be enough people using Vista yet.
The previous estimate was 14 billion years. The new estimate is "at least 1000 billion years". Hence, it's now at least 986 billion years older than the previous estimate. Of course, the precision of the new estimate is much lower, so subtracting them is nonsensical.
On the contrary, Win32 is going to become a de-facto standard for windowed applications. Microsoft is not going to be able to get away from the existing set of applications which use it, because Vista won't offer enough improvements over XP to make people switch like they did from DOS / Win 3.1 to Win95.
I predict that in 5 to 10 years, all OSs will support Win32 apps, though not necessarily as their primary app type. Wine will work well for existing binaries, and winelib will work even better for developers who want to make cross-platform apps from the start.
The back-button issue bothered me too, but I think they've fixed it. Through what magic, I don't know, but most of the time anyway (much more often than before) back does what you expect it to do.
Read the article...the reason (I'm beginning to believe) they didn't get authorization under FISA was because they couldn't. The wiretapping in question was done using broad analysis of a random sampling of phone calls. How can they go to a FISA judge with that?
They aren't stupid. They could easily have gone to the judges within 72 hours if this were normal wiretapping. It's not.
Well, NTP is not trying to shut down RIM. They're trying to get the money which they are entitled to by having the patents.
If NTP was a true 'little guy' and RIM had gone on to make a ton of money off their ideas, would still think that the smart little guy (who never produced and sold a product) who wants his money is trying to 'shut down the economically successfull RIM'?
If it's okay for a little guy to do it, it's okay for a big company to do it. The actions of corporations do not suddenly become evil when they transition from 'little guy' to 'big company'.
What do you think will happen if no one enforces their legitimate copyright, and everyone (not just technophiles) has push-button access to free copies of Rome on their TV, indistinguishable from the original? Do you think HBO will make money? Do you think they will continue to make high-budget shows when their subscriber base shrinks because no one has the incentive to subscribe?
Their most likely source of income would be incoporating ads into the scripts in a way which is impossible to skip, like references to how well Tide gets their togas cleaned. Is that better than paying for HBO?
The technology isn't wrong. But don't go bullshitting yourself thinking that downloading copyrighted material anonymously and in large quantities is somehow justifiable.
Fine, people trying to kill Bittorrent altogether because some people use it for copyright infringement is bad. But for christs sake, HBO puts a ton of money into those episodes and it deserves to get paid for them. It's illegal and immoral to download them, and I think it's perfectly fine to attack transfers of obviously copyrighted material.
How do you justify it morally? On a very small scale, filling in an episode you haven't seen, sure, no big deal. Massive redistribution of an entire series is obviously going to harm HBO, whose only crime was creating something which people like to watch. Do you think that HBO is some soulless bunch of corporate assholes who deserve to get screwed? Where do you draw the line between small artists and these corporate assholes? HBO hires the best screenwriters, directors, actors, and technical people in the business, and the result is the show that you like to watch. Do you think you're benefiting anyone by downloading it for free en masse?
What do you think will happen if no one enforces their legitimate copyright, and everyone has push-button access to free copies of Rome. Fast-forward to a time when most houses in America would have the ability to watch freely downloaded episodes on their TV, as an alternative to subscribing to HBO. Do you think HBO will make money? Do you think they will continue to make high-budget shows when their subscriber base shrinks? Their most likely source of income is incoporating ads into the scripts in a way which is impossible to skip, like references to how well Tide gets their togas cleaned. Is that better than paying for HBO?
The technology isn't wrong. But don't go bullshitting yourself thinking that downloading copyrighted material anonymously and in large quantities is somehow justifiable.
The big difference is that now, file trading can completely replace legitimate purchases, because:
1. You can get basically anything you want, at any time, in any amount. Previously, you were limited to what your friends were willing to give you a copy of. It took time and effort to make the copies. Now, I can copy an entire persons music collection with no skin off their back.
2. You get it in a format which is identical to the most modern format (mp3). Previously you could make tapes, or hand-labelled CDs, but there was still incentive to go buy real CDs for albums you like. Now, if you listen to your music via mp3s, you won't see the slightest bit of difference between a purchased song and a traded song.
I don't think that casual music sharing is a problem, and I think that if it were still limited to that, the recording companies wouldn't care much either. I recognize that they don't make a distinction (i.e. they don't tell you that a little sharing is okay, they say any sharing is bad) but that's not surprising.
However, it completely ruins the business model of music distribution to allow for anonymous, limitless music sharing. At some level, it's that business model which funds music. The landscape of music production and distribution will definitely change in the next decade, and I'm not sure it's going to be for the better.
Our kids may look back at the days when you could buy good songs by good artists, rather than having free access to mediocre music.
The best thing about grub is that as long as it's correctly loaded onto your MBR (not particularly difficult) you don't really need a config file at all. Anything which can go in the config file can be typed at the command-line. That's really useful in a bootloader, which otherwise can be particularly difficult to fix when you make a boneheaded mistake. GRUB has saved me several times with this feature. It's also handy when you want to boot into a different runlevel or whatever, it's easy to edit the kernel parameters before booting.
The other best thing is that you don't need to do much when you do something like install a new kernel or initrd; just update the config file (and you really don't even need to do that, see above). With Lilo you have to remember to run/sbin/lilo to update the MBR.
The worst thing about GRUB is that it was written for the Hurd, and has a Hurd-centric view of disks which can be difficult to work with in other operating systems.
That's because most of the problems we have these days aren't viruses. They're worms. Viruses (and trojans) are transmitted slowly, via a user's actions. Worms spread proactively, and do so quickly enough that there isn't time for a virus company to put out a signature. Generally, the effect of a worm isn't anything it does to your computer, but what it does to the network. The only way to stop worms is to make sure there are no security holes in your operating system.
Virus scanning for anything other than emails is a waste of time. And even email viruses aren't very hard to avoid; it's much easier to secure an email client than a network-connected operating system.
Linux isn't an operating system. GNU/Linux isn't even an operating system. Linspire is an operating system, and it seems to do a pretty good job of it. Someone who runs Linspire shouldn't say "I run Linux" any more than someone who runs Mac OS should say "I run FreeBSD".
We need to get away from thinking that Red Hat and SuSE are "basically the same". People shouldn't expect things to work the same on these different systems. I can't necessarily download a program for "Linux" and install it on all of these systems, at least not how modern users mean "install" (creates desktop interface links, can be easily uninstalled, etc).
Evaluate them on their own merits as an OS, and (mostly) forget the fact that many of them happen to share the same low-level implementations. It's great that it's easy to port software between these various GNU/Linux-based operating systems, but that's what it is--porting. If we lump them all together under a single umbrella, then each of them loses the ability to differentiate in positive ways (like Linspire is trying to do).
1. Let me play the game at the most difficult level RIGHT AWAY. I absolutely hate playing a game through twice. I'd rather play a hard than a game where I have to play it on an easy setting, and then 'get' to play it again on hard.
2. Include more multiplayer options. Why can't we play multiplayer on every one of the singleplayer maps? They're obviously in the game. I understand if MP isn't the focus, they don't want to spend a lot of time on extra maps, but at least let us play on the ones that are there.
Yeah, I kind of thought there should be some sort of 'dark ritual' which Anakin reluctantly agreed to, which bonded him to Palpatine and the Dark Side. He mentions in ROTJ that the dark side is too powerful for him to leave, even if he wanted to. At the end of ROTS, When Padme questioned what he was doing, he could have said "I did it for you; I wish I could undo it, but I can't".
I agree Obi-Wan and Anakin's fight would have been much more interesting if there had been that internal conflict in Anakin. It's my favority part of ROTJ, when you see that Vader would like to leave the dark side, but can't.
I know very little about chip design, but I've heard that power and heat are becoming the big limiting factors in the scalability of processors.
Does that mean that AMD's processors have much more room to scale, since they draw less power and generate less heat today?
Are there any estimates for how high, and when, AMD will be able to scale up to signifigantly higher frequencies?
Why doesn't AMD ramp up the frequencies now, and put out processors now with approximately the same power draw as Intel's chips, but with signifigantly higher performance? Except in laptops (which Intel owns) and blade servers (which maybe AMD is going after), low power isn't a big deal. Performance is a big deal everywhere.
Perhaps part of the problem is that crawlable pages have had to avoid actions using GET, but dynamic or authenticated pages haven't, because crawlers can't get to them. Now that Google is effectively crawling sites while users are browsing authenticated or dynamic content, the problem is occuring in places it hadn't before.
If they leave it up permanently, I'm sure it will be hacked once the next exploit is available. It's not impossible to secure a system like IIS, but it's much more difficult to make it secure permanently, as new exploits are found.
If this is a test of IIS's security (for example as opposed to Apache) they should make it an ongoing test, and measure it not by whether it was hacked within a certain short time period, but how many times it is hacked over a long period of time.
Can't the kernel also do a "soft" unmount, which flushes everything to disk to get it in a clean state, but leaves the filesystem mounted in the kernel?
If the user unplugs the drive, the kernel does a complete unmount, and if they access it again, the drive is transparently remounted.
Considering how many operating systems have this problem, I doubt the solution is this simple...
I had to do this the other day, and found it very quickly in the documentation, by searching for JetDirect: like this.
I know that doesn't prove that the docs are great, but the example you wanted is right there.
Personally I think CUPS is great. The Web interface isn't perfect, and I always forget the port, but it makes managing jobs and printers very easy. Plus all modern Linux systems will easily pick up the CUPS server over the network and allow me to manage jobs that way.
What stops all this? A real, heretofore unknown high-level security model, that actually says "The email program can access stored email data, preferences, and can talk to the network on this port, to these hosts" and "the word processor cannot talk IRC" and so forth.
What you are referring to is SELinux. There's not a lot of work going on in the desktop portions of this, but Fedora is doing a lot to get fine-grained permissions set up for server-side daemons.
Eventually SELinux will be applied to desktop apps too, and then Linux will be an extremely secure system. It will be a while, however, because it's very difficult to define exactly what a program is allowed to do; the goal is to restrict anything it doesn't need to do while letting it do anything it does need to do. If you err on one side it's a security hole (though admittedly small compared to today's apps) and if you err on the other side the app will fail when it tries to perform some action.
Slashdot Mirror
A zombie process is dead already, there's no point in killing it. It only exists to provide a return code to its parent. Processes whose parent has exited are reparented to init, which immediately checks and discards the return code. Therefore, zombie processes always have parents which still exist, and those parents may be interested in the return code from their child. If you were allowed to kill the zombie child process, then what do you propose waitpid() return to the parent when it's called?
I read on an Ubuntu wiki that Dapper, being their current long-term-support version, does not automatically suggest upgrading to Edgy, which is the 'bleeding edge' version. Presumably Edgy will notify users when Feisty is ready.
You don't have to enter anything twice. Everything is a single transaction which refers to two accounts. So as soon as you enter something saying "$200 from Checking into Expenses" there is automatically a "$200 into Expenses from Checking". It's just two ways of looking at the same transaction.
Indeed. I work on an MMOG which will be released in about two years, and we've considered and rejected using DX10 because there won't be enough people using Vista yet.
The previous estimate was 14 billion years. The new estimate is "at least 1000 billion years". Hence, it's now at least 986 billion years older than the previous estimate. Of course, the precision of the new estimate is much lower, so subtracting them is nonsensical.
On the contrary, Win32 is going to become a de-facto standard for windowed applications. Microsoft is not going to be able to get away from the existing set of applications which use it, because Vista won't offer enough improvements over XP to make people switch like they did from DOS / Win 3.1 to Win95.
I predict that in 5 to 10 years, all OSs will support Win32 apps, though not necessarily as their primary app type. Wine will work well for existing binaries, and winelib will work even better for developers who want to make cross-platform apps from the start.
The back-button issue bothered me too, but I think they've fixed it. Through what magic, I don't know, but most of the time anyway (much more often than before) back does what you expect it to do.
Read the article...the reason (I'm beginning to believe) they didn't get authorization under FISA was because they couldn't. The wiretapping in question was done using broad analysis of a random sampling of phone calls. How can they go to a FISA judge with that?
They aren't stupid. They could easily have gone to the judges within 72 hours if this were normal wiretapping. It's not.
Is this a new meme or something i've missed?
Well, NTP is not trying to shut down RIM. They're trying to get the money which they are entitled to by having the patents.
If NTP was a true 'little guy' and RIM had gone on to make a ton of money off their ideas, would still think that the smart little guy (who never produced and sold a product) who wants his money is trying to 'shut down the economically successfull RIM'?
If it's okay for a little guy to do it, it's okay for a big company to do it. The actions of corporations do not suddenly become evil when they transition from 'little guy' to 'big company'.
What do you think will happen if no one enforces their legitimate copyright, and everyone (not just technophiles) has push-button access to free copies of Rome on their TV, indistinguishable from the original? Do you think HBO will make money? Do you think they will continue to make high-budget shows when their subscriber base shrinks because no one has the incentive to subscribe?
Their most likely source of income would be incoporating ads into the scripts in a way which is impossible to skip, like references to how well Tide gets their togas cleaned. Is that better than paying for HBO?
The technology isn't wrong. But don't go bullshitting yourself thinking that downloading copyrighted material anonymously and in large quantities is somehow justifiable.
Fine, people trying to kill Bittorrent altogether because some people use it for copyright infringement is bad. But for christs sake, HBO puts a ton of money into those episodes and it deserves to get paid for them. It's illegal and immoral to download them, and I think it's perfectly fine to attack transfers of obviously copyrighted material.
How do you justify it morally? On a very small scale, filling in an episode you haven't seen, sure, no big deal. Massive redistribution of an entire series is obviously going to harm HBO, whose only crime was creating something which people like to watch. Do you think that HBO is some soulless bunch of corporate assholes who deserve to get screwed? Where do you draw the line between small artists and these corporate assholes? HBO hires the best screenwriters, directors, actors, and technical people in the business, and the result is the show that you like to watch. Do you think you're benefiting anyone by downloading it for free en masse?
What do you think will happen if no one enforces their legitimate copyright, and everyone has push-button access to free copies of Rome. Fast-forward to a time when most houses in America would have the ability to watch freely downloaded episodes on their TV, as an alternative to subscribing to HBO. Do you think HBO will make money? Do you think they will continue to make high-budget shows when their subscriber base shrinks? Their most likely source of income is incoporating ads into the scripts in a way which is impossible to skip, like references to how well Tide gets their togas cleaned. Is that better than paying for HBO?
The technology isn't wrong. But don't go bullshitting yourself thinking that downloading copyrighted material anonymously and in large quantities is somehow justifiable.
The big difference is that now, file trading can completely replace legitimate purchases, because:
1. You can get basically anything you want, at any time, in any amount. Previously, you were limited to what your friends were willing to give you a copy of. It took time and effort to make the copies. Now, I can copy an entire persons music collection with no skin off their back.
2. You get it in a format which is identical to the most modern format (mp3). Previously you could make tapes, or hand-labelled CDs, but there was still incentive to go buy real CDs for albums you like. Now, if you listen to your music via mp3s, you won't see the slightest bit of difference between a purchased song and a traded song.
I don't think that casual music sharing is a problem, and I think that if it were still limited to that, the recording companies wouldn't care much either. I recognize that they don't make a distinction (i.e. they don't tell you that a little sharing is okay, they say any sharing is bad) but that's not surprising.
However, it completely ruins the business model of music distribution to allow for anonymous, limitless music sharing. At some level, it's that business model which funds music. The landscape of music production and distribution will definitely change in the next decade, and I'm not sure it's going to be for the better.
Our kids may look back at the days when you could buy good songs by good artists, rather than having free access to mediocre music.
The best thing about grub is that as long as it's correctly loaded onto your MBR (not particularly difficult) you don't really need a config file at all. Anything which can go in the config file can be typed at the command-line. That's really useful in a bootloader, which otherwise can be particularly difficult to fix when you make a boneheaded mistake. GRUB has saved me several times with this feature. It's also handy when you want to boot into a different runlevel or whatever, it's easy to edit the kernel parameters before booting.
/sbin/lilo to update the MBR.
The other best thing is that you don't need to do much when you do something like install a new kernel or initrd; just update the config file (and you really don't even need to do that, see above). With Lilo you have to remember to run
The worst thing about GRUB is that it was written for the Hurd, and has a Hurd-centric view of disks which can be difficult to work with in other operating systems.
That's because most of the problems we have these days aren't viruses. They're worms. Viruses (and trojans) are transmitted slowly, via a user's actions. Worms spread proactively, and do so quickly enough that there isn't time for a virus company to put out a signature. Generally, the effect of a worm isn't anything it does to your computer, but what it does to the network. The only way to stop worms is to make sure there are no security holes in your operating system.
Virus scanning for anything other than emails is a waste of time. And even email viruses aren't very hard to avoid; it's much easier to secure an email client than a network-connected operating system.
Linux isn't an operating system. GNU/Linux isn't even an operating system. Linspire is an operating system, and it seems to do a pretty good job of it. Someone who runs Linspire shouldn't say "I run Linux" any more than someone who runs Mac OS should say "I run FreeBSD".
We need to get away from thinking that Red Hat and SuSE are "basically the same". People shouldn't expect things to work the same on these different systems. I can't necessarily download a program for "Linux" and install it on all of these systems, at least not how modern users mean "install" (creates desktop interface links, can be easily uninstalled, etc).
Evaluate them on their own merits as an OS, and (mostly) forget the fact that many of them happen to share the same low-level implementations. It's great that it's easy to port software between these various GNU/Linux-based operating systems, but that's what it is--porting. If we lump them all together under a single umbrella, then each of them loses the ability to differentiate in positive ways (like Linspire is trying to do).
1. Let me play the game at the most difficult level RIGHT AWAY. I absolutely hate playing a game through twice. I'd rather play a hard than a game where I have to play it on an easy setting, and then 'get' to play it again on hard.
2. Include more multiplayer options. Why can't we play multiplayer on every one of the singleplayer maps? They're obviously in the game. I understand if MP isn't the focus, they don't want to spend a lot of time on extra maps, but at least let us play on the ones that are there.
Yeah, I kind of thought there should be some sort of 'dark ritual' which Anakin reluctantly agreed to, which bonded him to Palpatine and the Dark Side. He mentions in ROTJ that the dark side is too powerful for him to leave, even if he wanted to. At the end of ROTS, When Padme questioned what he was doing, he could have said "I did it for you; I wish I could undo it, but I can't".
I agree Obi-Wan and Anakin's fight would have been much more interesting if there had been that internal conflict in Anakin. It's my favority part of ROTJ, when you see that Vader would like to leave the dark side, but can't.
Of course, when you say base 2, you mean 2 in base 10.
I know very little about chip design, but I've heard that power and heat are becoming the big limiting factors in the scalability of processors.
Does that mean that AMD's processors have much more room to scale, since they draw less power and generate less heat today?
Are there any estimates for how high, and when, AMD will be able to scale up to signifigantly higher frequencies?
Why doesn't AMD ramp up the frequencies now, and put out processors now with approximately the same power draw as Intel's chips, but with signifigantly higher performance? Except in laptops (which Intel owns) and blade servers (which maybe AMD is going after), low power isn't a big deal. Performance is a big deal everywhere.
Perhaps part of the problem is that crawlable pages have had to avoid actions using GET, but dynamic or authenticated pages haven't, because crawlers can't get to them. Now that Google is effectively crawling sites while users are browsing authenticated or dynamic content, the problem is occuring in places it hadn't before.
If they leave it up permanently, I'm sure it will be hacked once the next exploit is available. It's not impossible to secure a system like IIS, but it's much more difficult to make it secure permanently, as new exploits are found.
If this is a test of IIS's security (for example as opposed to Apache) they should make it an ongoing test, and measure it not by whether it was hacked within a certain short time period, but how many times it is hacked over a long period of time.
Can't the kernel also do a "soft" unmount, which flushes everything to disk to get it in a clean state, but leaves the filesystem mounted in the kernel?
If the user unplugs the drive, the kernel does a complete unmount, and if they access it again, the drive is transparently remounted.
Considering how many operating systems have this problem, I doubt the solution is this simple...
I had to do this the other day, and found it very quickly in the documentation, by searching for JetDirect: like this.
I know that doesn't prove that the docs are great, but the example you wanted is right there.
Personally I think CUPS is great. The Web interface isn't perfect, and I always forget the port, but it makes managing jobs and printers very easy. Plus all modern Linux systems will easily pick up the CUPS server over the network and allow me to manage jobs that way.
What stops all this? A real, heretofore unknown high-level security model, that actually says "The email program can access stored email data, preferences, and can talk to the network on this port, to these hosts" and "the word processor cannot talk IRC" and so forth.
What you are referring to is SELinux. There's not a lot of work going on in the desktop portions of this, but Fedora is doing a lot to get fine-grained permissions set up for server-side daemons.
Eventually SELinux will be applied to desktop apps too, and then Linux will be an extremely secure system. It will be a while, however, because it's very difficult to define exactly what a program is allowed to do; the goal is to restrict anything it doesn't need to do while letting it do anything it does need to do. If you err on one side it's a security hole (though admittedly small compared to today's apps) and if you err on the other side the app will fail when it tries to perform some action.