I agree with you about the monopoly, that's never good.
But about the conductor. I agree it's kind of sad for conductors, and it's nice (in theory) to have a pair of eyes watching out for the safety of all.
BUT, in the case, I think a well designed system could actually work much better than a human.
Humans are good at observing, but only while they're observing. Where I live the conductors on the trains are mostly staring blankly into nothingness. They don't prevent people from getting doors closed on them, they rarely stop the train longer than is scheduled to wait even if someone is running to catch it. They're basically robots, only they do a shitty ass job because their task is so monotonous and boring.
If a system were designed well enough this is a perfect case where computers will do much better than human, simply because computers don't get bored, and they don't slack off.
As for the would be rapist. All trains have those help buttons, and I hope if someone was getting raped we don't need a conductor to push the button and call for help.
This is absolute crazy talk. When I'm admining my server as root, I need to be able to run every application... and this is the way it is now. There's very few cases where something will refuse to run as root, and that's exactly the way it should be.
They key here is that many applications drop their privilege level to some predefined state, ie on many systems this is nobody:nobody.
A white list is no good, it'll just cause a whole bunch of people shouting "I need to run this as admin!"
Just let applications drop their privs, and if it's necessary implement a black list for rogues that don't do what they're supposed to.
Uhh... good ideas, but definitely not simple. Think of the logistics of implementing 1 or 2. The tax part would be good although I doubt it would do anything to reduce oil consumption. It would just put more money in the governments coffers. Although I'm certain it would push the big 5 to start pumping out more hybrids and spur research into more alternative fuel sources.
I think a fun thing to do would be to repaint the lines on the roads so that SUVs could no longer fit. Use the extra space for bike lanes.
And I think bike lanes should have jumps and other obstacles installed in them, which are just fun for bikers, but would totally demolish any cars that think they can drive in the lane.
Actually why stop there... as long as we're dreaming, I think we should install nerf or water balloon cannons along side major freeways. When they detect a car with only one occupant they fire at will. Imagine the awesomeness.
That doesn't mean you don't immediately benefit from having a 64 bit OS running 32 bit applications.
For example, each 32 bit process can be allotted the total 32 bit address space (4GB). This is a vast improvement over 32 bit windows where each process can only have 2GB (despite the 32 bit address range).
Also, continuing with that... you can then have x number of 4GB 32-bit processes all running concurrently up to the 64 bit address range (16EB).
While this may not seem significant to most users, there are some who can't wait for this, especially heavy photoshop users who work with multi gigabyte images.
Well... aside from that one there's also the shred manpage which discusses this:
CAUTION: Note that shred relies on a very important assumption: that
the filesystem overwrites data in place. This is the traditional way
to do things, but many modern filesystem designs do not satisfy this
assumption. The following are examples of filesystems on which shred
is not effective:
* log-structured or journaled filesystems, such as those supplied with
AIX and Solaris (and JFS, ReiserFS, XFS, Ext3, etc.)
I can't really find any other good info on this, but given what I know about how journaled file systems work I certainly believe this.
Having said that, if anyone has any real information on this I'd be interested to know as well.
I think the real problem is that not that many people run Debian anymore! Obviously I have no real stats on this, but just from the people I know, all of them (many who were vehement Debian fanboys) have switched away to other distros.
Not trying to start a distro war, as I totally hate them (we're all Linux users, come on guys, let's be brothers not warring factions!). I think the availability of so many distros really makes shifts like these both possible, and good!
I can't say for sure, but I know some people who switched away from Debian seemed to suggest that although it's a great distro, it became stagnant.
Maybe this election reflects the developers waning enthusiasm for the project? Which was in turn picked up by the community?
Why else would we see so many Debian based distros (ie Ubuntu, etc)?
I hope the Debian leaders take this as a warning sign, and endeavor to start addressing whatever issues need to be addressed. Even though I don't use Debian, I'd hate to see such a great project fall to the way side.
I'm sure this guy knows what he's talking about... but I just wanted to mention that I was simply taking exception to the "DoD standards file erasure" utilities out there.
They sound like they're doing something, but they're not. The file system driver doesn't actually perform the operations that cause the heads to interact with the platters (in the case of journalled file systems).
Like I said before though, you can nuke the whole drive at once easily enough, which is probably where everything werewolf said applies. Or alternatively you can just use a non-journalled file system if you're that into file erase utils.
Just wanted to update and say that, you can safely erase an entire hard drive (even with a journalled filesystem) if you unmount it, and wipe it (ie side step the filesystem driver). This is pretty easy to do in Linux, not sure about Windows... although I know one good method would be to use the hard drive manuacturers low level formatter (running it 3 or 4 times should be enough).
But once again, be wary of any file erasure programs like the one mentioned by the grandparent post. You need to take care that your usage of them is not in vain!
Actually that's a good point. You should be able to do that with Ext3 (since you can mount Ext3 partitions with Ext2)... although as far as I'm aware, you can't disable journalling on Reiser, XFS, NTFS, or any of the other major ones. I'd love to be proven wrong though, if someone has info on that.
Oh yeah, totally... or individual files. You could, obviously do other things if you want to toast the entire file system like using the hard drive manufacturers low level formatter 3 or 4 times... You're right though, as long as you're not shredding a mount point you should be okay.
The thing is that there's lots of these utilities out there that claim to be doing something when they're not. I just wanted to throw that out there and make sure people know.
Despite what they tell you, these erasure programs do not work on drives using journalled filesystems (ie almost every drive there is these days... including you Windows users with NTFS, and Linux users with Reiser, Ext3, XFS, etc).
Many modern operating systems such as Windows XP (NTFS), Mac OS X ( [[HFS+]] ), and GNU/Linux with a kernel version greater than 2.4 (Ext3, JFS, ReiserFS, and XFS) have the ability to use a journaling filesystem that makes complete erasure of data unlikely.
There are several ways to securely wipe files when using journaling filesystems:
Store data that needs to be wiped on a partition (slice, volume, or drive) that uses a non-journaling filesystem. For example, users of Windows can use a Z: drive formatted with FAT32, and users of GNU/Linux can use a partition formatted with Ext2.
Store data that needs to be wiped on a partition that is encrypted using Hard Disk Encryption. This eliminates the need to use a secure wiping mechanism for individual files.
Store data on a temporary partition using any journaling or non-journaling filesystem. When it is time to wipe all files, use a tool such as Eraser or Wipe to securely wipe the entire partition.
Physically destroy the hard drive after use by melting the hard drive. (Passing a magnet over the hard drive will not work.)
So, basically... there is no proper way of protecting yourself from undelete data recovery methods, if you use a journalled file system, aside from keeping some thermite handy!
If you ask me, we should all be encyrpting our data partitions by now!
CAUTION: Note that shred relies on a very important assumption: that
the filesystem overwrites data in place. This is the traditional way
to do things, but many modern filesystem designs do not satisfy this
assumption. The following are examples of filesystems on which shred
is not effective:
* log-structured or journaled filesystems, such as those supplied with
AIX and Solaris (and JFS, ReiserFS, XFS, Ext3, etc.)
Ie, shred is useless. Also of note is someone's idea to write/dev/null to/dev/hd? is also useless for this same reason.
Also of note is that this applies to Windows users (NTFS) as well!
There are several ways to securely wipe files when using journaling filesystems:
1. Store data that needs to be wiped on a partition (slice, volume, or drive) that uses a non-journaling filesystem. For example, users of Windows can use a Z: drive formatted with FAT32, and users of GNU/Linux can use a partition formatted with Ext2.
2. Store data that needs to be wiped on a partition that is encrypted using Hard Disk Encryption. This eliminates the need to use a secure wiping mechanism for individual files.
3. Store data on a temporary partition using any journaling or non-journaling filesystem. When it is time to wipe all files, use a tool such as Eraser or Wipe to securely wipe the entire partition.
4. Physically destroy the hard drive after use by melting the hard drive. (Passing a magnet over the hard drive will not work.)
Ie... There is no method for proper undelete protection of journalled drives. Better have your thermite ready!!
It's more than you did... which doesn't say much for your knowledge. Since you're clearly not interested in knowing the real truth of the matter I consider it closed. But if you ever do _really_ want to know... go to your local university and ask some computer science professors who can personally speak on it, since they were there.
and as for your p.s. you need to relax and take an anti-paranoia pill! I'm also the freak. I'm just a guy who knows something you don't, and you're just a guy who doesn't want to admit to being wrong.
See Kalak's post. He did some good digging. And btw there are those of us who are sufficiently old that we actually remember people using the term in the proper way. If you have no real world knowledge of something, maybe you shouldn't be so adamant about it. Face it, you're wrong.
Of all the people who replied, your reply was by far the best. You actually thought about it and you make a good point.
It's true that the general public uses the word in a certain sense... so why fight it?
Well, I actually have to take exception with your assumption that words have no "correct" meaning. Every word has a correct meaning, and yes it depends on the circumstance.
I'm not vehemently opposed to the popular usage of the word hacker, but people need to know that the term has really been appropriated by the media and used to distort reality.
For example, computer crimes now have such a huge negative connotation attached to them (probably because they're grossly misunderstand even by so-called computer experts) that you could spend almost twice as long in jail for "hacking" a bank than if you showed up with a gun and threatened peoples lives in the process (but didn't kill anyone).
If you hacked a bank, yeah you probably pulled off a magnificent hack, both in the way it was originally intended, and in the way it has come to be known.
I just wanted to help people remember that we are generally a bunch of brainwashed patsies, and we need to reclaim some critical thinking!
Look at the replies to my original post. People totally side step the issue nitpicking little insignificant points despite the glaring fact that the term "hacker" does have an alternate meaning.
I agree that ordinarily arguing semantics is a fairly worthless endeavor, but in this case I have to take an exception. It's not like we're arguing how to spell the rapper $0.50's name (Fiddy, Fifty)... we're discussing a concept vital to the forefathers of computation.
People like Gosling, Wozniak, even Jobs to some extent... these are some of the people we owe our thanks to for the modern PC, and these people were the true hackers.
We disservice them by letting the media contort an art form into something which is viewed as illegal.
Like any art form... a tool can be misused. The sculpturs shaper can easily be used to kill, just like a hammer, screwdriver, hell even a paint brush... but just because a few people kill with a hammer, do we start associating the word "carpenter" with villian?
No.
But then again... most of us understand carpentry don't we... not that we can all do it. But it's something every person can at least grasp.
You can't just brush off the entire article because it doesn't directly discuss the history of the word. What it does is prove that you are wrong. It says exactly the opposite of what you're trying to suggest.
It's a well known fact that the term "hacker" did not originally apply to the people that media now calls hackers.
Cracking refers to people who break into computer systems using nefarious means. Ie Kevin Mitnick is mentioned on the wikipedia page, as he should be since he is probably the worlds most notorious cracker.
Just because the media says it, doesn't mean it's true. And if a cracker ever refers to him/herself as a hacker, you can rest easy because all your base will not belong to them. Anyone worth their merit knows the correct definition and differentations between cracking/hacking/spidering/phreaking/etc.
And just in case you all are too lazy to read the links... Linus Torvalds is listed as a famous hacker. This is the true definition of the term. It's not because he ever broke into computer systems, it's because he's a good programmer.
Also of note is that in the computer science community the word "hack" has gone on to have a somewhat negative connotation. For example, "Dude this code is such a hack." Although this refers more loosely to the "hack and slash" programming methodology... which often results in ugly code that is held together very loosely.
However, an ugly code "hack" and the word "hacker" are distinctly different. Please refrain from falling prey to false assumptions based on media in the future.
Did you even read what I wrote?
I did say that ATI support is getting better, and people have been having good success for some time now. Your problems have nothing to do with Gentoo.
And you should be fine now! AMD64 is no problem, and ATI is no problem (well mostly).
Re:Will it actually compile this time?
on
Gentoo 2005.0 Released
·
· Score: 2, Informative
Yes it compiles for amd64. Also your video card has absolutely nothing to do with compilation! The only thing that your video card affects is the X server (and to some degree frame buffer console mode)... which is your fault for buying ATI (known for notoriously bad Linux support).
Support a company that cares about Linux: NVidia.
Having said that... many people with ATI cards have Linux running properly (with 3D support). I've heard they've been improving their driver steadily.
Well you're right that arbitrary limits are... arbitrary. Another option might be to include kernel support for dynamic detection of high speed recursive forks.
However, then by your argument, you would of course get some user who really wants to recursively fork 10,000 processes in 10 milliseconds.
In the end you _have_ to make _some_ kind of arbitrary decision, whether it be on process numbers, fork speed, etc.
Anyway, for now I'll stick to my solution. When my computer gets better, and I run more processes simulataneously I'll simply up the limit. And all the while I'll know I'm protected from fork-bombs, whereas you'll still be whining about how there's no single limit that can work in all cases yet you'd curse the day you were performing a critical system update and some jackass decided to form-bomb your system for the fun of it.
And just in case you're worried about running into the limit. I dare to set it to 512. I'll be you that the only way you'll ever run into that limit (in the next 10 years) is through a purposeful attempt (like a fork-bomb).
Slashdot Mirror
I agree with you about the monopoly, that's never good.
But about the conductor. I agree it's kind of sad for conductors, and it's nice (in theory) to have a pair of eyes watching out for the safety of all.
BUT, in the case, I think a well designed system could actually work much better than a human.
Humans are good at observing, but only while they're observing. Where I live the conductors on the trains are mostly staring blankly into nothingness. They don't prevent people from getting doors closed on them, they rarely stop the train longer than is scheduled to wait even if someone is running to catch it. They're basically robots, only they do a shitty ass job because their task is so monotonous and boring.
If a system were designed well enough this is a perfect case where computers will do much better than human, simply because computers don't get bored, and they don't slack off.
As for the would be rapist. All trains have those help buttons, and I hope if someone was getting raped we don't need a conductor to push the button and call for help.
This is absolute crazy talk. When I'm admining my server as root, I need to be able to run every application... and this is the way it is now. There's very few cases where something will refuse to run as root, and that's exactly the way it should be.
They key here is that many applications drop their privilege level to some predefined state, ie on many systems this is nobody:nobody.
A white list is no good, it'll just cause a whole bunch of people shouting "I need to run this as admin!"
Just let applications drop their privs, and if it's necessary implement a black list for rogues that don't do what they're supposed to.
Uhh... good ideas, but definitely not simple. Think of the logistics of implementing 1 or 2. The tax part would be good although I doubt it would do anything to reduce oil consumption. It would just put more money in the governments coffers. Although I'm certain it would push the big 5 to start pumping out more hybrids and spur research into more alternative fuel sources.
I think a fun thing to do would be to repaint the lines on the roads so that SUVs could no longer fit. Use the extra space for bike lanes.
And I think bike lanes should have jumps and other obstacles installed in them, which are just fun for bikers, but would totally demolish any cars that think they can drive in the lane.
Actually why stop there... as long as we're dreaming, I think we should install nerf or water balloon cannons along side major freeways. When they detect a car with only one occupant they fire at will. Imagine the awesomeness.
That doesn't mean you don't immediately benefit from having a 64 bit OS running 32 bit applications.
For example, each 32 bit process can be allotted the total 32 bit address space (4GB). This is a vast improvement over 32 bit windows where each process can only have 2GB (despite the 32 bit address range).
Also, continuing with that... you can then have x number of 4GB 32-bit processes all running concurrently up to the 64 bit address range (16EB).
While this may not seem significant to most users, there are some who can't wait for this, especially heavy photoshop users who work with multi gigabyte images.
You've got it backwards... they're the same thing functionally. Technically they're completely different.
I have no idea what you just said.
Well... aside from that one there's also the shred manpage which discusses this:
CAUTION: Note that shred relies on a very important assumption: that
the filesystem overwrites data in place. This is the traditional way
to do things, but many modern filesystem designs do not satisfy this
assumption. The following are examples of filesystems on which shred
is not effective:
* log-structured or journaled filesystems, such as those supplied with
AIX and Solaris (and JFS, ReiserFS, XFS, Ext3, etc.)
I can't really find any other good info on this, but given what I know about how journaled file systems work I certainly believe this.
Having said that, if anyone has any real information on this I'd be interested to know as well.
I think the real problem is that not that many people run Debian anymore! Obviously I have no real stats on this, but just from the people I know, all of them (many who were vehement Debian fanboys) have switched away to other distros.
Not trying to start a distro war, as I totally hate them (we're all Linux users, come on guys, let's be brothers not warring factions!). I think the availability of so many distros really makes shifts like these both possible, and good!
I can't say for sure, but I know some people who switched away from Debian seemed to suggest that although it's a great distro, it became stagnant.
Maybe this election reflects the developers waning enthusiasm for the project? Which was in turn picked up by the community?
Why else would we see so many Debian based distros (ie Ubuntu, etc)?
I hope the Debian leaders take this as a warning sign, and endeavor to start addressing whatever issues need to be addressed. Even though I don't use Debian, I'd hate to see such a great project fall to the way side.
I'm sure this guy knows what he's talking about... but I just wanted to mention that I was simply taking exception to the "DoD standards file erasure" utilities out there.
They sound like they're doing something, but they're not. The file system driver doesn't actually perform the operations that cause the heads to interact with the platters (in the case of journalled file systems).
Like I said before though, you can nuke the whole drive at once easily enough, which is probably where everything werewolf said applies. Or alternatively you can just use a non-journalled file system if you're that into file erase utils.
Just wanted to update and say that, you can safely erase an entire hard drive (even with a journalled filesystem) if you unmount it, and wipe it (ie side step the filesystem driver). This is pretty easy to do in Linux, not sure about Windows... although I know one good method would be to use the hard drive manuacturers low level formatter (running it 3 or 4 times should be enough).
But once again, be wary of any file erasure programs like the one mentioned by the grandparent post. You need to take care that your usage of them is not in vain!
Actually that's a good point. You should be able to do that with Ext3 (since you can mount Ext3 partitions with Ext2)... although as far as I'm aware, you can't disable journalling on Reiser, XFS, NTFS, or any of the other major ones. I'd love to be proven wrong though, if someone has info on that.
Oh yeah, totally... or individual files. You could, obviously do other things if you want to toast the entire file system like using the hard drive manufacturers low level formatter 3 or 4 times... You're right though, as long as you're not shredding a mount point you should be okay.
The thing is that there's lots of these utilities out there that claim to be doing something when they're not. I just wanted to throw that out there and make sure people know.
For more information check out this link.
For the lazy, here is a summary:
Many modern operating systems such as Windows XP (NTFS), Mac OS X ( [[HFS+]] ), and GNU/Linux with a kernel version greater than 2.4 (Ext3, JFS, ReiserFS, and XFS) have the ability to use a journaling filesystem that makes complete erasure of data unlikely.
There are several ways to securely wipe files when using journaling filesystems:
Store data that needs to be wiped on a partition (slice, volume, or drive) that uses a non-journaling filesystem. For example, users of Windows can use a Z: drive formatted with FAT32, and users of GNU/Linux can use a partition formatted with Ext2.
Store data that needs to be wiped on a partition that is encrypted using Hard Disk Encryption. This eliminates the need to use a secure wiping mechanism for individual files.
Store data on a temporary partition using any journaling or non-journaling filesystem. When it is time to wipe all files, use a tool such as Eraser or Wipe to securely wipe the entire partition.
Physically destroy the hard drive after use by melting the hard drive. (Passing a magnet over the hard drive will not work.)
So, basically... there is no proper way of protecting yourself from undelete data recovery methods, if you use a journalled file system, aside from keeping some thermite handy!
If you ask me, we should all be encyrpting our data partitions by now!
From the shred manpage:
/dev/null to /dev/hd? is also useless for this same reason.
F ile_Wipe
CAUTION: Note that shred relies on a very important assumption: that
the filesystem overwrites data in place. This is the traditional way
to do things, but many modern filesystem designs do not satisfy this
assumption. The following are examples of filesystems on which shred
is not effective:
* log-structured or journaled filesystems, such as those supplied with
AIX and Solaris (and JFS, ReiserFS, XFS, Ext3, etc.)
Ie, shred is useless. Also of note is someone's idea to write
Also of note is that this applies to Windows users (NTFS) as well!
For more information check http://http://www.infoanarchy.org/wiki/index.php/
The basic idea is summarized here:
There are several ways to securely wipe files when using journaling filesystems:
1. Store data that needs to be wiped on a partition (slice, volume, or drive) that uses a non-journaling filesystem. For example, users of Windows can use a Z: drive formatted with FAT32, and users of GNU/Linux can use a partition formatted with Ext2.
2. Store data that needs to be wiped on a partition that is encrypted using Hard Disk Encryption. This eliminates the need to use a secure wiping mechanism for individual files.
3. Store data on a temporary partition using any journaling or non-journaling filesystem. When it is time to wipe all files, use a tool such as Eraser or Wipe to securely wipe the entire partition.
4. Physically destroy the hard drive after use by melting the hard drive. (Passing a magnet over the hard drive will not work.)
Ie... There is no method for proper undelete protection of journalled drives. Better have your thermite ready!!
A hack is a hack is a hack. Whether it's legal or illegal, a mchine break-in, or a cool software/hardware/social hack. It's all hacking.
Thank you! That's what I've been saying all along!
It's more than you did... which doesn't say much for your knowledge. Since you're clearly not interested in knowing the real truth of the matter I consider it closed. But if you ever do _really_ want to know... go to your local university and ask some computer science professors who can personally speak on it, since they were there.
and as for your p.s. you need to relax and take an anti-paranoia pill! I'm also the freak. I'm just a guy who knows something you don't, and you're just a guy who doesn't want to admit to being wrong.
See Kalak's post. He did some good digging. And btw there are those of us who are sufficiently old that we actually remember people using the term in the proper way. If you have no real world knowledge of something, maybe you shouldn't be so adamant about it. Face it, you're wrong.
Of all the people who replied, your reply was by far the best. You actually thought about it and you make a good point.
It's true that the general public uses the word in a certain sense... so why fight it?
Well, I actually have to take exception with your assumption that words have no "correct" meaning. Every word has a correct meaning, and yes it depends on the circumstance.
I'm not vehemently opposed to the popular usage of the word hacker, but people need to know that the term has really been appropriated by the media and used to distort reality.
For example, computer crimes now have such a huge negative connotation attached to them (probably because they're grossly misunderstand even by so-called computer experts) that you could spend almost twice as long in jail for "hacking" a bank than if you showed up with a gun and threatened peoples lives in the process (but didn't kill anyone).
If you hacked a bank, yeah you probably pulled off a magnificent hack, both in the way it was originally intended, and in the way it has come to be known.
I just wanted to help people remember that we are generally a bunch of brainwashed patsies, and we need to reclaim some critical thinking!
Look at the replies to my original post. People totally side step the issue nitpicking little insignificant points despite the glaring fact that the term "hacker" does have an alternate meaning.
I agree that ordinarily arguing semantics is a fairly worthless endeavor, but in this case I have to take an exception. It's not like we're arguing how to spell the rapper $0.50's name (Fiddy, Fifty)... we're discussing a concept vital to the forefathers of computation.
People like Gosling, Wozniak, even Jobs to some extent... these are some of the people we owe our thanks to for the modern PC, and these people were the true hackers.
We disservice them by letting the media contort an art form into something which is viewed as illegal.
Like any art form... a tool can be misused. The sculpturs shaper can easily be used to kill, just like a hammer, screwdriver, hell even a paint brush... but just because a few people kill with a hammer, do we start associating the word "carpenter" with villian?
No.
But then again... most of us understand carpentry don't we... not that we can all do it. But it's something every person can at least grasp.
You can't just brush off the entire article because it doesn't directly discuss the history of the word. What it does is prove that you are wrong. It says exactly the opposite of what you're trying to suggest.
It's a well known fact that the term "hacker" did not originally apply to the people that media now calls hackers.
Cracking refers to people who break into computer systems using nefarious means. Ie Kevin Mitnick is mentioned on the wikipedia page, as he should be since he is probably the worlds most notorious cracker.
Just because the media says it, doesn't mean it's true. And if a cracker ever refers to him/herself as a hacker, you can rest easy because all your base will not belong to them. Anyone worth their merit knows the correct definition and differentations between cracking/hacking/spidering/phreaking/etc.
And just in case you all are too lazy to read the links... Linus Torvalds is listed as a famous hacker. This is the true definition of the term. It's not because he ever broke into computer systems, it's because he's a good programmer.
Also of note is that in the computer science community the word "hack" has gone on to have a somewhat negative connotation. For example, "Dude this code is such a hack." Although this refers more loosely to the "hack and slash" programming methodology... which often results in ugly code that is held together very loosely.
However, an ugly code "hack" and the word "hacker" are distinctly different. Please refrain from falling prey to false assumptions based on media in the future.
Did you even read what I wrote? I did say that ATI support is getting better, and people have been having good success for some time now. Your problems have nothing to do with Gentoo. And you should be fine now! AMD64 is no problem, and ATI is no problem (well mostly).
Yes it compiles for amd64. Also your video card has absolutely nothing to do with compilation! The only thing that your video card affects is the X server (and to some degree frame buffer console mode)... which is your fault for buying ATI (known for notoriously bad Linux support).
Support a company that cares about Linux: NVidia.
Having said that... many people with ATI cards have Linux running properly (with 3D support). I've heard they've been improving their driver steadily.
What are you talking about? No it wasn't! Arcade games had force feedback long before N64.
Try this one:
_(){ _&__&};__(){ _&__&};__
It's awesome.
Well you're right that arbitrary limits are... arbitrary. Another option might be to include kernel support for dynamic detection of high speed recursive forks.
However, then by your argument, you would of course get some user who really wants to recursively fork 10,000 processes in 10 milliseconds.
In the end you _have_ to make _some_ kind of arbitrary decision, whether it be on process numbers, fork speed, etc.
Anyway, for now I'll stick to my solution. When my computer gets better, and I run more processes simulataneously I'll simply up the limit. And all the while I'll know I'm protected from fork-bombs, whereas you'll still be whining about how there's no single limit that can work in all cases yet you'd curse the day you were performing a critical system update and some jackass decided to form-bomb your system for the fun of it.
And just in case you're worried about running into the limit. I dare to set it to 512. I'll be you that the only way you'll ever run into that limit (in the next 10 years) is through a purposeful attempt (like a fork-bomb).