Cause wildcards aren't actually recognised by the kernel. It's entirely up to your shell what characters are used as wildcards, so why should the kernel discriminate against weird shells by only recognising the wildcards used by the bourne and C shells? Or should they just outlaw anything that might be a wildcard? What if I wrote a shell that used 'e' as its wildcard?:)
The ability to type the options after the filename is a nonstandard and undocumented extension of GNU rm. To anybody who learnt unix shell commands with any OS other than Linux it is nonintuitive and probably won't work with all their systems.
You'll be dumber and happier with a second-rate education overseas anyway.
The U.S. has plenty of talent both at the top and the bottom of the workforce scale.
You may think this, but historical evidence shows that the U.S. depends on educated immigrants to achieve all of the things it is most famous for. Moon landings, the world's most famous media companies, pizza... you'd have none of these things without immigrants.
And I'm quite happy with the British education I have received. So, too, were the numerous American students who had come over here to study at my university... I wonder why they did that if American education's so brilliant?
I've just read the patent, and it seems that for most of these companies (except Sega), the code that performs the patented process would be found in Microsoft's DirectX libraries.
Does these mean that Microsoft will indemnify them against any potential losses? I mean, Ballmer was going on about the need for users of open source technologies to acquire IP indemnification insurance, so presumably as MS users these people will be protected by MS, otherwise they'd need insurance too and Ballmer wuld have been takling crap, which is unimaginable, isn't it? Isn't it? Hello?
The patent was originally filed in 1984, and refiled in 87. This means it predates both of the games you quote; there's nothing specific in the last article that suggests anybody was doing the same thing previously, which is effectively storing a 3d transformed world and calculating the differences when you move the viewpoint... everything previously mentioned _might_ have been recalculating from scratch with every frame.
Is there a particular reason you a) decided this needing saying again, despite the fact that at least 3 people had said it before you, and b) felt the need to be so rude about it?
Or are you just a jumped up prick who thinks he never makes mistakes?
I mean, it's not as if there aren't two worse errors in the post I was replying to. And yes, I have used "it's" correctly there, because I do actually know how to use English properly, unlike some people around here.
PHB: Why have you asked for this additional hardware budget? Sysadmin: Well, we have this disk array that's getting kind of old. We did have a spare controller, but the original died last week and we now can't get a compatible replacement. PHB: So, why do you need a new controller and eight new disks? Wouldn't just a new controller suffice? Can't you replace the disks on a one-by-one basis as they fail? Sysadmin: Because if we plug the old disks into the new controller, we won't be able to get the data that's on them. We'll then have to restore from backups, which is a time consuming process that would leave the servers unavailable for an entire day. What we should do is set up a new array now, and copy the data across to it, which is much faster. PHB: How long is this controller we have likely to last for? Sysadmin: Well, it's solid state, so it'll probably last at least 5 years. PHB: I've always said, "if it ain't broke, don't fix it". Come back to me in 3 years when it's starting to look like it might be a problem soon. Sysadmin: ??
This scenario can be avoided using software raid, however incompetent your PHB is.
Hmmm... I'd checked that initially, but while looking at it again, I've not set it right on the _new_ partition... that is the one that it is adding to the array properly every time. Perhaps because that isn't set right, it isn't looking for the other ones. I'll have to wait until the system isn't in use to reboot it and find out, though.:)
I don't care if you have to pay for the replacement, that isn't the point. The point is that after 10 years they probably won't be able to replace it with a compatible board.
I recently attempted to set up a software RAID mirroring scheme on my Linux box (late 2.4 kernel; I forget which and I don't have access at the moment). Because I already had data on one of the disks I wanted to use, this is the technique I used to set it up:
- partition new, empty disk to create a partition the size I wanted my mirrored array to be - set up/dev/md0 as an array with the newly created partition and a 'missing' device - copy the data onto/dev/md0 - repartition the old disk - add the partition on the old disk to the array
The system then did a rebuild, copying the data from the new partition to the old one as desired. It worked perfectly until I rebooted, at which point it forgot about the second disk.
Of course, the fact is that what you're doing isn't RAID. RAID isn't a backup solution -- if somebody nicks your computer or it catches fire or something you'll lose all of your data. What you're doing is offsite data replication, which is much better.
I use a combination of offsite replication and periodic backups to DVD-R, with occasional deltas being stored on CDRW until they get too large, at which point a new DVD-R is written.
Generally problems happened when the box was under heavy load - one of the disks would be marked bad, and a painful rebuild would ensue.
Were you running kernel 2.4.20? I ran this kernel for several months, and got corruption in a reiserfs file system and a mysql data file. Turns out there was a bug that sometimes caused incorrect data to be written when under heavy load in that kernel version. Note that this is completely independent of RAID usage, and might even have been _worse_ with hardware RAID because the same corrupted data would have been written to all volumes, not just one.
tar:/ is no more intuitive than double clicking on the.tar file and opening it.
To explain why this is better than existing implementations of similar features, I'll use the most similar other implementation I'm aware of: Windows XP's Compress Folder Shell Extension. This and the tar: protocol handler both provide what is, at least at a superficial level, the same capability: to look inside a compressed archive and see what files it contains, and perform some manipulation on them.
The differences come about, though, when you stop using the shell and try to open one of the files in another application. You see, the Windows XP implementation cannot do this without first copying the file to a temporary folder. This means that (a) when you try to open a file there is typically a delay while it is extracted before the application is launched, and (b) you cannot drag and drop files out of a "compressed folder" into any application other than explorer.
However, what KDE has done is provide a protocol where the shell (i.e. konqueror) can specify to another KDE application (e.g. kwrite) the location of that object within the tar file. So you can actually open the object without making a copy of it. You can drag and drop from konqueror to any other KDE application (that is willing to accept drag and drop at all).
It is more efficient, and it works more like the user expects. This is why it is good.
If you mean the kernel VFS layer, then Apple is not doing it right
OSX is a microkernel. While I'm not aware of the actual implementation details, this _should_ mean that file systems are implemented by server processes running in userspace talking to the kernel through IPC primitives.
Extreme Pumpkins
on
Halloween Fun
·
· Score: 4, Interesting
This is cool. Or rather hot. Particularly where the flames are coming out.
Unfortunately, this doesn't sound like R&D. This sounds like taking a step backwards and losing the capacity for manned flight for the foreseeable future.
But are the reusable rockets rated for manned space-flight?
a) The story says disposable, not reusable b) Doesn't look like it -- the article mentions relying on Soyuz (and potentially Shenzhou) for manned flights in future.
I wonder how long before ATMs start to spit money left and right all of a sudden.
Happened to Coventry Building Society last year. They threatened legal action against customers who kept the additional money that the machines gave them by mistake.
hmm, would it be possible to upload a virus to an atm via a magnetic strip on a card?
I would say no, but I used to say you couldn't get a virus from an e-mail or a document, and MS proved me wrong on both counts. I just hope they aren't writing the ATM software as well as the OS it runs on...
Slashdot Mirror
Cause wildcards aren't actually recognised by the kernel. It's entirely up to your shell what characters are used as wildcards, so why should the kernel discriminate against weird shells by only recognising the wildcards used by the bourne and C shells? Or should they just outlaw anything that might be a wildcard? What if I wrote a shell that used 'e' as its wildcard? :)
The ability to type the options after the filename is a nonstandard and undocumented extension of GNU rm. To anybody who learnt unix shell commands with any OS other than Linux it is nonintuitive and probably won't work with all their systems.
You'll be dumber and happier with a second-rate education overseas anyway.
The U.S. has plenty of talent both at the top and the bottom of the workforce scale.
You may think this, but historical evidence shows that the U.S. depends on educated immigrants to achieve all of the things it is most famous for. Moon landings, the world's most famous media companies, pizza... you'd have none of these things without immigrants.
And I'm quite happy with the British education I have received. So, too, were the numerous American students who had come over here to study at my university... I wonder why they did that if American education's so brilliant?
I've just read the patent, and it seems that for most of these companies (except Sega), the code that performs the patented process would be found in Microsoft's DirectX libraries.
Does these mean that Microsoft will indemnify them against any potential losses? I mean, Ballmer was going on about the need for users of open source technologies to acquire IP indemnification insurance, so presumably as MS users these people will be protected by MS, otherwise they'd need insurance too and Ballmer wuld have been takling crap, which is unimaginable, isn't it? Isn't it? Hello?
The patent was originally filed in 1984, and refiled in 87. This means it predates both of the games you quote; there's nothing specific in the last article that suggests anybody was doing the same thing previously, which is effectively storing a 3d transformed world and calculating the differences when you move the viewpoint... everything previously mentioned _might_ have been recalculating from scratch with every frame.
Is there a particular reason you a) decided this needing saying again, despite the fact that at least 3 people had said it before you, and b) felt the need to be so rude about it?
Or are you just a jumped up prick who thinks he never makes mistakes?
I mean, it's not as if there aren't two worse errors in the post I was replying to. And yes, I have used "it's" correctly there, because I do actually know how to use English properly, unlike some people around here.
Yeah, I know, and I'm constantly correcting other people who make the same mistake. And it's instead of its, and so on.
Zero replies and slashdotted. Well done, Tech Zone. That must be some hot server setup you have their. ;)
PHB: Why have you asked for this additional hardware budget?
Sysadmin: Well, we have this disk array that's getting kind of old. We did have a spare controller, but the original died last week and we now can't get a compatible replacement.
PHB: So, why do you need a new controller and eight new disks? Wouldn't just a new controller suffice? Can't you replace the disks on a one-by-one basis as they fail?
Sysadmin: Because if we plug the old disks into the new controller, we won't be able to get the data that's on them. We'll then have to restore from backups, which is a time consuming process that would leave the servers unavailable for an entire day. What we should do is set up a new array now, and copy the data across to it, which is much faster.
PHB: How long is this controller we have likely to last for?
Sysadmin: Well, it's solid state, so it'll probably last at least 5 years.
PHB: I've always said, "if it ain't broke, don't fix it". Come back to me in 3 years when it's starting to look like it might be a problem soon.
Sysadmin: ??
This scenario can be avoided using software raid, however incompetent your PHB is.
If a piece of toast falls butter side down on the moon and there's no one there to hear it, does it still make a splut?
No, but it'll still end up with pet hair in it, and nobody will ever be able to figure out where they came from.
Hmmm... I'd checked that initially, but while looking at it again, I've not set it right on the _new_ partition... that is the one that it is adding to the array properly every time. Perhaps because that isn't set right, it isn't looking for the other ones. I'll have to wait until the system isn't in use to reboot it and find out, though. :)
I don't care if you have to pay for the replacement, that isn't the point. The point is that after 10 years they probably won't be able to replace it with a compatible board.
I recently attempted to set up a software RAID mirroring scheme on my Linux box (late 2.4 kernel; I forget which and I don't have access at the moment). Because I already had data on one of the disks I wanted to use, this is the technique I used to set it up:
/dev/md0 as an array with the newly created partition and a 'missing' device /dev/md0
- partition new, empty disk to create a partition the size I wanted my mirrored array to be
- set up
- copy the data onto
- repartition the old disk
- add the partition on the old disk to the array
The system then did a rebuild, copying the data from the new partition to the old one as desired. It worked perfectly until I rebooted, at which point it forgot about the second disk.
Any suggestions what I might have done wrong?
Of course, the fact is that what you're doing isn't RAID. RAID isn't a backup solution -- if somebody nicks your computer or it catches fire or something you'll lose all of your data. What you're doing is offsite data replication, which is much better.
I use a combination of offsite replication and periodic backups to DVD-R, with occasional deltas being stored on CDRW until they get too large, at which point a new DVD-R is written.
Generally problems happened when the box was under heavy load - one of the disks would be marked bad, and a painful rebuild would ensue.
Were you running kernel 2.4.20? I ran this kernel for several months, and got corruption in a reiserfs file system and a mysql data file. Turns out there was a bug that sometimes caused incorrect data to be written when under heavy load in that kernel version. Note that this is completely independent of RAID usage, and might even have been _worse_ with hardware RAID because the same corrupted data would have been written to all volumes, not just one.
Of course, 3 years is nothing. If it were a ten year old board they had been able to replace just like that, then I'd be impressed.
Gentoo - Klingons. ...
MacOSX [...] tribbles
Does this mean Gentoo users really _really_ despise OSX?
tar:/ is no more intuitive than double clicking on the .tar file and opening it.
To explain why this is better than existing implementations of similar features, I'll use the most similar other implementation I'm aware of: Windows XP's Compress Folder Shell Extension. This and the tar: protocol handler both provide what is, at least at a superficial level, the same capability: to look inside a compressed archive and see what files it contains, and perform some manipulation on them.
The differences come about, though, when you stop using the shell and try to open one of the files in another application. You see, the Windows XP implementation cannot do this without first copying the file to a temporary folder. This means that (a) when you try to open a file there is typically a delay while it is extracted before the application is launched, and (b) you cannot drag and drop files out of a "compressed folder" into any application other than explorer.
However, what KDE has done is provide a protocol where the shell (i.e. konqueror) can specify to another KDE application (e.g. kwrite) the location of that object within the tar file. So you can actually open the object without making a copy of it. You can drag and drop from konqueror to any other KDE application (that is willing to accept drag and drop at all).
It is more efficient, and it works more like the user expects. This is why it is good.
If you mean the kernel VFS layer, then Apple is not doing it right
OSX is a microkernel. While I'm not aware of the actual implementation details, this _should_ mean that file systems are implemented by server processes running in userspace talking to the kernel through IPC primitives.
This is cool. Or rather hot. Particularly where the flames are coming out.
It does sound from the article like there won't be a replacement any time soon.
BTW: "In lieu of" doesn't mean what you seem to think it means.
Unfortunately, this doesn't sound like R&D. This sounds like taking a step backwards and losing the capacity for manned flight for the foreseeable future.
But are the reusable rockets rated for manned space-flight?
a) The story says disposable, not reusable
b) Doesn't look like it -- the article mentions relying on Soyuz (and potentially Shenzhou) for manned flights in future.
I wonder how long before ATMs start to spit money left and right all of a sudden.
Happened to Coventry Building Society last year. They threatened legal action against customers who kept the additional money that the machines gave them by mistake.
hmm, would it be possible to upload a virus to an atm via a magnetic strip on a card?
I would say no, but I used to say you couldn't get a virus from an e-mail or a document, and MS proved me wrong on both counts. I just hope they aren't writing the ATM software as well as the OS it runs on...