Because Unix has a crude "all or nothing" security model - you're either the sysadmin, or you aren't.
That's not strictly true. While it doesn't have the fine grained security of, say, VMS (at least, not without using the various capabilities options available), it's not necessarily all or nothing. You can do an awful lot with sensible use of groups. In my experience, 99% of the times that people claim they need root access, they don't. It's just the quick and easy way out, so they take it.
Am I the only one that thinks Kill 'Em All isn't such a great album? With a few exceptions (e.g., Whiplash), it's really not up to much. Now Ride The Lightning, on the other hand, is an entirely different matter...
Therefore, I'm very glad to find Deneba packaged Canvas7 beta for Linux to be installed in/usr/local.
Well, I agree it's better than/usr, but/usr/local is reserved for me to use (according to the FHS), and not for third party software to install into. It should install into/opt/canvas7.
That said, I personally have a shared/opt across all my boxen, so I've had to impose some structure on it. I'd want it installed in/opt/x86/linux/canvas7 (/opt/sparc/linux/canvas7 would be great, too, but I guess I'm being too optimistic there). Ideally, any RPMs should be relocatable to allow for setups like mine.
If you're allowing root access to individual boxen, then I can't see any way of preventing access. No matter what you do with NFS (or Samba, for that matter), the user can simply su to the user whose home directory they wish to access, and then access it as the user themselves. I have to ask, though, why do individuals need root access to their personal boxen?
As far as I know [...] Multics was the only OS ever certified at the "B" level.
Nope. DG/UX has offered a B2 security option for many years now. Trusted Solaris and Trusted IRIX have both been certified at B1, along with several others. Even Trusted Xenix (!) of all things managed to get a B1 rating in the early 1990s. Personally, I wouldn't trust Xenix with anything. It ranks right up there with Interactive Unix as one of those OSes I'm glad I'll never have to use again:-)
It's not an issue of trying to run 40k+ copies of Linux, it's about having a stable, fast, and large platform to run an OS.
Yes and no. The 40k copies is one of those unfortunate statistics that leaks out every now and then, but has no practical value in the real world. The fact that it can run 40000 is irrelevant. However, the fact that is can run many copies is relevant. 40000 wouldn't be practical. 100 concurrent copies, on the other hand, is both practical and useful.
Well, every time there's a story on the RIAA or MP3s, there's always a million posts saying, "It's not stealing because the author still has it.
Whether or not the author still has it, copyright is still being infringed. I object to the RIAA because they're claiming that the MP3 format itself should be outlawed, because it makes copying music too easy. I personally still buy the CD, even if I could theoretically download it from the net instead. I do download MP3s, but only to see if I like the band enough to buy the CD.
I wouldn't condone copying MP3 music, just as I wouldn't condone cloning a website. It's absurd to claim that a site has zero value, just as it's absurd to claim that an MP3 doesn't have value. Claiming that MP3 as a format is inherently evil is an entirely different matter, however...
So how does this work in Oz where the population density is low.
Actually, it's not that low, at least, not if you ignore the middle of the country which is basically empty anyway. The vast majority of Australians live in one of the major cities (off the top of my head: Sidney, Melbourne, Brisbane, Perth and Canbera). Admittedly, for a city of only 3 million people, Melbourne occupies a huge amount of space, but the chances of not being within 10 miles of a local exchange are fairly slim. Of course, the rural areas are always going to lose out...
Finally, an eccentric englishman I can look up to!
Didn't you read the article? He's from Sheffield. They're barely human up there, let alone Englishmen! As the saying goes, it's grim up North!:-) BTW, is Alan Cox not eccentric enough for you?
I really hope the price of gas keeps climbing up to $2- $3 dollars a gallon here in the U.S. Maybe that will dissuade people from buying gas-guzzeling SUV's.
It won't work. Petrol (gas to those of you that choose to so describe the liquid with which you fuel your car) costs the equivalent of nearly US$6/gallon here in the UK, and the roads are still full of them.
My only regret is that I can't convince my department to drop support for them completely and pick up a more worthy UNIX (Like DG/UX. Man they rock...)
Moving from SCO to DG/UX isn't trivial. Unless you're already running SCO on a DG AViiON, you'll need to buy new hardware. Athough DG/UX can be persuaded to run on a standard PC, they won't sell it to you unless it's to run on one of their own machines. I agree that the DG/UX kernel is very nice, I just wish they'd ship a more complete userland. BTW, anyone know how to trace system calls on DG/UX (there's no truss, strace, ktrace etc.)?
NetBSD is probably the closest I'll get to running a decent Unix on my NeXT. However, although it has support for 68040 based machines, it explicitly doesn't support my Turbo Colorstations:-( The Linux port to NeXT hardware is still a long way from usability. Guess I'm stuck with NeXTStep/OpenStep for now.
The mozilla builds I have been playing with are still nowhere near stable enough to approach a full release, even if the code is labelled "beta".
Yep, I'm afraid I have to agree here. I've been using M14 as my primary browser recently, and although it's far, far better than the previous versions I'd looked at (M10/M11), it's still not at the point where I'd say it was ready for public release, even as a beta. It still crashes occasionally, without warning, and comletely unreproducably (this is on Win32, admittedly:-) Still, it's showing a lot of promise. I just hope AOL/NSCP know what they're letting themselves in for with a beta release of this quality. The MS-sympathetic parts of the press are going to give them a really hard time unless things rapidly improve over the next month.
An example of this lies in the vim package where there are a lot of different vim executables (plain vim, vim w/ python bindings, vim w/ perl bindings, etc) that all provide 'vim'. Any other program requiring any version of vim can rest easily:-)
Yep, that sounds like a really nice feature. All those packages can provide "vim", on which other packages can depend. That way, it doesn't matter which vim you've got installed, so long as at least one of them is... which is why RPM provides exactly the same functionality (you'll need a version newer than 3.0.3, but that's nearly a year old now -- almost prehistoric in Linux terms:-)
You download a source package guaranteed to compile on your system so you get an optimized system
You're thinking too much like a home user. Corporate users don't want to have a software development kit on every machine. Also, if it needs compiling, it's going to be too slow for any moderately large app.
Of course the dependency checks of either blow the pants off RPMS. Care to give some examples? I've yet to see any real world examples where.deb dependencies are better than RPM. Yes,.deb can have more complex boolean expressions in dependency checking, but in real life, RPMs dependencies are sufficient to achieve everything that's needed.
It looks like the LSB is going to want everyone to use RPM, so we have to take that into account.
This isn't quite true. The idea is to have a standard LSB packaging format, which can be used for distribution-neutral packages. Each distribution will be responsible for providing suitable tools for converting LSB packages to their native package format, and handling their subsequent installation. It is true that the current plan is to use RPM as the basis for LSB packages, simply because it's so widely supported already. As the spec says: If you don't like this, then please propose an alternative.
Re:XFree 4.0? Won't they have a problem with this?
on
SuSE 6.4 Announced
·
· Score: 3
No, they won't have a problem with it -- XFree86 has been officially released (albeit with limited supported cards, which is presumably why SuSE 6.4 doesn't use it by default).
Re:SuSE making inroads in the North American marke
on
SuSE 6.4 Announced
·
· Score: 2
KDE rather than GNOME (I prefer it)
Why is this a problem with the others? I'm not aware of any current distributions that ship GNOME but not KDE. The only difference is the GNOME is the default in Red Hat, while KDE is the default in SuSE.
SuSE may not be just for Europeans any more.
People in the US seem to assume that everyone in Europe uses SuSE. While it may have a stranglehold on the German market, that's certainly not the case in the UK. Yes, it's available, but Red Hat seems to be more prevalent.
Overclocking SMP is NOT suicide [...] What's the risk?
The risk is both damage to the physical hardware and data corruption. The hardware can easily be replaced when it's a cheap Celeron, but not when it's a dual core IBM Power CPU. The data corruption can't be ignored, though. Don't believe me? Maybe you'd like to hear it directly from someone you might trust.
Do you know any engineers? They overestimate everything, like Scotty telling Kirk how long repairs will take. You bet that you can run that chip faster than it is rated.
Yes, you can, if you're prepared to take the risk -- that's the whole basis of overclocking. Chips are rated at the speed the manufacturer can guarantee they'll operate as intended. Say you overclock your chip by 15%. You're now encroaching into the safety margin that the engineers and the manufacturer allowed to be sure that all chips will work correctly. Even so, perhaps 98% of all chips will be OK. Do you want to gamble on whether or not you've got one of the 1 in 50 chips that won't work? Personally, I don't like the odds, particularly when the chips cost as much as this one will...
Overclocking stupid, eh? Actually, many PowerPC processors overclock quite well, I know from personal experience.
Maybe they do, maybe they don't. You're missing the point though. If you want faster speeds, go buy faster processors (or more of them). Overclocking is only for those who can't afford to do that. People buying these chips aren't going to fall into that category.
The other point to consider is that overclocking an SMP system is tantamount to suicide, by all accounts. Now maybe that won't be the case here, because the cores are on the same die, and hence will be affected in exactly the same way, but I don't know enough about it to be sure, and I certainly wouldn't risk it.
Yeah, it would be compromised, but given the patents on codecs, you'd never be able to distribute the software legally. I hate to say it, but it's pretty damned effective protection.
Just like the Frauhofer patents have prevented the distribution of MP3 encoding software? OK, so I know the situation is slightly different for MP3, but it's incredibly difficult to prevent the spread of software whose time has come, legal or not...
By free, I assume you mean open. The problem with open codecs is (at least in the eyes of content providers) copy protection. They see an inability to save streamed RealAudio/ReadVideo, and are happy because their copyrighted works aren't being distributed around the net. An open codec would allow anyone to save the streamed data to a file for later use. In actual fact, the current closed codecs only provide security through obscurity anyway, and will eventually be compromised.
Get rid of it and get better software that downloads the headers and lets you mark obvious spam so you don't waste bits downloading it later.
Firstly, my ISP only supports POP3 mail access. They're moving to IMAP, but it's not there yet. Secondly, assuming that I have access to my mail via IMAP, what software should I be using? fetchmail is working fine for me at the moment, and will continue to do so, but I don't know how I could persuade it to downlaod headers only and then selectively choose which messages to junk and which to download. So what else is there?
I have a little button on my heyboard that corresponds to a function known as a "delete message" function that really works wonders.
That's nice for you. However, before you can get to that stage, you have to download the message. Those of us in the UK (and, in fact, most of Europe) don't have free local calls. The upshot is, that to download a 2.5MB spam (yes, I've been spammed with PDF product catalogues this size before) over my modem costs me the equivalent of about US$2.00. Now if someone shoved some junk mail through your door with a bill for $2.00 that you had no choice but to pay, would you complain? No? Then please send me your address, I've got a once in a lifetime opportunity for you...
Slashdot Mirror
That's not strictly true. While it doesn't have the fine grained security of, say, VMS (at least, not without using the various capabilities options available), it's not necessarily all or nothing. You can do an awful lot with sensible use of groups. In my experience, 99% of the times that people claim they need root access, they don't. It's just the quick and easy way out, so they take it.
Am I the only one that thinks Kill 'Em All isn't such a great album? With a few exceptions (e.g., Whiplash), it's really not up to much. Now Ride The Lightning, on the other hand, is an entirely different matter...
Well, I agree it's better than /usr, but /usr/local is reserved for me to use (according to the FHS), and not for third party software to install into. It should install into /opt/canvas7.
That said, I personally have a shared /opt across all my boxen, so I've had to impose some structure on it. I'd want it installed in /opt/x86/linux/canvas7 (/opt/sparc/linux/canvas7 would be great, too, but I guess I'm being too optimistic there). Ideally, any RPMs should be relocatable to allow for setups like mine.
If you're allowing root access to individual boxen, then I can't see any way of preventing access. No matter what you do with NFS (or Samba, for that matter), the user can simply su to the user whose home directory they wish to access, and then access it as the user themselves. I have to ask, though, why do individuals need root access to their personal boxen?
Nope. DG/UX has offered a B2 security option for many years now. Trusted Solaris and Trusted IRIX have both been certified at B1, along with several others. Even Trusted Xenix (!) of all things managed to get a B1 rating in the early 1990s. Personally, I wouldn't trust Xenix with anything. It ranks right up there with Interactive Unix as one of those OSes I'm glad I'll never have to use again :-)
Yes and no. The 40k copies is one of those unfortunate statistics that leaks out every now and then, but has no practical value in the real world. The fact that it can run 40000 is irrelevant. However, the fact that is can run many copies is relevant. 40000 wouldn't be practical. 100 concurrent copies, on the other hand, is both practical and useful.
Whether or not the author still has it, copyright is still being infringed. I object to the RIAA because they're claiming that the MP3 format itself should be outlawed, because it makes copying music too easy. I personally still buy the CD, even if I could theoretically download it from the net instead. I do download MP3s, but only to see if I like the band enough to buy the CD.
I wouldn't condone copying MP3 music, just as I wouldn't condone cloning a website. It's absurd to claim that a site has zero value, just as it's absurd to claim that an MP3 doesn't have value. Claiming that MP3 as a format is inherently evil is an entirely different matter, however...
Actually, it's not that low, at least, not if you ignore the middle of the country which is basically empty anyway. The vast majority of Australians live in one of the major cities (off the top of my head: Sidney, Melbourne, Brisbane, Perth and Canbera). Admittedly, for a city of only 3 million people, Melbourne occupies a huge amount of space, but the chances of not being within 10 miles of a local exchange are fairly slim. Of course, the rural areas are always going to lose out...
Didn't you read the article? He's from Sheffield. They're barely human up there, let alone Englishmen! As the saying goes, it's grim up North! :-) BTW, is Alan Cox not eccentric enough for you?
It won't work. Petrol (gas to those of you that choose to so describe the liquid with which you fuel your car) costs the equivalent of nearly US$6/gallon here in the UK, and the roads are still full of them.
Moving from SCO to DG/UX isn't trivial. Unless you're already running SCO on a DG AViiON, you'll need to buy new hardware. Athough DG/UX can be persuaded to run on a standard PC, they won't sell it to you unless it's to run on one of their own machines. I agree that the DG/UX kernel is very nice, I just wish they'd ship a more complete userland. BTW, anyone know how to trace system calls on DG/UX (there's no truss, strace, ktrace etc.)?
NetBSD is probably the closest I'll get to running a decent Unix on my NeXT. However, although it has support for 68040 based machines, it explicitly doesn't support my Turbo Colorstations :-( The Linux port to NeXT hardware is still a long way from usability. Guess I'm stuck with NeXTStep/OpenStep for now.
Yep, I'm afraid I have to agree here. I've been using M14 as my primary browser recently, and although it's far, far better than the previous versions I'd looked at (M10/M11), it's still not at the point where I'd say it was ready for public release, even as a beta. It still crashes occasionally, without warning, and comletely unreproducably (this is on Win32, admittedly :-) Still, it's showing a lot of promise. I just hope AOL/NSCP know what they're letting themselves in for with a beta release of this quality. The MS-sympathetic parts of the press are going to give them a really hard time unless things rapidly improve over the next month.
Yep, that sounds like a really nice feature. All those packages can provide "vim", on which other packages can depend. That way, it doesn't matter which vim you've got installed, so long as at least one of them is... which is why RPM provides exactly the same functionality (you'll need a version newer than 3.0.3, but that's nearly a year old now -- almost prehistoric in Linux terms :-)
You're thinking too much like a home user. Corporate users don't want to have a software development kit on every machine. Also, if it needs compiling, it's going to be too slow for any moderately large app.
Of course the dependency checks of either blow the pants off RPMS. Care to give some examples? I've yet to see any real world examples where .deb dependencies are better than RPM. Yes, .deb can have more complex boolean expressions in dependency checking, but in real life, RPMs dependencies are sufficient to achieve everything that's needed.
This isn't quite true. The idea is to have a standard LSB packaging format, which can be used for distribution-neutral packages. Each distribution will be responsible for providing suitable tools for converting LSB packages to their native package format, and handling their subsequent installation. It is true that the current plan is to use RPM as the basis for LSB packages, simply because it's so widely supported already. As the spec says: If you don't like this, then please propose an alternative.
No, they won't have a problem with it -- XFree86 has been officially released (albeit with limited supported cards, which is presumably why SuSE 6.4 doesn't use it by default).
Why is this a problem with the others? I'm not aware of any current distributions that ship GNOME but not KDE. The only difference is the GNOME is the default in Red Hat, while KDE is the default in SuSE.
SuSE may not be just for Europeans any more.
People in the US seem to assume that everyone in Europe uses SuSE. While it may have a stranglehold on the German market, that's certainly not the case in the UK. Yes, it's available, but Red Hat seems to be more prevalent.
The risk is both damage to the physical hardware and data corruption. The hardware can easily be replaced when it's a cheap Celeron, but not when it's a dual core IBM Power CPU. The data corruption can't be ignored, though. Don't believe me? Maybe you'd like to hear it directly from someone you might trust.
Yes, you can, if you're prepared to take the risk -- that's the whole basis of overclocking. Chips are rated at the speed the manufacturer can guarantee they'll operate as intended. Say you overclock your chip by 15%. You're now encroaching into the safety margin that the engineers and the manufacturer allowed to be sure that all chips will work correctly. Even so, perhaps 98% of all chips will be OK. Do you want to gamble on whether or not you've got one of the 1 in 50 chips that won't work? Personally, I don't like the odds, particularly when the chips cost as much as this one will...
Maybe they do, maybe they don't. You're missing the point though. If you want faster speeds, go buy faster processors (or more of them). Overclocking is only for those who can't afford to do that. People buying these chips aren't going to fall into that category.
The other point to consider is that overclocking an SMP system is tantamount to suicide, by all accounts. Now maybe that won't be the case here, because the cores are on the same die, and hence will be affected in exactly the same way, but I don't know enough about it to be sure, and I certainly wouldn't risk it.
Just like the Frauhofer patents have prevented the distribution of MP3 encoding software? OK, so I know the situation is slightly different for MP3, but it's incredibly difficult to prevent the spread of software whose time has come, legal or not...
By free, I assume you mean open. The problem with open codecs is (at least in the eyes of content providers) copy protection. They see an inability to save streamed RealAudio/ReadVideo, and are happy because their copyrighted works aren't being distributed around the net. An open codec would allow anyone to save the streamed data to a file for later use. In actual fact, the current closed codecs only provide security through obscurity anyway, and will eventually be compromised.
Firstly, my ISP only supports POP3 mail access. They're moving to IMAP, but it's not there yet. Secondly, assuming that I have access to my mail via IMAP, what software should I be using? fetchmail is working fine for me at the moment, and will continue to do so, but I don't know how I could persuade it to downlaod headers only and then selectively choose which messages to junk and which to download. So what else is there?
That's nice for you. However, before you can get to that stage, you have to download the message. Those of us in the UK (and, in fact, most of Europe) don't have free local calls. The upshot is, that to download a 2.5MB spam (yes, I've been spammed with PDF product catalogues this size before) over my modem costs me the equivalent of about US$2.00. Now if someone shoved some junk mail through your door with a bill for $2.00 that you had no choice but to pay, would you complain? No? Then please send me your address, I've got a once in a lifetime opportunity for you...