Reasons for using it over any other major linux distro:
# Choose from among XFS, ext2, ext3, reiserfs, and others during install
Nope, Mandrake's the same.
# choose to encrypt your filesystem
Mmm, same.
# free security updates, unlike RedHat
Wrong.
# improved YaST2, the ultamite in system configuration utilities, let's you configure everything from a DHCP server to CUPS
Just what we need, and also covered (badly) by webmin/linuxconf on other distros.
# YaST Online Update, for automatic upgrading of your RPMs
Same as mandrake.
I've yet to see anything major to swing me. I'm far too used to the old tools (chkconfig, ifconfig) and the like that it doesn't make much difference which distro I use, linux or otherwise. That's one of the main reasons I'm against pushing things like YaST2 too hard, or people will simply turn linux into open source M$.
It's amazing how fast the numbers go up in distributions compared to that of the packages within...
Maybe companies should be forced to number their distros based on the average version number of the packages within; that'd cause an interesting non-linear numbering scheme.;)
As far as I am aware, the anti-dumping legislation in the US is far more significant than elsewhere. I believe in the US though, anti-dumping laws only affected foreign companies. Yet another interesting US law that seeks to protect US business through block out clauses. Seems like the free market americans can't cope with their foreign counterparts...
You've modded this post up when he says Yast is far easier than mandrake's equivalent (which he hasn't tried). Please. urpmi has proved to be a good apt-get equivalent, which is easy to configure and doesn't fall over. "SuSE is easier to get going in 3D gaming"? Where does this hold any water? All my kit has been correctly detected and configure first time by mandrake, and I'm confused as to how that can improve.
This one's a semi myth. They're not quite the same thing. MRI is a use of NMR. NMR is the principle you're using, and MRI is an application of the principle. It can be called NMRI, and that *is* equivalent to MRI.
As a footnote, I note that my local hospital (Leeds General Infirmary) calls it NMRI.
That assumes that RTF really is standardised properly. I've had people refused applix generated RTF as being not proper RTF.
I think we have to wait for those *lazy* application developers to write apps that actually produce the correct output. I suspect in this case it wasn't applix that was at fault, but the M$ product at the other end trying to read it.
While I was working at Leeds University we had similar programs, but it wasn't exact matches we were searching for. It's far harder to detect plagiarism than that. You have to tokenize the work to get rid of variable names (although you can take that into account later) get rid of all the formatting, try shuffling it all round a bit, and then seeing if it matches.
Most of the people we caught cheating had at least made a cheap attempt to hide their wrongdoing.
Look MHz and Voltage do not directly relate to power usage. You seem to have completely lost the plot here.
I want a fast CPU because occasionally I load the machine lots, and when I do, I want it to do things fast.
I might also want to render a horribly complicated OpenGL scene. Don't assume that everyone is doing games the whole time, nor that all games now run at 200fps. It's total crap. People want an 1000000THz processor that uses power management so that it only works 0.00001% of the time on average. That'd be perfect for me.
SCSI performance analysis
on
2.2 vs 2.4
·
· Score: 1
Pretty poorly written article that comments on the improvements in SCSI are shown in the results, then gives you a graph showing a performace decrease...
I'm not convinced in the slightest by the argument that things run in a more stable manner on X distribution. Yes some distros might ship with a broken compiler whatever, but once that's fixed, I've not found any difference whatsoever between distros in this respect.
Compiling your own packages I would agree tends to increase your chances of having a nice stable system...
>> I thought that was part of the reason behind a server side scripting language.
> Nope. Client-side stuff is for things that happen on the client-side (like menus and > stuff), and server-side is for getting data from the server and doing things there.
> There's really no overlap, and with today's modern websites that demand snazzy effects, > there's really no money to support crap like Netscape.
Mmm nice. Server side scripting does have the advantage of not relying on features at the browser end to some extend, so the guy had a point. But if it produces non-standard HTML as its output, then it's not going to work unless the browser supports those 'extensions'. Write good scripts that make no unreasonable assumptions about the other end's browser, and all is dandy.
Just put your anti-netscape opinion back in the closet, and start talking some sense. Snazzy effects can be acheived with javascript/flash, and if the browser doesn't support it, then you can have a less flashy site to fall back on. Then everyone's happy.
> you should use Explorer (unless you're a real > masochist) over Netscape.
Who said I had to use either? I use a variety of browsers, including text based (links). Users of Opera/Mozilla/Galeon/lynx are all gonna be pretty pissed if the web goes this way. The Internet, since it is based on open standards (in theory) shouldn't tie you to a particular vendor. The best browser should support the standards 100%. It shouldn't simply tack on bits that make it no longer a web browser. It starts to become a MS-Net-Info browser...
Surely that's something that all too many people are sloppy about. The only commercial site I've written, I made the effort to be pedantic, and pushed all the pages through the w3c validator service. If anything was non-standard then it didn't go in.
I'm against using Javascript for essential sections of the page. Anything that is even borderline non-standard, has to merely make the page look in some way nicer, but no detract from the functionality. To make a page simply reject the browser fullstop goes against the cross-platform model that the Internet and its users should subscribe to.
What percentage of pages actually meet the DTD? It'd be a pretty sad figure if you ask me. Admittedly it would help if browsers properly supported the standards...
jh
But we all know that you can disable macrovision on a G400 thanks to the wonderful guys at matrox who 'accidentally' released a set of drivers where macrovision could be disabled by a a roundabout route.
Which was nice.
Linux is not a difficult operating system at all. It is virtually impossible to use at all without being educated properly in how to use it. The university I work at (as a TA) is now adding additional material to their course to improve linux skills, with the belief that a solid unix grounding at the start will actually save students' time later. I would have no interest in teaching C with VC++.
What do you say to someone when their Roaming Profile goes mad with NT swallowing all their work, or it GPFs? I get less grief on the whole with linux.
In fact when it came to teaching my computer illiterate sister programming (as part of a career change) I tried that route and it failed horribly. Falling back on Linux/vi/g++ had remarkably few problems.
Anyone that attempts to learn principles from a pathetically loosely designed language like VB should go home. I'm all in favour of starting with C and moving slowly up. We've entirely stopped teaching arrays and pointers now in the belief (I assume) that they are no longer necessary to teach. It doesn't fill me with delight.
The university involved here is a good old traditional English university BTW.
Slashdot Mirror
Not something to get too excited about. I'm part of a team working with Grid Engine on Globus, and it's not a pretty sight.
jh
Reasons for using it over any other major linux distro:
# Choose from among XFS, ext2, ext3, reiserfs, and others during install
Nope, Mandrake's the same.
# choose to encrypt your filesystem
Mmm, same.
# free security updates, unlike RedHat
Wrong.
# improved YaST2, the ultamite in system configuration utilities, let's you configure everything from a DHCP server to CUPS
Just what we need, and also covered (badly) by webmin/linuxconf on other distros.
# YaST Online Update, for automatic upgrading of your RPMs
Same as mandrake.
I've yet to see anything major to swing me. I'm far too used to the old tools (chkconfig, ifconfig) and the like that it doesn't make much difference which distro I use, linux or otherwise. That's one of the main reasons I'm against pushing things like YaST2 too hard, or people will simply turn linux into open source M$.
jh
It's amazing how fast the numbers go up in distributions compared to that of the packages within...
;)
Maybe companies should be forced to number their distros based on the average version number of the packages within; that'd cause an interesting non-linear numbering scheme.
jh
As far as I am aware, the anti-dumping legislation in the US is far more significant than elsewhere. I believe in the US though, anti-dumping laws only affected foreign companies. Yet another interesting US law that seeks to protect US business through block out clauses. Seems like the free market americans can't cope with their foreign counterparts...
jh
You've modded this post up when he says Yast is far easier than mandrake's equivalent (which he hasn't tried). Please. urpmi has proved to be a good apt-get equivalent, which is easy to configure and doesn't fall over. "SuSE is easier to get going in 3D gaming"? Where does this hold any water? All my kit has been correctly detected and configure first time by mandrake, and I'm confused as to how that can improve.
I just can't believe modding sometimes...
This one's a semi myth. They're not quite the same thing. MRI is a use of NMR. NMR is the principle you're using, and MRI is an application of the principle. It can be called NMRI, and that *is* equivalent to MRI.
As a footnote, I note that my local hospital (Leeds General Infirmary) calls it NMRI.
jh
That assumes that RTF really is standardised properly. I've had people refused applix generated RTF as being not proper RTF.
I think we have to wait for those *lazy* application developers to write apps that actually produce the correct output. I suspect in this case it wasn't applix that was at fault, but the M$ product at the other end trying to read it.
jh
While I was working at Leeds University we had similar programs, but it wasn't exact matches we were searching for. It's far harder to detect plagiarism than that. You have to tokenize the work to get rid of variable names (although you can take that into account later) get rid of all the formatting, try shuffling it all round a bit, and then seeing if it matches.
Most of the people we caught cheating had at least made a cheap attempt to hide their wrongdoing.
How does this make a story?
jh
Look MHz and Voltage do not directly relate to power usage. You seem to have completely lost the plot here. I want a fast CPU because occasionally I load the machine lots, and when I do, I want it to do things fast. I might also want to render a horribly complicated OpenGL scene. Don't assume that everyone is doing games the whole time, nor that all games now run at 200fps. It's total crap. People want an 1000000THz processor that uses power management so that it only works 0.00001% of the time on average. That'd be perfect for me.
Pretty poorly written article that comments on the improvements in SCSI are shown in the results, then gives you a graph showing a performace decrease...
>...so I can take advantage of any compile time optimazation etc...
Another person that's missed the point of source rpms?
I'm not convinced in the slightest by the argument that things run in a more stable manner on X distribution. Yes some distros might ship with a broken compiler whatever, but once that's fixed, I've not found any difference whatsoever between distros in this respect.
Compiling your own packages I would agree tends to increase your chances of having a nice stable system...
jh
jh
> Nope. Client-side stuff is for things that happen on the client-side (like menus and
> stuff), and server-side is for getting data from the server and doing things there.
> There's really no overlap, and with today's modern websites that demand snazzy effects,
> there's really no money to support crap like Netscape.
Mmm nice. Server side scripting does have the advantage of not relying on features at the browser end to some extend, so the guy had a point. But if it produces non-standard HTML as its output, then it's not going to work unless the browser supports those 'extensions'. Write good scripts that make no unreasonable assumptions about the other end's browser, and all is dandy.
Just put your anti-netscape opinion back in the closet, and start talking some sense. Snazzy effects can be acheived with javascript/flash, and if the browser doesn't support it, then you can have a less flashy site to fall back on. Then everyone's happy.
jh
> masochist) over Netscape.
Who said I had to use either? I use a variety of browsers, including text based (links). Users of Opera/Mozilla/Galeon/lynx are all gonna be pretty pissed if the web goes this way. The Internet, since it is based on open standards (in theory) shouldn't tie you to a particular vendor. The best browser should support the standards 100%. It shouldn't simply tack on bits that make it no longer a web browser. It starts to become a MS-Net-Info browser...
jh
Surely that's something that all too many people are sloppy about. The only commercial site I've written, I made the effort to be pedantic, and pushed all the pages through the w3c validator service. If anything was non-standard then it didn't go in.
I'm against using Javascript for essential sections of the page. Anything that is even borderline non-standard, has to merely make the page look in some way nicer, but no detract from the functionality. To make a page simply reject the browser fullstop goes against the cross-platform model that the Internet and its users should subscribe to.
What percentage of pages actually meet the DTD? It'd be a pretty sad figure if you ask me. Admittedly it would help if browsers properly supported the standards... jh
But we all know that you can disable macrovision on a G400 thanks to the wonderful guys at matrox who 'accidentally' released a set of drivers where macrovision could be disabled by a a roundabout route. Which was nice.
Linux is not a difficult operating system at all. It is virtually impossible to use at all without being educated properly in how to use it. The university I work at (as a TA) is now adding additional material to their course to improve linux skills, with the belief that a solid unix grounding at the start will actually save students' time later. I would have no interest in teaching C with VC++. What do you say to someone when their Roaming Profile goes mad with NT swallowing all their work, or it GPFs? I get less grief on the whole with linux. In fact when it came to teaching my computer illiterate sister programming (as part of a career change) I tried that route and it failed horribly. Falling back on Linux/vi/g++ had remarkably few problems. Anyone that attempts to learn principles from a pathetically loosely designed language like VB should go home. I'm all in favour of starting with C and moving slowly up. We've entirely stopped teaching arrays and pointers now in the belief (I assume) that they are no longer necessary to teach. It doesn't fill me with delight. The university involved here is a good old traditional English university BTW.