And you can do something else really neat - watch...
Take a drive out of a Mac and put it into a PC - Check.
I can do that with my Linux machines - a drive out of a Mac machine can easily be read, formatted, and/or backed up (dd the data off the disk or whatever), with no problem. That Linux thing sure is compatible, huh?
What's really funny is explaining this to a Mac user, who insists that Mac hard drives are DIFFERENT from ALL OTHER HARD DRIVES, and that CAN'T POSSIBLY WORK...
Yep, the stability that is possible when you're writing software for a single known hardware spec can be quite annoying. (note my facetiousness)
When someone can do that successfully, let me know. Apple's famous for not really knowing their own hardware - they often don't have internal specs/documentation for their own hardware, so no one (other than the engineer(s) who designed it) knows how it really works. Sure, they have a smaller set of hardware to support (and the hardware pool shrinks a bit more with each new release of OS X - 10.2 killed everything pre-G3, 10.3 killed everything pre-NewWorld). But when you don't know your own hardware, you still have problems.
I am not in the 90%. I have my home desktop, a PowerBook Pismo (purchased by my employer), a PowerMac 7500 (acquired from a previous coworker when I mentioned I wanted to play around on an OldWorld PPC), and my dual-head work desktop. I run Linux on them all. I wouldn't have it any other way.
I've dealt with MacOS X - sure, you have your shell prompt, but it just doesn't feel like Un*x - to me, it's closer to using Cygwin on Windows than actually running Un*x. It still feels too constraining to me. The Mac has always seemed overly constraining - as someone else pointed out, there's one way to do each thing, or a very small group of choices. You do it the way the machine expects. I've gotten used to the Un*x way - you tweak and customize, and make the machine do what you want it to do any way you want.
I'm not saying MacOS X is bad - if you like it and are happy with it, good for you. However, claiming it's the one universal one-size-fits-all do-everything perfect platform is just wrong. And I've had to wrestle with it enough that I know it's far from perfection - Apple still needs to finish certain parts of the OS (no, it's not "finished" just because it has a GUI), and they need to document the hell out of how it _actually_ works, not how they think it should work (their current documentation, especially for MacOS X Server, is abominably inaccurate and generally worthless for real use).
I will agree that as far as commercial alternatives, it's a lot better than the majority of what's out there, however...
Similar story here, with a Pismo 400MHz, running Debian. This thing is a rock. Sleep is so much more reliable than on any x86-based system, and it doesn't drag ass the way OS X would on this hardware./me waits to get flamed, and for the next OS X release to not support G3s anymore (just wait! they killed the oldworlds!)
Yes, and it also has terrible color-endianness problems. I tried using it with Mac-on-Linux from a PowerMac 7500 running Debian, and it butchered the colors, whether using PseudoColor or TrueColor. I'd go for XFree86's port to Windows - it actually works right, and yes Virginia, it does rootless too. (See some other comments.)
Yeah, I found that rather humorous as well. When I first heard SCO talking about how Linux distribution vendors and solution providers should be "indemnifying" their customers, I figured it was only a matter of time before someone would do it and SCO would throw it back in their face and say "Told you! They must have done what we said, why else would they indemnify their customers?" To me, that just proves yet again what a sham SCO's case is - full of cheap tactics that are supposed to make the people they're suing look bad, but just serve to make them look like bigger morons.
I don't even know wtf this guy was on about with the "floppy.o" thing anyway. I have a desktop with no floppy drives, the floppy controller disabled - and it's been this way for years. I use GRUB for booting, and I've never had a problem with that. Can someone explain it to me?
I'm pretty sure that Data General was NOT bought by Digital. Far as I know, Data General is still around, or was not too long ago, and they were making large multi-processor Windows machines after they stopped making multi-processor 68060 systems to run DG/UX on. They didn't get eaten, just abandoned UNIX-based systems as a core market. It's obviously not done well for them, considering they're little more than a footnote anymore.
First, as others will point out, the plants and animals used for food today have been genetically manipulated for years - if not by direct manipulation of their genes, by breeding in desirable traits, and breeding out undesirable ones. Also, supposedly all these "glo-fish" are made male. It's awfully hard for them to breed and make more glowing fishies. Also, THEY ARE TROPICAL FISH. If you release them into your local lake or stream, the odds of them surviving for long are slim to none - because they're simply not suited to that environment at all.
What is up with people and this hysteria? C'mon kids, let's try to show some reason.
Hah. Gee, what a shock. Paul Thurrott whores himself out to Microsoft again. I'm SHOCKED, SHOCKED I say. He's only done it a few times before... not so surprising that he should do it yet again. He's just a pro-MS troll who happens to get paid for it.
And what of the fact that practically every LS-120 drive self-destructed a couple years ago? The place I was working then suddenly had them dying almost by the truckload, and the LS-120 internal drive I had in my home system stopped reading the LS-120 disks, then after awhile, just stopped working altogether. (The place I worked, a public school district, had bought them for their NewWorld Mac systems, because they had so much material on floppies, and needed to access it.)
I'm talking about 10.2 (Jaguar) prerelease builds, but I know 10.0 (and possibly some early 10.1 releases) dealt poorly with SCSI. Apple had kind of given up on SCSI for awhile, which is kind of funny, considering it was their mainstay for a long time - not to mention the fact that SBP-2 is just SCSI-3 protocol over a serial bus, so they have to implement some SCSI support code anyway.
Yes, I understand that prerelease software may have bugs, but the fact that it would chew up partitions on SCSI disks to the point where the automount bit got cleared, and then both OS 9 and OS X would just not see them anymore - that's getting a little ridiculous.
No, it was around 1.0 MB, actually. I don't recall the exact reasoning for that limit, but I know that I could fit GRUB and an over-1 MB kernel on a floppy, but couldn't boot from the same kernel dd'd onto a floppy, and that was the reason why.
Several early versions (and several prereleases) of OS X had really broken SCSI support. A coworker at my previous job was always trying the latest OS X version, and had an Adaptec SCSI controller with a SCSI drive hanging off it. She was always having problems with OS X chewing up and spitting out partitions on the SCSI disk.
I'm running 2.6.0 on Debian sid right now. It looks like the necessary additional package (module-init-tools) isn't available on woody (maybe as a backport, but...), so you'll probably have to go with at least sarge (testing) to do it. Works well for me though - running patched versions of the nVidia and VMware kernel modules, and everything seems good so far.
Um, just 'make oldconfig && make' is all you need to do once you copy your.config - if the kernel is too big for zImage (which kernels anymore generally are), the build procedure will generate the bzImage automatically.
I know you could do it with 2.4 bzImage kernels, as long as they were below a certain size limit. If they were too big, however (I don't recall the exact size limit), it just wouldn't work.
The true artist loves his tools, and so then do his tools love him back. Saying that "oh, I don't care, whatever tools they give me I'll use" is either (a) a lie, or (b) means you really don't care about the output. My dad, an auto mechanic for many years, prefers certain tools over others - he knows his tools, and chooses them carefully. Not just "right tool for the job", but "who made the tool?" It's the same thing with a development environment. I can't imagine trying to write code on Windows anymore - it's Linux and gvim for me, thanks. It takes time to learn the ins and outs, sure, but once you learn, you're that much more effective at what you do.
Complete system? Yeah, I suppose... for looking pretty. Or did you mean for getting work done? Because you definitely won't be getting any serious work done on a Windows box without installing (hence, usually _buying_) a whole lot of extra software. That's not my idea of a "complete" system.
You're deluding yourself if you really think the PATRIOT Act's expanded powers won't be (and haven't been) used outside the scope of terrorism. There have already been news stories about the use of those broad, sweeping powers in cases that have nothing whatsoever to do with "terrorism". Read the news, man. These limit _everyone's_ freedoms. Don't give me the "oh, they'll REALLY only use it against X" - the government and law enforcement will use it whenever they damn well please. They're just following our "great leader's" *cough* *cough* example, after all...
Here's a hint for you. The grandparent post, as well as many other posts, are simply pointing out (and rightly so, IMO) that the ends don't justify the means. I mean, yeah, Saddam Hussein was not a good guy. I think we can all agree on that point (ok, most of us, maybe not all). However, I think that it's not really a good precedent to set that "oh, we don't like what these guys are doing, let's give some dubious reason and start a war with them". That just can't be allowed. It's not cool, and that kind of flimsy reasoning won't cut it. While yes, Saddam Hussein's capture is a net positive, that doesn't mean we should say "whatever it takes, immoral or not, to achieve our desired result". Please, just think about that the next time you are going to tell someone not to speak up about methods they think are dubious, or just plain wrong. The ends do not, can not, and should not be used to justify the means, no matter how good the outcome.
It's probably the FrontPage server extensions, which provide a way to upload content to a web site without using WebDAV or FTP. It's been known as a security hole for a long, long time now. If you value security, you know better than to enable it in the first place.
Once upon a time, way back when, I'd have agreed with you. I'm still a happy Linux user (running Debian on a PowerBook, no less), but I know a lot of the/. readership anymore is comprised of people who've probably never used anything other than Windows. Too bad, really.
Re:the most annoying thing
on
PC Annoyances
·
· Score: 1
Yes, but so many users spend so much time on changing the wallpaper, colors, fonts, sounds, mouse cursors, and on and on, that (a) they don't get anything else done, and (b) no one else can make heads or tails of their desktop. Some people would say "so? this means they can customize to their hearts' content!" But considering how Windows' big benefit, as a "universal platform" is anybody can supposedly sit down and use any Windows machine... that kinda throws that out the window, hm?
Um, IMAP != MAPI. MAPI == Microsoft's Messaging API, which allows integration into the Windows desktop for sending mail. Personally, I don't run Windows at all (except for certain things for work, but then I run Win2K inside VMware).
Slashdot Mirror
And you can do something else really neat - watch...
Take a drive out of a Mac and put it into a PC - Check.
I can do that with my Linux machines - a drive out of a Mac machine can easily be read, formatted, and/or backed up (dd the data off the disk or whatever), with no problem. That Linux thing sure is compatible, huh?
What's really funny is explaining this to a Mac user, who insists that Mac hard drives are DIFFERENT from ALL OTHER HARD DRIVES, and that CAN'T POSSIBLY WORK...
Yep, the stability that is possible when you're writing software for a single known hardware spec can be quite annoying. (note my facetiousness)
When someone can do that successfully, let me know. Apple's famous for not really knowing their own hardware - they often don't have internal specs/documentation for their own hardware, so no one (other than the engineer(s) who designed it) knows how it really works. Sure, they have a smaller set of hardware to support (and the hardware pool shrinks a bit more with each new release of OS X - 10.2 killed everything pre-G3, 10.3 killed everything pre-NewWorld). But when you don't know your own hardware, you still have problems.
I am not in the 90%. I have my home desktop, a PowerBook Pismo (purchased by my employer), a PowerMac 7500 (acquired from a previous coworker when I mentioned I wanted to play around on an OldWorld PPC), and my dual-head work desktop. I run Linux on them all. I wouldn't have it any other way.
I've dealt with MacOS X - sure, you have your shell prompt, but it just doesn't feel like Un*x - to me, it's closer to using Cygwin on Windows than actually running Un*x. It still feels too constraining to me. The Mac has always seemed overly constraining - as someone else pointed out, there's one way to do each thing, or a very small group of choices. You do it the way the machine expects. I've gotten used to the Un*x way - you tweak and customize, and make the machine do what you want it to do any way you want.
I'm not saying MacOS X is bad - if you like it and are happy with it, good for you. However, claiming it's the one universal one-size-fits-all do-everything perfect platform is just wrong. And I've had to wrestle with it enough that I know it's far from perfection - Apple still needs to finish certain parts of the OS (no, it's not "finished" just because it has a GUI), and they need to document the hell out of how it _actually_ works, not how they think it should work (their current documentation, especially for MacOS X Server, is abominably inaccurate and generally worthless for real use).
I will agree that as far as commercial alternatives, it's a lot better than the majority of what's out there, however...
Similar story here, with a Pismo 400MHz, running Debian. This thing is a rock. Sleep is so much more reliable than on any x86-based system, and it doesn't drag ass the way OS X would on this hardware. /me waits to get flamed, and for the next OS X release to not support G3s anymore (just wait! they killed the oldworlds!)
Yes, and it also has terrible color-endianness problems. I tried using it with Mac-on-Linux from a PowerMac 7500 running Debian, and it butchered the colors, whether using PseudoColor or TrueColor. I'd go for XFree86's port to Windows - it actually works right, and yes Virginia, it does rootless too. (See some other comments.)
Yeah, I found that rather humorous as well. When I first heard SCO talking about how Linux distribution vendors and solution providers should be "indemnifying" their customers, I figured it was only a matter of time before someone would do it and SCO would throw it back in their face and say "Told you! They must have done what we said, why else would they indemnify their customers?" To me, that just proves yet again what a sham SCO's case is - full of cheap tactics that are supposed to make the people they're suing look bad, but just serve to make them look like bigger morons.
I don't even know wtf this guy was on about with the "floppy.o" thing anyway. I have a desktop with no floppy drives, the floppy controller disabled - and it's been this way for years. I use GRUB for booting, and I've never had a problem with that. Can someone explain it to me?
I'm pretty sure that Data General was NOT bought by Digital. Far as I know, Data General is still around, or was not too long ago, and they were making large multi-processor Windows machines after they stopped making multi-processor 68060 systems to run DG/UX on. They didn't get eaten, just abandoned UNIX-based systems as a core market. It's obviously not done well for them, considering they're little more than a footnote anymore.
First, as others will point out, the plants and animals used for food today have been genetically manipulated for years - if not by direct manipulation of their genes, by breeding in desirable traits, and breeding out undesirable ones. Also, supposedly all these "glo-fish" are made male. It's awfully hard for them to breed and make more glowing fishies. Also, THEY ARE TROPICAL FISH. If you release them into your local lake or stream, the odds of them surviving for long are slim to none - because they're simply not suited to that environment at all.
What is up with people and this hysteria? C'mon kids, let's try to show some reason.
Hah. Gee, what a shock. Paul Thurrott whores himself out to Microsoft again. I'm SHOCKED, SHOCKED I say. He's only done it a few times before... not so surprising that he should do it yet again. He's just a pro-MS troll who happens to get paid for it.
And what of the fact that practically every LS-120 drive self-destructed a couple years ago? The place I was working then suddenly had them dying almost by the truckload, and the LS-120 internal drive I had in my home system stopped reading the LS-120 disks, then after awhile, just stopped working altogether. (The place I worked, a public school district, had bought them for their NewWorld Mac systems, because they had so much material on floppies, and needed to access it.)
I'm talking about 10.2 (Jaguar) prerelease builds, but I know 10.0 (and possibly some early 10.1 releases) dealt poorly with SCSI. Apple had kind of given up on SCSI for awhile, which is kind of funny, considering it was their mainstay for a long time - not to mention the fact that SBP-2 is just SCSI-3 protocol over a serial bus, so they have to implement some SCSI support code anyway.
Yes, I understand that prerelease software may have bugs, but the fact that it would chew up partitions on SCSI disks to the point where the automount bit got cleared, and then both OS 9 and OS X would just not see them anymore - that's getting a little ridiculous.
No, it was around 1.0 MB, actually. I don't recall the exact reasoning for that limit, but I know that I could fit GRUB and an over-1 MB kernel on a floppy, but couldn't boot from the same kernel dd'd onto a floppy, and that was the reason why.
Several early versions (and several prereleases) of OS X had really broken SCSI support. A coworker at my previous job was always trying the latest OS X version, and had an Adaptec SCSI controller with a SCSI drive hanging off it. She was always having problems with OS X chewing up and spitting out partitions on the SCSI disk.
I'm running 2.6.0 on Debian sid right now. It looks like the necessary additional package (module-init-tools) isn't available on woody (maybe as a backport, but...), so you'll probably have to go with at least sarge (testing) to do it. Works well for me though - running patched versions of the nVidia and VMware kernel modules, and everything seems good so far.
Um, just 'make oldconfig && make' is all you need to do once you copy your .config - if the kernel is too big for zImage (which kernels anymore generally are), the build procedure will generate the bzImage automatically.
I know you could do it with 2.4 bzImage kernels, as long as they were below a certain size limit. If they were too big, however (I don't recall the exact size limit), it just wouldn't work.
The true artist loves his tools, and so then do his tools love him back. Saying that "oh, I don't care, whatever tools they give me I'll use" is either (a) a lie, or (b) means you really don't care about the output. My dad, an auto mechanic for many years, prefers certain tools over others - he knows his tools, and chooses them carefully. Not just "right tool for the job", but "who made the tool?" It's the same thing with a development environment. I can't imagine trying to write code on Windows anymore - it's Linux and gvim for me, thanks. It takes time to learn the ins and outs, sure, but once you learn, you're that much more effective at what you do.
Complete system? Yeah, I suppose... for looking pretty. Or did you mean for getting work done? Because you definitely won't be getting any serious work done on a Windows box without installing (hence, usually _buying_) a whole lot of extra software. That's not my idea of a "complete" system.
You're deluding yourself if you really think the PATRIOT Act's expanded powers won't be (and haven't been) used outside the scope of terrorism. There have already been news stories about the use of those broad, sweeping powers in cases that have nothing whatsoever to do with "terrorism". Read the news, man. These limit _everyone's_ freedoms. Don't give me the "oh, they'll REALLY only use it against X" - the government and law enforcement will use it whenever they damn well please. They're just following our "great leader's" *cough* *cough* example, after all...
Here's a hint for you. The grandparent post, as well as many other posts, are simply pointing out (and rightly so, IMO) that the ends don't justify the means. I mean, yeah, Saddam Hussein was not a good guy. I think we can all agree on that point (ok, most of us, maybe not all). However, I think that it's not really a good precedent to set that "oh, we don't like what these guys are doing, let's give some dubious reason and start a war with them". That just can't be allowed. It's not cool, and that kind of flimsy reasoning won't cut it. While yes, Saddam Hussein's capture is a net positive, that doesn't mean we should say "whatever it takes, immoral or not, to achieve our desired result". Please, just think about that the next time you are going to tell someone not to speak up about methods they think are dubious, or just plain wrong. The ends do not, can not, and should not be used to justify the means, no matter how good the outcome.
It's probably the FrontPage server extensions, which provide a way to upload content to a web site without using WebDAV or FTP. It's been known as a security hole for a long, long time now. If you value security, you know better than to enable it in the first place.
Once upon a time, way back when, I'd have agreed with you. I'm still a happy Linux user (running Debian on a PowerBook, no less), but I know a lot of the /. readership anymore is comprised of people who've probably never used anything other than Windows. Too bad, really.
Yes, but so many users spend so much time on changing the wallpaper, colors, fonts, sounds, mouse cursors, and on and on, that (a) they don't get anything else done, and (b) no one else can make heads or tails of their desktop. Some people would say "so? this means they can customize to their hearts' content!" But considering how Windows' big benefit, as a "universal platform" is anybody can supposedly sit down and use any Windows machine... that kinda throws that out the window, hm?
Um, IMAP != MAPI. MAPI == Microsoft's Messaging API, which allows integration into the Windows desktop for sending mail. Personally, I don't run Windows at all (except for certain things for work, but then I run Win2K inside VMware).