Main "Hating Microsoft in a nutshell" thread - Page 12


View Poll Results: Do you think making Linux and MS interactable (kinda) a good idea?

Voters
11. You may not vote on this poll
  • Yes, this is a great idea

    3 27.27%
  • Yes, it's an ok idea

    1 9.09%
  • It wouldn't hurt

    3 27.27%
  • No, Linux should stick to Linux and Microsoft should stick to Microsoft

    4 36.36%
  • Or just use CrossOver Office

    0 0%
Page 12 of 76 FirstFirst ... 289101112131415162262 ... LastLast
Results 166 to 180 of 1139

Thread: Main "Hating Microsoft in a nutshell" thread

  1. #166
    Join Date
    Jan 2003
    Posts
    270
    DOS based stuff will be around a long time. Partly because of WinXP activation (spyware), customers are moving to Linux in droves. Partly because the hardware, - you need a P4 to run XP well - is still not fully expensed by companies.

    For those who like DOS, Linux is the natural next step. Where else can I burn an ISO at the command line?

  2. #167
    Join Date
    Mar 2002
    Location
    SE Michigan
    Posts
    111
    If that's a fake, he did a *very* good job.
    --
    Gentoo
    AMD64 3200+
    http://www.penguicon.org/

  3. #168
    Join Date
    Aug 1999
    Location
    Canada
    Posts
    64
    Well I've never used XP and I'm not one to go around yelling that "SuchandSuch an OS sux!!!", but let me add this story:

    My daughter went from Win2000 to WinXP on her main box at home, because she wanted remote desktop. Everything went well for about a month - no unneeded reboots and she never shuts her box. Then she *did* turn off her computer one night, and when she re-opened it the morning after, WinXP was dead...

    Okay, fine, freak accident she thinks, pissed off but what the heck, she's got a backup and so she wipes and re-installs... and again things are just peachy for a while, until the day she has to turn off her computer for maintenance.

    And guess what ?

    Yup, WinXP dead... really dead, no offers to run scandisk or anything, deleted drivers, corrupted files, the works... and this second time she didn't have a recent backup, so she was sitting in front of a useless computer, thinking about all the data she had lost **

    Needless to say she did wipe/reinstalled again, but this time went back to Win2000. As far as she's concerned, WinXP is utter crap

    ** Actually she didn't lose anything - I had just sent her a Knoppix CD for her to try, and with it she managed to mount her data drive and pull all her stuff to safety...
    -------------------------
    One reason for the bustle was that over large parts of the continent other people preferred to make money without working at all, and since the Disc had yet to develop a music recording industry they were forced to fall back on older, more traditional forms of banditry.

  4. #169
    Join Date
    Apr 2001
    Location
    SF Bay Area, CA
    Posts
    14,947
    Originally posted by nextbillgates
    And if a Windows program oversteps it's bounds, the kernel kills it as well with an error message. I'm not sure what you're getting at here
    Irrelevant. What you're talking about (Windows program overstepping its bounds) is a fault in the program, not the kernel. When that happens in Linux, you do NOT get a kernel OOPS, and you do NOT get a panic! You get a segfault in the program itself, and the program dies. Same behavior.

    Now, if there's a bug in the Linux kernel (if it's running driver code that's talking through /dev to a program, and it hits a bad memory word, which causes a segfault, for example), then the kernel will either panic or OOPS. And the program will die with the same segfault, because the kernel was executing on its behalf. If the panic wasn't while executing on the behalf of a userspace program, it'll lock up (or reboot, if you configure the kernel that way).

    What Windows does when the kernel finds bad memory by accident is, it BSOD's, writes a 64KB log file (that's readable by no one except Microsoft -- how useful is that?), and reboots. Unless you configure it not to reboot, then it just sits there until you turn it off. Regardless of whether it was executing on behalf of a userspace program or not. That is what I'm getting at. Well, that and the fact that the 2K kernel never "recovers" from a BSOD, it just reboots, same as a panic inside the kernel in Linux.

  5. #170
    Join Date
    Jun 2003
    Posts
    76
    Originally posted by chatins
    Fat32 is not journaled like ext3. This is the monkey's point. Windows file system is vastly inferior. Windows XP files systems in general are more sensitive to bad hardware, shutdown then any previous version of windows.
    FAT32 is also not recommended for use under Windows XP. Microsoft recommends NTFS and offers to upgrade your drives during the upgrade.

  6. #171
    Join Date
    Feb 2003
    Posts
    309
    That's not an ATM, it's an internet booth. Don't you see the VoIP handset?

  7. #172
    Join Date
    Jun 2002
    Location
    Montreal, Qué, Canada
    Posts
    281
    Originally posted by Nuada Storm
    FAT32 is also not recommended for use under Windows XP. Microsoft recommends NTFS and offers to upgrade your drives during the upgrade.
    Note that if the user didn't choose to convert to ntfs at installation time, it can be done, without loosing data from a command prompt: convert driveletter: /fs:ntfs

  8. #173
    Join Date
    Mar 2002
    Location
    SE Michigan
    Posts
    111
    ...and the "insert cash here" slot.
    --
    Gentoo
    AMD64 3200+
    http://www.penguicon.org/

  9. #174
    Join Date
    Sep 2000
    Location
    So. Cal
    Posts
    655
    Originally posted by bwkaz
    Irrelevant. What you're talking about (Windows program overstepping its bounds) is a fault in the program, not the kernel. When that happens in Linux, you do NOT get a kernel OOPS, and you do NOT get a panic! You get a segfault in the program itself, and the program dies. Same behavior.

    Now, if there's a bug in the Linux kernel (if it's running driver code that's talking through /dev to a program, and it hits a bad memory word, which causes a segfault, for example), then the kernel will either panic or OOPS. And the program will die with the same segfault, because the kernel was executing on its behalf. If the panic wasn't while executing on the behalf of a userspace program, it'll lock up (or reboot, if you configure the kernel that way).

    What Windows does when the kernel finds bad memory by accident is, it BSOD's, writes a 64KB log file (that's readable by no one except Microsoft -- how useful is that?), and reboots. Unless you configure it not to reboot, then it just sits there until you turn it off. Regardless of whether it was executing on behalf of a userspace program or not. That is what I'm getting at. Well, that and the fact that the 2K kernel never "recovers" from a BSOD, it just reboots, same as a panic inside the kernel in Linux.
    To each his own, I suppose. I've had kernel OOPSes on several machines for various reasons, and although the KERNEL went on, the system was no longer stable. I would rather have the computer go down for two minutes or so and return to a stable state than have it remain up in an unstable state. Thankfully, I have yet to have a Linux server OOPS on me, so I don't really consider it an issue, but then again, I have never seen one of my Windows servers BSOD in a production environment, either.

  10. #175
    Join Date
    Jun 2003
    Posts
    76

    Re: MS haters, look here...

    Originally posted by Sepero
    I was googling for a new "blue-screen-of-death" wallpaper and came across this. The BSOD on an ATM. I'm not sure, but it looks pretty real.

    http://wolfox.werewolves.org/Photojo...20BSOD_JPG.htm
    It is actually a public Internet Access Kiosk.

  11. #176
    Join Date
    Apr 2001
    Location
    SF Bay Area, CA
    Posts
    14,947
    Heh.

    At work, we had a 100% repeatable BSOD on a 2K machine. When we started up a certain program (RSLogix -- it's a program used to create ladder logic networks for PLC's that monitor the presses that they use), it would bluescreen with a "Registry access error" or something like that. Every time.

    The fix ended up being to create another account, and run RSLogix under that account. Must have been something in the profile (or more likely the ntuser.dat Registry file... bloated binary database crap... ) had gotten corrupted, or set wrong, or something.

    There are also issues with 2K and COM, when you install a crap-old version of InstallShield. IS4 (I think it was version 4, but it might have been 5, in any case, it was pretty ancient), it installs a bunch of .dll and .ocx files in its own directory. Which wouldn't be a big deal, except that these were copies of like the NT 3.51 version of the DLLs, they had the same name as other 2K DLLs (just earlier versions), and the IS4 installer insisted on running regsvr32 on them. Which hosed up the Registry keys for any COM classes that were implemented in the system version of those files (because the CLSIDs / IIDs were the same). Net result -- any VB program that you run (yes, I have to write stuff in VB at work... I know, I hate it too ) gives a "this interface is too old"-type message, and dies. Wonderful.

    When I enable IO-APIC support in my 2.4 kernels, and then try to modprobe the UHCI driver to get mouse support, the UHCI driver oopses. No idea why -- probably the motherboard's USB bus is crappy or out of spec or something, and doesn't work with IO-APIC interrupts. The kernel oopses, modprobe dies, uhci.o gets unloaded, and the system is still stable (of course, with no mouse). That's about the only OOPS that I think I've seen (well, except for the panic that happened when I formatted the root directory as reiserfs and then both neglected to include reiserfs support in the kernel and neglected to build an initrd, but that was my own fault ).

  12. #177
    Join Date
    Feb 2003
    Posts
    309

    Re: Re: MS haters, look here...

    Originally posted by Nuada Storm
    It is actually a public Internet Access Kiosk.
    ...like I said

  13. #178
    Join Date
    May 2003
    Posts
    1,241
    I wonder how long the BSOD has been there. Sometimes, in places like libraries, they won't do anything about the BSOD for days...

    ~psi42

  14. #179
    Join Date
    Sep 2000
    Location
    So. Cal
    Posts
    655
    Originally posted by bwkaz
    Heh.

    At work, we had a 100% repeatable BSOD on a 2K machine. When we started up a certain program (RSLogix -- it's a program used to create ladder logic networks for PLC's that monitor the presses that they use), it would bluescreen with a "Registry access error" or something like that. Every time.

    The fix ended up being to create another account, and run RSLogix under that account. Must have been something in the profile (or more likely the ntuser.dat Registry file... bloated binary database crap... ) had gotten corrupted, or set wrong, or something.

    There are also issues with 2K and COM, when you install a crap-old version of InstallShield. IS4 (I think it was version 4, but it might have been 5, in any case, it was pretty ancient), it installs a bunch of .dll and .ocx files in its own directory. Which wouldn't be a big deal, except that these were copies of like the NT 3.51 version of the DLLs, they had the same name as other 2K DLLs (just earlier versions), and the IS4 installer insisted on running regsvr32 on them. Which hosed up the Registry keys for any COM classes that were implemented in the system version of those files (because the CLSIDs / IIDs were the same). Net result -- any VB program that you run (yes, I have to write stuff in VB at work... I know, I hate it too ) gives a "this interface is too old"-type message, and dies. Wonderful.
    I never did like Windows' reliance on the registry. It makes it a real ***** to fix.

    When I enable IO-APIC support in my 2.4 kernels, and then try to modprobe the UHCI driver to get mouse support, the UHCI driver oopses. No idea why -- probably the motherboard's USB bus is crappy or out of spec or something, and doesn't work with IO-APIC interrupts. The kernel oopses, modprobe dies, uhci.o gets unloaded, and the system is still stable (of course, with no mouse). That's about the only OOPS that I think I've seen (well, except for the panic that happened when I formatted the root directory as reiserfs and then both neglected to include reiserfs support in the kernel and neglected to build an initrd, but that was my own fault ).
    I have had repeatable OOPs with the latest Gentoo installer and older Slackware discs that have left Linux in running, but in a state where nearly everything I do causes a segfault.

  15. #180
    Join Date
    Nov 2000
    Location
    /dev/3.14Pie/HowRandom
    Posts
    6,700
    I have seen a BSOD on the ATM at my local grocery store- kind of scary that your financial transactions depend on software as stable and secure as MS's.


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •