No Microsoft Patches This Month After All

PCWorld/ IDG News Service ^ | 9/9/2005 | Robert McMillan

[js1138]

Well. last night I updated my WIndows XP OS and everything was great until I updated the audio driver

That can happen with any driver, which is why XP device manager has a button to roll back drivers to an earlier version.

Try it.

[js1138]

You might try getting drivers from the hardware manufacturer -- either the sound card or motherboard.

As someone pointed out, Windows supplied drivers are generic and are there for convenience.

I've only seen MS supplied drivers be bad a couple of times in the the last few years, and I maintain dozens of computers.

[Redleg Duke]

I tried it and it said that there were no previous versions!

[ShadowAce]

Look, there's NO SUCH THING as an OS or software package that doesn't need some amount of patching. I've been involved in applying patches to IBM 360 System OS, VMS and BSD 4.1 on Vaxen, OS/2, Novell NetWare, HP-UX, Ultrix, Linux, and of course Windows (starting with Lan Manager 2.0 and Windows 3.0) and a few others too!

True. As I do not have the experience with patching that you do, can you tell me how many of those OS' patches break things as often as Windows patches seem to do? I know Linux does not have the same track record as Windows as those patches actually fix the problem without introducing others.

[js1138]

I build computers out of leftover parts from upgrades and such. I'm kind of used to finding drivers on the internet, and I'm kind of used to them not working.

I think I've seen this situation, but after a while they blur together.

You should be able to get the original driver from the hardware manufacturer.

[Redleg Duke]

I bought a sound card on Ebay this weekend. That is probably the brute force solution.

[MarkL]

True. As I do not have the experience with patching that you do, can you tell me how many of those OS' patches break things as often as Windows patches seem to do? I know Linux does not have the same track record as Windows as those patches actually fix the problem without introducing others.

Here you've hit on a major difference between a Linux distribution and Windows, in their architecture. Windows is a monolithic OS, while a Linux distribution is made up of a lot of different, independant modules. Your question would be better served by asking "what is the track record of Linux kernal patches breaking things, compared to Windows." For instance, you can't compare a Windows security patch that deals with the OS, to a Bind or SendMail security patch. I have seen a number of Novell patches pulled off the Internet almost as soon as they hit it. I don't recall the same for Linux, but I would be surprised if that didn't happen. You'll see this a lot more often in the Windows world, simply because of the sheer number of Windows systems out there, as well as all the different configurations, running different software.

Mark

[Golden Eagle]

I know Linux does not have the same track record as Windows as those patches actually fix the problem without introducing others.

Actually, this is more bunk from you peddled as fact. Read the US Cert vulnerability list, AGAIN, and you'll see not only are there far more patches for Unix/Linux, there are significantly more "updated" patches for bugs.

http://www.us-cert.gov/cas/bulletins/SB05-250.html

Feel free to click the Archive button in the top right as well, and you'll see this is not only common, it is standard for Unix/Linux to have more holes, and more updates of those patches required. Why you keep trying to claim the opposite is the question?