Windows can kiss my hairy white ass. I know this rant has seen the world over, but give me a fucking break. Not only will it not allow me to update Windows from a browser other than Internet Explorer 5+ (I tried with FireFox), but the fucking server is overloaded, and if I want the update on CD I have to pay for it.
Seems to me like Mr. Gates should ask AOL if they have any extra blank “1000 Hours Free!” CD’s. Oh, and while he’s at it, he can get a fucking clue. I got one and this is the last time I pay for an OS.
The update site is working fine, now. No system can afford to build in enough capacity to handle the largest conceivable peak load.
I do agree, however, that making it difficult to obtain updates on CDs is reprehensible. If you’ve got a nice broadband connection, multi-megabyte patches and service packs aren’t a problem, but if you’re on a shaky dial-up, it can be damn near impossible to get the stuff you need. CDs are cheap, and so is postage. Microsoft should make it easy - they can certainly afford the marginal expense.
I agree with you. Unfortunately, Microsoft’s system seems to have a single point of access, and with many machines automatically running Windows Update, there’s bound to be a bottleneck. If the patches were mirrored to servers around the world, and checksummed with say an MD5 key (so that you could be sure of their authenticity) things would be better.
What really annoyed me about Windows Update was that the really critical patches … specifically the two for the MSBlaster hole … are not prioritised to be the first updates installed.
So if you create a machine with a fresh Windows 2000 installation (as I did yesterday), you have to spend forever downloading IE and OS service packs and a million other less important things, occasionally having to reboot and restart the downloads, cos the msblaster attacks keep nobbling the machine!
My point being that search queries, while resource-intensive in the sense that the index must be searched, do not result in a physically large download. Microsoft, while not requiring much resource usage to locate the file, does need a large bandwidth and continual use of server resources to serve up the download of large files.
Comparing Google and Microsft, as you did, is inapposite.
Whilst I take your point, Bricker, it too is somewhat inapposite to the complaint at hand. The original error message, HTTP/1.1 Server Too Busy, usually reflects the number of connections to the web server equaling the maximum number of connections supported by the web server.
Indeed, Microsoft has a Knowledge Base article on this subject, with some suggestions for remedying the situation.
