Can anyone explain why ACPI would want to know what OS is running? Shouldn't it just report what it can do and let the OS decide what to do with it?

You would think so, but unfortunately ACPI is horrendously overengineered by committee and basically does have code running in it which can contain these sort of checks. I once had to recompile some part of it (possibly the DSDT - differentiated system description table - whatever that is) on an old Dell laptop of mine which literally had

if (operating_system == "Microsoft Windows NT" || operating_system == "Microsoft Windows 95" || operating_system == "Microsoft Windows") { // make stuff work }

baked into it. Linux worked fine once those checks were removed, of course.

Needless to say, my opinion of the whole mess has been pretty tainted by that. I try to assume that it is incompetence rather than actual malice on the part of Microsoft and/or the manufacturers, but it does do an awfully good impression of the latter.

Your experience with ACPI is very common.

Here's what Linus Torvalds had to say in 2003: "ACPI is a complete design disaster in every way. But we're kind of stuck with it. If any Intel people are listening to this and you had anything to do with ACPI, shoot yourself now, before you reproduce." http://www.linuxjournal.com/article/7279

This is why Linux now claims to be Windows.

The striking parallel with user agents makes a good case for a generalized design rule of "don't assume anything about the ability of the other side to handle your side's specifications. Let them make the decision."

> horrendously overengineered by committee

Yes, but it doesn't explain the rationale behind such a decision.

You're totally right, but i didn't see that article until you posted it, and I thought the content might be interesting to HN.

I didn't mean it as a criticism, just trying to add to the conversation.