>To me, as someone who actually wants to use the software Oh, is that what I'm doing? Gosh, all this time I thought I was doing meaningful work using the softw

>To me, as someone who actually wants to use the software Oh, is that what I'm doing? Gosh, all this time I thought I was doing meaningful work using the software. >woefully unstable Is that really your experience? I mean, I run OSX on all the machines in my house, and work in it all day every day, and I haven't found it to be unstable at ALL. It might not be as rock-solid as it was 5 years ago, but it still manages to stay up and reasonable for months at a time, so it's a distinction without a difference. My Windows machines can barely manage a few days without something coming along that requires a reboot to resolve. >There is no such a thing as "most users" That's not how math works. When I said "If it's not a documented part of the programming environment, use it at your peril," I absolutely mean the developer. Don't use undocumented behavior. It's a dumb idea, and it absolutely WILL bite you. I mean, this isn't hard; if $vendor says "that's not documented or supported, and we can and will change that behavior," then it's malpractice to build a product that depends on it. That MSFT has been less strict about this is a problem, not a feature. >Your argument would be wrong It, like all arguments, might well be -- but you certainly haven't demonstrated that it is. Windows still enjoyed desktop hegemony because of momentum, and because you can buy a shitty Windows laptop for $200. It's not because Paint Shop Pro from 1999 still runs on it. >i'd rather them learning nothing. Well, you're absolutely getting your wish, because that appears to be what they've learned.