You’re confusing a post on Joel’s forums with a post by Joel. In fact, Joel responded later in that thread to counter the original poster who asked the question.

Also, it’s becoming a bit of a cliche that MS is at its best when the compeition is the fiercest. Think of the arms race / space race analogy: Google’s success is having the sort of effect on MS that Sputnik had on the complacent US science and educational system nearly 50 years ago, or Netscape and Java had on the grandiose Cairo / Blackbird / etc. visions at MS 10 years ago. Lots of pontifications were issued back then about how the US was losing the Cold War or MS’ glory days were behind it, which looks pretty absurd in retrospect. Of course the prognosticators may be right this time around, but I’m betting that new blood and a new attitude will keep MS in the game.

Finally, I can’t resist a cheap shot: “new generation of distributed network computing technologies that will use the internet as the platform for new scenarios across new topologies, and leave the client to being an abstraction layer for UI” Ahh, that sure brings back fond memories of the good ol’ .bomb days, when all you needed was a domain name and a phrase like that to get a few million out of the VCs. Of course, actually delivering real value to people who don’t give a rat’s patootie about network topologies and abstraction layers turned out to be a tiny bit harder …