It was a crowd of Apple Mac OS X developers though, with a great deal of expertise in Obj-C, not the web, so it shouldn't be taken as an objective verdict on whether it was a good strategy or not.
I think it was inevitable. And probably planned for a second release too. Not that web APIs were lacking, but purely from performance perspective. By then, web was designed for 2GHz/2Gb desktop, not a 200MHz cpu from a dvd player.
Well he changed his mind more often and quickly than you can say "magical thinking". Still, better to seem certain when speaking to investors, developers and other get-rich-quick types.
On the other hand: it's okay to be wrong if you can admit that you're wrong, it's okay to change your mind, and, of course, circumstances can change.