The WINE project will turn 15 years old this June, and to celebrate there will be the version 1.0 release of WINE (WINE Is Not an Emulator). The 1.0 release plan puts two release candidates in May and then the final release on June 6, 2008...
Wine reaching a 1.0 stable version is very important for distributions - we may even see some "Wine by default" things spring up.
I'm pragmatic about it, personally, but I do see the bad things that could come with Win32 support (see: OS/2 - The windows compatibility meant that devs only released windows versions of software.) Likewise, if I can support Linux and windows (and OSX, and BSDs, and so on since Wine is everywhere) with one executable, why am I going to bother with any other versions of my software?
On the other hand, people clamor to be able to play their favorite games in non-windows OSes, the latest version of quickbooks, <insert proprietary software here>, and they can finally make the jump to Linux.