Implementing the correct behaviour is not the best choice if the rest of the structure suffers functional regressions due to it by not being prepared. I know the 1.7.x versions are development releases but if something is not done its go into the stable release.
My understanding is that Ken's work also fixed real applications, and therefore reverting it would also create regressions. At least when we take the correct approach, the bugs can be fixed without introducing new ones.
We shouldn't be making stable releases when a feature like WGL is in the middle of major work.
This bug http://bugs.winehq.org/show_bug.cgi?id=33125 and the latter fix.
If it's the patch adding DECLSPEC_HOTPATCH to CreateSemaphoreExW, that's already in wine and has been since 1.5.28.
"Patch which disables optimization for CreateEventExW" would be rejected in seconds. It looks like something that only works by coincidence.
It seems Anastasius Focht figured out what was really going on and wrote this comment: http://bugs.winehq.org/show_bug.cgi?id=33125#c110 . So it just needs someone to write the patch and maybe add a test.