Hey folks, Dan and Hans have been working on locating memory leaks the past few days. The culminating event has been Dan's work on adding garbage collection to DBMail.
I really don't want to shoot you guys down on this one, I like the idea of moving to a future with GC and memory pools and all sorts of useful stuff... just that I've really put my butt on the line with this temporary thaw thing, and I *really* don't want to see it go on a single minute longer than is needed to fix the specific items that we identified last week and which I talked most people into believing were worth fixing with breakage before the 2.0 release. Basically, I think it's a bad idea to change the memory management scheme right before a release. That might sound funny given that I've been advocating for lots of other breakage before the release, but my thinking is that memory leaks and even GC are internal issues which can be changed during the stable series without disrupting how people and other software interacts with DBMail. Dan, Hans, defenses for garbage collection? Thoughs, questions, comments? Aaron
