let things be clear. We know produce frequent enough pkg snapshots that people who are interested in the quality of OpenBSD ports can test, all the time.
Problems we're not aware of DON'T EXIST. If it's not reported to the maintainer, or this ml, but vaguely on some irc channel where no openbsd developer ever sets foot, then there's NO BUG. As far as the release process goes, you can see the system go to beta. If you don't test the rest of the time, you should test THEN. People are still under the misconception that when we announce lock, there's still time to do 10s of innocuous bug-fixes, or fix critical problems that have been there for 2 months (see above paragraph, we need to be aware of said critical problems first). Nope. Lock is lock. It's darn near to actual release builds. Each monkey-wrench will not help, and there are usually enough actual critical last minute issues without new stuff that pops up suddenly because "hey, it's not release-time, so let's wait to report that bug". Just to put things in perspective, over the last 6 months, there have been a HUGE number of bug fixes and improvements... being able to build chrome in a fraction of the time it used to take helped bulk builds. Some rather large file system bugs have been quashed. The whole ports building network has seen some fairly major restructuring for speed. (and that's just a very small selection of things I like which I took from things I'm aware of among other things which I haven't followed closely at all).
