> Would you consider this a failure of the new commercialization of open source?
not really ... I'm about to set up a similar feed for the PHP bug system on bugs.php.net and think that the result will be sort of similar ...
A comparison of some bug statistics like average and median time to close over time may give some further insight here, but i don't have direct access to the actual bug database anymore and while retrieving such information would be possible by crawling the complete bugs.mysql.com contents i don't want to put that much stress on that web frontend ...
> I've been thinking, the days of anyone being able to contribute to a project like MySQL has been coming to an end.
Copyright assignment agreements and cathedral style development have been contribution barriers for quite a while already.
The rush to MySQL 5.0 ("Enterprise ready!") didn't help either and the code base is still suffering from that today.
So i won't completely blame the difficulties in being able to contribute to the current owner alone. As far as i can tell the current contributors agreement process is actually way less complicated as it had been when i joined MySQL in 2004.
The delayed release of ongoing code changes to the public source trees on launchpad, part of the bug handling process having moved to an internal tracker, hidden test cases and references to hidden internal bug reports or to CVE entries that only provide information like "unspecified bug in unspecified component" do not help the process either though.
Fortunately there are other places where your contributions can go now (MariaDB, Percona, Drizzle ...) and all of them seem to be more open to accepting code from the outside than MySQL has ever been in the last 8+ years ...