Rick Hillegas (JIRA) wrote:
I would allow a patch which didn't work on Java 6 if it got other people unstuck
> and if I trusted the developer to submit a fix soon.
Apart from the issues of causing other folks grief as they need that
platform working, one issue is that in an all volunteer development
effort it's impossible to assume that anyone will follow up. We have had
at least one patch where follow up work was promised but never happened.
Stuff happens, people move on, if they are being paid to work on Derby
their boss tells them to do something else.
Much better to always try to have the code in a working state.
Of course from time to time a patch will unexpectedly break a platform
or a specific feature as it's not required that all tests are run on all
possible platforms. Then someone just volunteers to fix it by some
means, but in this case the J2ME issue was raised over a year ago.
Dan.
