On 7/13/12 11:39 AM, Logan Bell wrote:
I agree it's a small fix, but we must ask the obvious question what
justifies a bug release? Looking through our wiki I don't see any sort
of guidelines for this. My instincts tell me that since we're a small
project that quick bug releases to satisfy users of the project is the
best way to grow the project's user/dev base. Or do we wait until we
have N number of bug fixes? In my mind, a bug fix releases should be
fast and prudent, and maybe each bug fix release should be decided upon
as needed.
I'll be happy to outline a wiki page describing what justifies a bug fix
pending the outcome of this chain.
documenting some guidelines would be a Good Thing.
I guess I'm going off my own instincts based on my CPAN contributions.
Unless a commit represents a fix to a failing test or a bug someone has
reported, I don't bother releasing a new version until I've accumulated
what feels like a critical mass of change. It's a fairly arbitrary
decision-making process, I admit, and propelled mostly by feeling rather
than rational guidelines. Here are some of the questions I find myself
posing to myself:
* without this release, will people fail to use the code?
* how long has it been since the last release?
* will releasing this code gain me users?
* will releasing this code lose me users?
* will my mental cache, where I carry around a seemingly infinite set
of TODOs, feel better being flushed of responsibility for this set of
changes?
Releasing my personal code to CPAN is easy: I just decide, and can push
out a release in a minute or two.
Releasing the Apache Way requires a VOTE thread, which requires building
of artifacts, testing those, several svn commits, etc.
So my burden of proof, my release threshold, feels significantly higher
for Lucy than it does for my CPAN code. That's because I'm lazy. I find
solace in the fact that laziness is sometimes a virtue.
I have no philosophical objection to releasing a 0.3.3 to fix this CPAN
metadata issue, as long as I don't have to do any work myself. That's my
laziness talking. OTOH, I did make the effort to take Grant's patch and
commit it, within a few hours of seeing it on the list, so my laziness
is selective.
/end-of-brain-log
--
Peter Karman . http://peknet.com/ . [email protected]
