Trac: release management

[Martin Dengler]

On Fri, Jun 13, 2008 at 11:37:32AM -0700, Wade Brainerd wrote:
> Hello,
> 
> I agree wholeheartedly that ticket triage and statistical analysis is
> a worthwhile effort, so I think I support what Garrett is talking
> about.

To clarify, I advocated ticket triage in my response and I attempted
to make clear that statistical analysis was fine as long as it didn't
repress the development it was meant to analyse.

> At my employer we have teams of producers constantly watching
> individual and per-component bug counts, transfering bugs from
> overworked team members, ensuring progress is being made according to
> priority levels, and tracking the rate of change of the blocking
> ticket count compared with previous projects to estimate our
> completion date, among other things.

I think that's a great idea.  When OLPC/SugarLabs has the producers,
let's do it.  But don't make the developers do that, since any good
developer that I've known has always done all those things, already,
to the maximal extent it doesn't interfere with doing development
itself.

> I don't care how the open source world "usually" does it, releases
> don't happen unless you're on top of your tickets.

Just in case anyone thought I was arguing that doing something
suboptimally is OK because 'the open source world "usually" does
[that]', I wasn't.  I was just saying the cost/benefit case had to be
made if success is plausible without whatever's being argued for.

> Regards,
> 
> -Wade

Martin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.laptop.org/pipermail/devel/attachments/20080613/493e2c0f/attachment.sig>