[Koha-devel] RM and QA roles (important!)

Paul Poulain paul.poulain at biblibre.com
Mon Jan 14 09:36:57 CET 2013


Le 13/01/2013 15:29, Jared Camins-Esakov a écrit :
> Please consider lesson 1, but this e-mail is actually about lesson 2. QA
> is an inherently time-consuming process, and the pool for QAers is much
> smaller than that for signing off. I expect that as we approach the
> deadlines for 3.12 the number of bugs awaiting QA will decrease.

Hi Jared,

This morning, while I was under my shower, I was thinking of this
problem. I can't dedicate as much time as I'd like (I've a business to
run, hdl has left now). Jonathan does a great job, but we (BibLibre)
can't do everything. So I'm very happy with this mail !!!


lesson 1=
Everybody should also realize that having patches waiting for signoff
for more than 3 months results in rebase needed. And rebase needed
result in time consuming, that could be used elsewhere. Last week, the
10 oldest bugs where all enhancements waiting for more than 4 months.
8 where from BibLibre, 5 have been rebased, Julian spent around 6 hours
to do that. Hours that could have been used to signoff someone else
patch. I don't know what to do with this situation, I just can say that
it's unefficient. If someone has a suggestion... (I made some,
months/years ago. You know my opinion, I know many of you disagree, but
any proposal would be carefully examined by me, for sure !)

lesson 1 (again)=
I've started last week to send an email to some french librarians that
were at the hackfest, and have a personal sandbox (that BibLibre setup
for them). In this mail, I point 10 of the oldest patches that they
could signoff. I hope it will produce some results. I'll do that for at
least 2 months, then see if the effort produces some effect.

lesson 2=
+1 to have you QAing patches, of course. That's how I made it last year,
I think 3.10 was pretty stable. I'm sure you'll be wise enough to know
which patch would be worth having another set of eyes, and ask the QA team.

PS: I see on the signed-off list that 1/3 of patches don't have a
complexity set. That's a shame, because "trivial" and "small patches"
could be QAed by you fist.
-- 
Paul POULAIN - BibLibre
http://www.biblibre.com
Free & Open Source Softwares for libraries
Koha, Drupal, Piwik, Jasper



More information about the Koha-devel mailing list