Date: prev next · Thread: first prev next last
2013 Archives by date, by thread · List index


On Sat, Dec 7, 2013 at 6:30 AM, Norbert Thiebaud <nthiebaud@gmail.com> wrote:
On Fri, Dec 6, 2013 at 3:31 PM, Robinson Tryon
<bishop.robinson@gmail.com> wrote:
I felt like I was working hard to go through appropriate channels...

Yeah, you did well. I'm sorry I apparently started a bikeshed on this...
I have missed ESC of nov 21st (my fault), but 'the guys doing the work
should choose'
is not 'the guys doing the work should organize a beauty pageant election'

IMHO 'the guys doing the work should choose' means 'we're going to
leave it up to the best judgment of those people to decide'.

I thought that we had a bunch of okay abbreviations, and aside from
tdf# (which I wanted to reserve for future use), all of the other
names seemed good enough to me, and didn't raise any red flags when I
checked-in with the ESC.

So, with that information in hand, I figured that I would ask for
input from QA, first in the form of searching for consensus, and
second (if need be) as a voting process. QA doesn't get to make
decisions like this very often, so I thought it would be something fun
for us to decide as a team.

Alternatively, if we want to interpret this message as ' *only* the
guys doing the work should choose', then that's fine with me, too. By
my reckoning, that list includes

- Cloph (for fixing bots and websites, creating DNS redirects, and
generally putting up with all of my requests)
- Dennis Roczek (for a bunch of detailed bot work translating urls on the wiki)
- Andras Timar (for updating urls in the translations and core git repos)
- Tollef (Doing all of the work on the FDO end)
- Me (for doing a lot of legwork on the migration)
- Some others I've undoubtedly forgotten

On the dev side we are not used to 'vote'. decisions are usually just
taken. When there is some controversy, the interested parties expose
the merit of their respective positions, explaining the rational for
their choice (and I mean _rational_ not _feelings_).

ok

That usually lead to either a compromise to address each other points,
or the parties rallying around the rational of one side (we all have
'opinions' on anything when asked... but more often than not we do not
_care_ that much about a given topic, so unless we have a strong
argument to offer we usually do not demand that our opinions be
counted as strongly as the one of the people intimately involved with
the work and problems associated with it.. iow 'pick your battle
wisely'  :-) )), or more often a combination of both.
in 3 years we where driven to a vote only once.. and even then that
was more to have each position 'on the record'.

ok

So in that light I would point out again the criteria _I_ think are
relevant for the name here:

1/ short. the summary commit message is the 1st line of a commit
message, and is limited to 80 chars (72 preferred), and when a commit
refer to a bug we want the bug reference in that message.
2/ obvious meaning and easy to remember and type, as much as possible.

The opposition I've seen so far to lo# are centered around 'it can be
confused with the abbreviation of the product'. I think that is a
feature not a bug.
in the context of bugzilla the use will always be lo#<number> the #
make it clear that it is a bug number and remove any ambiguity...
furtheremore these _are_ libreoffice bugs.. lo = libreoffice, # = bug
here.  So, other abbreviation may have merit, and may prevails, but
discarding lo# for that reason seems a red herring to me.

for reference a quick grep of the log message give use the following uses:

fdo#nnnn
deb#nnnn
n#nnnnn
#innnnnn#
rhbz#nnnnnn
CID#nnnnn
cp#nnnn
bnc#nnnn
abi#nnnn
i#nnnn
#nnnnn#
lp#nnnnn

and a mix bag of some mistyped variations thereof (like #fdonnnn)

Up at the top of your email you said:
"I'm sorry I apparently started a bikeshed on this..."

...but the length and breadth of your email belies that
characterization of this particular decision about abbreviations. This
is clearly not a bikeshed decision, or you and Mat would (hopefully)
not be so invested in what abbreviation is chosen.

If it's a big deal, then yes, I agree we should pay more attention to
it. If it's not a big deal, then we go ahead and pick from the list. I
just need to get a clear message, one way or the other.

I'm sympathetic to input from the developers, even at a late stage in
this process, which is precisely why I listened to Eike's comments and
wrote "The devs really like lo# (as it gives them more room in commit
messages)."

Cheers,
--R

Context


Privacy Policy | Impressum (Legal Info) | Copyright information: Unless otherwise specified, all text and images on this website are licensed under the Creative Commons Attribution-Share Alike 3.0 License. This does not include the source code of LibreOffice, which is licensed under the Mozilla Public License (MPLv2). "LibreOffice" and "The Document Foundation" are registered trademarks of their corresponding registered owners or are in actual use as trademarks in one or more countries. Their respective logos and icons are also subject to international copyright laws. Use thereof is explained in our trademark policy.