Consensus on the kdelibs coding style

[Aaron J. Seigo]

On Saturday 05 July 2008, David Jarvie wrote:
> On Sat 5 July 2008 20:25:19 Maksim Orlovich wrote:
> > > this discussion was had quite thoroughly, including on this list. this
> > > is the
> > > one point there was broad consensus on, in fact. restarting
> > > conversations evern N months is not useful, and we have mailing list
> > > archives for this purpose.
>
> When this was discussed previously, I didn't think to question the
> desirability of relatively detailed coding rules. I happen to have had the
> new (to me) thought that this isn't actually necessary. That's why I voiced
> this opinion. Or is that not allowed?

it's not a matter of "allowed" or "not allowed". it's a request to keep things 
moving rather than stalling by re-iterating conversations. it's a bad habit we 
have in kde: keep repeating ourselves until we get what we want.

we don't do that to people's applications, but we certainly do it for the 
libraries. this is because the libraries are shared by many people, even 
though the number of people who work on them isn't huge. we exhibit this 
behaviour for many of our shared resources and it usually prevents us from 
coming to a decision and then sticking with it. sometimes we need to respect 
the effort made in coming to a decision versus the decision itself, mostly when 
that decision is one of those judgement call type things and which isn't have 
a measured, detrimental affect on things.

which pretty much describes the coding style to a 'T'.

now, perhaps you weren't aware of the conversations previously had, so i 
thought i'd mention it. i'm sorry you saw that as dissalowing things. not the 
intention.

> > If it's a broad consensus, why are there so many disagreements, and why
> > you telling people who disagree to shut up?
> >
> > May be the reason there is so much disagreement again and again is that
> > coding style is the matter of personal preference, and people dislike it
> > being forced on them?
>
> KDE is a free software project.

no argument.

> Beyond setting some basic rules, why can't
> freedom apply to coding style as well, especially since it's something that
> a lot of people have strong (and divergent) views about. AFAICS, the only

you are free to code however you want in your projects. when you are working 
with others, however, there's a certain amount of "working with others" that 
ought to come into it. that "working with others" extends to coding style for 
people working on kdelibs, just as it does to people working on kdepim, 
plasma, etc. 

what you're really asking for is for those who are working on these code bases 
to not set guidelines for their projects. that's a freedom each team should 
have, imho.

just like we have user interface guidelines, policies against non-translatable 
UIs and other such things, this is just one more agreement.

> justification for an official coding style is to ensure code is readable.

readable, maintainable, etc. yes.

> That doesn't require detailed rules.

they are guidelines, not rules, and they aren't exactly detailed. i have a 
binder here that's 2 inches thick on coding style from a government project 
that's detailed ;) so yes we can also discuss what is detailed and what isn't.

the current guidelines are not onerous. they haven't caused problems in the 
code or the pace at which the code is being worked on.

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Trolltech

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 194 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20080705/c2d88b9f/attachment.sig>