[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: KDE development with git
From:       "David Jarvie" <lists () astrojar ! org ! uk>
Date:       2007-07-12 13:30:23
Message-ID: 37198.S0VKQUNDQxA=.1184247023.squirrel () www ! sensical ! net
[Download RAW message or body]

On Thursday 12 July 2007 12:40, Thiago Macieira wrote:
> Maurizio Monge wrote:
>>For instance being able to register a git repository on kde.org would
>>be a very good step in this direction (consider also that git does not
>>even require a "git" server, http or ssh will just do!).
>>I really don't understand your accept_or_die policy, if a project
>>would like to use a different VCS i think that no *very good reason*
>>is required.
>>Considering that KDE has the very liberal policy "you have write
>>access for almost everything" (an i'm happy of it), i don't understand
>>why you should impose "but you must use SVN!!!"
>
> 4) many people build all of KDE and do fixing commits here and there as
> needed. It depends on their goodwill to move to another system. Heck! I
> know many developers who took over a YEAR to move to Subversion after we
> decommissioned CVS! If we bring in several more repositories, those
> people will probably not build your code and test it.
>
> I think the most important points are 2 and 4. I believe that moving code
> away from Subversion means it'll decrease in quality. And it'll have no
> translations...

I completely agree. I don't think the transition to KDE 4 would have been
really practical if the repository had been split between different vcs's.
There have been many many API changes in kdelibs, and the authors have
gone through the entire repository making the necessary changes to
applications to ensure that they still build. If KDE had been split up,
probably only those parts actually in SVN would have benefited from this
attention, and the maintainers of the rest would have had to work out for
themselves how to adapt to API changes. That would have been a major drain
on the project.

Admittedly, now that the KDE 4 API is on the verge of being frozen, that
becomes less of an issue. But although KDE 4.1 will involve lesser
changes, we need to be ready to cope with that when the time comes.

-- 
David Jarvie.
KAlarm author & maintainer.
http://www.astrojar.org.uk/kalarm

[prev in list] [next in list] [prev in thread] [next in thread]