[FFmpeg-devel] [GIT] ready by the weekend

[Jason Garrett-Glaser]

On Wed, Nov 17, 2010 at 11:14 AM, Reimar D?ffinger
<Reimar.Doeffinger at gmx.de> wrote:
> On Wed, Nov 17, 2010 at 11:06:31AM -0800, Jason Garrett-Glaser wrote:
>> On Wed, Nov 17, 2010 at 11:03 AM, Reimar D?ffinger
>> <Reimar.Doeffinger at gmx.de> wrote:
>> > On Wed, Nov 17, 2010 at 07:30:19PM +0100, Luca Barbato wrote:
>> >> On 11/17/2010 06:47 PM, Reimar D?ffinger wrote:
>> >> > On Wed, Nov 17, 2010 at 05:01:31PM +0100, Luca Barbato wrote:
>> >> >> j-b and janne should have all the issues sorted during this weekend (hw
>> >> >> problems, latest repo checks)
>> >> >>
>> >> >> I'll host a mirror of it and keep in sync a copy for github and
>> >> >> gitorious (somebody wants a repo.or.cz as well?).
>> >> >>
>> >> >> So I guess by the next week we can switch anytime.
>> >> >>
>> >> >> Is everybody ready? =)
>> >> >
>> >> > No. We don't have a "cheat sheet" for anyone only familiar
>> >> > with SVN,
>> >>
>> >> I thought there cheat sheets links got already provided many times.
>> >> Rafael sent them again anyway.
>> >
>> > You can send them 50 times and I still don't care.
>> > I can use search myself, the point is that not everyone has to use
>> > search themselves, find some document, figure it isn't quite ideal
>> > if you're already used to FFmpeg and then some things are explained
>> > in a way we do _not_ want them to be done.
>> >
>> >> > and we have not documented (have we even decided?)
>> >> > how to handle bad commit messages (and different levels of
>> >> > bad, ranging from spelling mistakes over non-understandable
>> >> > to wrong patch-by attribution).
>> >>
>> >> there isn't patch-by in git... You send your patch and it gets signed
>> >> off by whoever push it to the main tree....
>> >
>> > What? Are you suggesting we no longer document who wrote a patch?
>> > I really have not the slightest clue what you mean by this.
>>
>> git distinguishes "author" from committer, svn does not. ?Thus one no
>> longer has to manually track authorship as in svn.
>
> So? Git still certainly doesn't contain magic smoke that makes it impossible
> to have the wrong thing as "author", so I fail to see how that is relevant
> (except for the point that I used the wrong term).

The only reason anyone ever gets author wrong is because it's not part
of the patch.

Dark Shikari