[Kde-games-devel] Move to git now?

[Ian Wadham]

On 01/02/2012, at 3:26 AM, Frederik Scharzer wrote:
> Am Montag, 30. Januar 2012, 20:27:21 schrieb Albert Astals Cid:
>> El Dilluns, 30 de gener de 2012, a les 18:28:59, Stefan Majewsky va 
> escriure:
>> 
>>> I'm just thinking about git vs. SVN again. Git would be very
>>> beneficial because the libkdegames5 work can start right now, in
>>> a branch. One of the main arguments against a git move currently
>>> does not hold: We are just past a major release, so it's not a
>>> big problem if development stalls a bit now because of the
>>> conversion.
>>> 
>>> But still, the problems with data files in git are not solved.
>>> Can we please do this now? In the other mail, I thought about
>>> migrating only libkdegames to git, and leaving the games in SVN.
>>> But that won't work. The games need to have a v5 branch, too, so
>>> changes in libkdegames5 can be tested.
>> 
>> Things that need to be decided:
>> * Data goes into the same repo than code or not?
>>     My vote is that data is one with code and they should stick
>> together
>> 
>> * One repo per app or a global repo
>>     My vote is that since I want data to stick with code and we do
>> not want a huge repo we might want to split the repos
>> 
>> Anyway, whatever the decision is we still *need* someone to do the
>> *work*.
> 
> Since we did these opinion gathering several times now, I decided to 
> start a wiki page where we could manage all this so it will be more 
> visible than the mailing list.
> 
> Please add yourself to the first table. I hope it is not too messy. I 
> did not figure out how to get lined between the cells otherwise.
> http://techbase.kde.org/Projects/Games/MoveToGit

Thanks, Frederik, I made an entry.

Re this whole discussion, I have just had two full days of baby sitting
grandchildren, followed by a family gathering to welcome home my son
from his trip to Japan, so I have not had time to reply before.

Firstly, I deplored Christian's contribution and also Lindsay's response,
but I was very pleased to see the way this group handled both.  You
showed great maturity, guys, better than kde-devel a few months back.
Thank goodness we did not get an endless thread about the use of four
letter words… And we did not need to go to full moderation of the list.

Re the move to git, I cannot see any real benefit in it, neither for myself
nor for the group, so why do it?

OTOH I am prepared to go along with the group changing "now",
but was not so prepared a couple of months ago.

I had huge commits to do in KSudoku, as well as being under pressure
from several other directions.  I asked back then what the benefits are,
but got no answer.

AFAICS, the only reasons given for the change are like "it's the latest",
"Linus wrote it", "the rest of KDE is doing it", "SVN is old hat", etc.  I do
not see these as being valid reasons on their own.

I am very happy with SVN.  It keeps the code safe, has a full audit
trail, allows a revision to affect more than one file and is easier to use 
than CVS was.

So if we move to git, what will be the benefits for individual programmers
and for the KDE Games group in making the change?  Will the gains in
productivity (churning out games code) outweigh the effort involved in
changing?  How do we know?

Making the change for its own sake does not appeal to me.  I would
rather spend time working on games code and producing some benefits
for our users.  Besides, admin of source code control was one of my
tasks before I retired and joined KDE Games.  I have no desire to go
on a bus driver's holiday … the one where he ends up driving the bus :-)

Assuming we change, I am in favour of splitting each game into its own
repository.  That will save everybody bandwidth when they only wish to
work on their own game.

Re the data-files problem, why bother with the history? I see
on http://techbase.kde.org/Projects/MoveToGit/UsingSvn2Git that you can
select the revisions you want, so why not just archive earlier data files
on SVN or its branches?  How often do you really need that history?
Is it a big problem to follow another link to get it?

One worry about going non-monolithic is how anyone outside KDE Games
(or inside, for that matter) can get a comprehensive list of all available
games and can build all games as a block if so desired.  And how will
people know when there is a new game or a major re-write or upgrade
of an existing game.  I think we will need a "superbuild" at least, but
that may not solve all problems.

Also, how will releases, mods to games and mods to libraries be
co-ordinated?  I do not want us to repeat any of the problems that
arose in the first release of KDE with git repositories, nor any of the
build problems I encountered early last year in trying to build Phonon,
KDE libs and all its dependencies after they had been ported to git.
There were teething problems there.

All the best, Ian W.