On 05/02/2011 03:50 AM, Jürgen Spitzmüller wrote:
> Vincent van Ravesteijn wrote:
>> P.S. You are pretty quiet in that discussion ;).
> 1. I do not know git well enough. My impression from my spurious contacts
> with
> it are that it feels quite "different" and that I would need invest some
> learning energy in order to use it productively. But I really do not feel
> informed enough to give any serious judgement. (And currently, I'm lacking
> time to dive into the facts).
>
I have to say that I was quite worried about the learning curve here,
but having spent not very long playing with git, I've already adopted it
for my own work and will be trying to figure out how to integrate it
into LyX's VC system.
For anyone who's interested, these pages:
https://git.wiki.kernel.org/index.php/GitTips#git_svn
http://git-scm.com/course/svn.html
http://andy.delcambre.com/2008/03/04/git-svn-workflow.html
give some good advice on getting started using git with our existing svn
repositories.
Using git really effectively, with branches and merging and such, will
take more time for me to learn. But the upside is really big.
> 2. I leave the decision to the new crew in the cockpit. Obviously, the
> current
> workflow is something I find not so bad, since I have driven that workflow
> myself (of course, I agree with everybody that the 2.0 cycle was too long).
> But this does not mean that I'm opposed in trying something new. On the
> contrary, now is a good moment for that. So while I do not think the
> established model is in need for a radical change, I won't try to defend it
> either, in general.
>
One point I've been meaning to make for a while is that, from my point
of view, a major reason the 2.0 cycle was too long was the email outage
we had a year ago. Not only did it completely suspend work for quite a
long time, it came at exactly the wrong time for me: when I otherwise
would have had time to do a lot of LyX work.
Richard
