[Kde-scm-interest] git layout planning

Richard Dale richard.dale at telefonica.net
Thu Sep 23 17:11:14 CEST 2010


On Friday, September 10, 2010 06:06:36 pm Chani wrote:
> okay, so, we've got a summary of the modules vs split repos thread at
> http://techbase.kde.org/Projects/MoveToGit/Layout
> 
> we haven't actually made a decision yet. :) however, that page also doesn't
> show all the information we need to make an informed decision.
> 
> the sysadmins intentionally didn't talk about the social implications; we
> now (as I type) have people working on a paper to address exactly that
> (btw, thank you, sysadmin and social paper-writers. *hugs*).
> 
> so as much as I hate to discourage work... writing split-layout rules is a
> bit premature at this point. :)
I am very happy with the technical work that the sysadmins have done, and 
can't fault it. 

It is not true that all kde svn modules should be migrated in the same way, as 
they have very different needs. Personally, I 'm thinking of  the kdebindings 
module, and I'm sure there are several other KDE modules that don't fit into 
some kind of cookie cutter pattern that the forthcoming paper might describe.

However, is 'people working on a paper' the best way to discuss the social 
implications of the move to git. Do the people working on the paper have some 
insights that the rest of us don't? I don't see why it can't be discussed on 
some mailing list like other topics. However, 'social implications' doesn't 
belong on this mailing list perhaps. Or the implications on KDE release 
schedules of the git move don't belong here either. Or the i18n implications 
of the git vs svn workflow. It is very hard to find some central place in the 
KDE community where all the aspects of the git move can be discussed.

> on the other hand, rules for big modules are also applicable to split
> modules, so feel quite free to help out with those ;) or I suppose there's
> extragear, where we've already decided the sensible thing is to split on
> each app. there's no shortage of tasks to do that don't depend on the
> outcome of this layout discusion. :)
If people are 'publishing papers' and handing down decisions by implication, 
then it isn't a discussion. I'm sure it can be an excellent basis for a 
discussion perhaps, but telling people to hold off doing practical things is a 
bad idea in my opinion.

-- Richard


More information about the Kde-scm-interest mailing list