[Kde-scm-interest] Meeting minutes

[Oswald Buddenhagen]

On Tue, Nov 17, 2009 at 11:22:50AM +0100, Thomas Zander wrote:
> On Monday 16. November 2009 21.43.18 Oswald Buddenhagen wrote:
> > irrespective of the actual numbers i'd venture the guess that on
> > average, individual developers would still save disk space, as they
> > wouldn't have to download stuff they have no interest in. that
> > applies to app developers in particular.
> 
> This assumes something that is in general currently not the case; it
> assumes you can build something like marble without having the cmake
> files of the module present.
>
everyone is perfectly aware of the effort required to make it work.

> Anyone claiming this works today using svn is not being completely
> honest with himself.
> 
nobody is claiming it.
but i am claiming that it is less expensive with svn. svn doubles the
size of the checkout compared to cvs or a tarball, git typically triples
it (that's just collected from a few projects. generally, the more
disorganized a project is, the more revisions it has, and the bigger the
history is). and don't tell me anything about shallow clones, as that's
besides the point: i cannot make a shallow clone of the module i'm
interested in, as otherwise i cannot push anything to it. which means i
*have* to accept the overhead of the other projects which happen to be
in my module.

and before you start again, remember that the size issue is not my only
argument - it isn't even my main one (though it certainly *is* for many
people out there).

> Oswald; the earliest suggestion from you to have split modules like
> kdelibs and kdebase and kdeedu split up goes back to well over 2 years
> ago[1]. While I think its fine that you have an opinion, I think its
> not so fine if you continue to object to people doing the actual work
> without actually doing work to move in the direction you want.
> 
uhm, no. i'm just fine with the status quo (git-svn, svn and
websvn). this configuration is suboptimal, but it kinda works. now, if
you want to change things, you better make sure that a) it doesn't get
worse as such and b) that you first fix things which can't be fixed
later exactly *due* to the change.

and yes, i don't care how much work others put into it if the result is
still sub-par. i didn't ask for anything. do it right, or don't do it at
all.