[Dune-devel] Private write access for Andreas

Steffen Müthing steffen.muething at ipvs.uni-stuttgart.de
Wed May 8 14:23:31 CEST 2013


Hi Martin,


Am 07.05.2013 um 21:17 schrieb Martin Nolte:

> Hi Markus, hi all,
> 
> my opinion on this matter has not changed. There is simply no need to keep private branches on the dune-project.org repositories. Even worse, everyone who clones these repository gets a full copy of all these private branches.
> 
> As for my private branches: I have already moved them to separate repositories. Their copies on dune-project.org are abandoned and can be removed. Would someone with full access and sufficient knowledge (e.g., Steffen?) be so kind as remove them, possibly also garbage collecting the repositories?

I purposely didn't delete any branches that were converted as part of the automatic process from the Git repositories
on the principle that deleting them is easy, but restoring them less so. If there are any personal or feature branches you don't
need anymore, you can easily remove them with

git push origin :branch/to/delete

(note the colon in front of the branch, that triggers the deletion). Manually repacking the repositories isn't necessary, that happens
automatically at scheduled intervals (and when you clone the repository, Git's smart-http protocol will only send you reachable
revisions anyway).

> 
> Best,
> 
> Martin
> 
> PS: Funny enough, my private dune-common branch (which I deleted from svn) has reappeared during git conversion.

That's part of the problem that Git and SVN don't match up perfectly, and the conversion sometimes doesn't pick up branch
deletes / moves correctly (to SVN, they are just file operations after all, and the conversion script tries to make an educated
guess). That was especially fun with the releases/ -> tags/ rename in some repositories…

Steffen

> 
> On 05/07/2013 06:16 PM, Markus Blatt wrote:
>> Hi,
>> 
>> On Tue, May 07, 2013 at 04:31:14PM +0200, Christian Engwer wrote:
>>> we already have private branches, namely for Robert and Martin. We
>>> explicitly suggested this setup in the proposal for the repository
>>> layout. All I want to do is use what we decided upon.
>> 
>> I beg to differ. I already ranted against this _during_ the proposal
>> and was ignored by everyone but Martin, who actually agreed with me
>> but wanted to keep his private branch during an interim period.
>> 
>> IMHO these branches were there due to a limitation in
>> subversion. With using git this picture has changed and so should our
>> mind. There is simply no need for these branches anymore as we can use
>> additional remote repositories for this and decrease the
>> administration effort, see my other mail.
>> 
>>> 
>>>> If Andreas works on some features that are/might be of a general
>>>> interest, shouldn't we create a feature branch with a speaking name
>>>> for it? Then people/developers at least know the context the commits
>>>> belong to. This would at least allow for meaningfully filtering the commit
>>>> messages. (Filtering on a person's name seems rather inadequate to me).
>>> 
>>> We will not have feature branches for all the bug fixes. The idea here
>>> is not to have a private repo with all private new features that
>>> nobody else will use. Still a feature branch can only be created by a
>>> developer, while Andreas can create them easily in his own
>>> "namespace".
>> 
>> He can also do this in his own mirror of DUNE without any
>> administration overhead on you besides adding his repository as a
>> remote.
>> 
>>> 
>>>> Additionally, your approach seems to implement another caste to our
>>>> developer system. Before we had: normal people/contributors,
>>>> developers with write access, and core developers. Now there is
>>>> another one (contributors with private write access) in between the
>>>> first two. Thus DUNE might even more be perceived like an elite clube and
>>>> further keep others from contributing. Do we really want this?
>>> 
>>> That is the reason to add this. We have some "rules" about when to
>>> grant write access. There are good reason for these rules and there
>>> reasons against these rules. in my opinion these branches are making
>>> it easierr to contribute.
>> 
>> Consequently we would need to establish user ids and private branches
>> for everybody who has/will ever contribute...
>> 
>> Markus
>> 
> 
> -- 
> Dr. Martin Nolte <nolte at mathematik.uni-freiburg.de>
> 
> Universität Freiburg                                   phone: +49-761-203-5630
> Abteilung für angewandte Mathematik                    fax:   +49-761-203-5632
> Hermann-Herder-Straße 10
> 79104 Freiburg, Germany
> 
> _______________________________________________
> Dune-devel mailing list
> Dune-devel at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune-devel

Steffen Müthing
Universität Stuttgart
Institut für Parallele und Verteilte Systeme
Universitätsstr. 38
70569 Stuttgart
Tel: +49 711 685 88429
Fax: +49 711 685 88340
Email: steffen.muething at ipvs.uni-stuttgart.de

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 495 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <https://lists.dune-project.org/pipermail/dune-devel/attachments/20130508/4bd4e81e/attachment.sig>


More information about the Dune-devel mailing list