Re: Report from the debconf11 sponsoring/mentors BoF.
On Sun, Sep 11, 2011 at 1:44 PM, David Bremner wrote:
> Also it was discussued how to avoid follow-up RFS mails for successive
> uploads to the debian-mentors mailing list. Those are typically a sign
> of a non-functional mentoring process, as it implies that either the
> maintainer did not know about the suggested tight relationship between
> sponsor and maintainer, or the sponsor did not reply in time. In the
> first case the maintainer should learn about the whole concept of
> mentoring. Here, the Debian Women project was mentioned who apparently
> provide every new contributor a fixed mentor, e.g. similar to the
> application managers for NM. In the later case no obvious answer was
> found. Maybe a sponsor should not consider sponsoring a package, if he's
> conceivable short of time or unwilling to sponsor successive uploads.
> This is, because it leaves a contributor without possibility to maintain
> his own packages, without sponsor who would advocate him at some point
> to come over this problem and the software archive with
> old/outdated/buggy software.
I personally encourage people I have sponsored to send future RFS
mails to debian-mentors instead of me personally. I do this for two
I prefer to do the sponsoring/mentoring work in public so that it
benefits other sponsees and provides example things for other sponsors
to check for or think about when reviewing packages. Also this way of
working is part of the culture of Debian and the spirit of our social
If I am busy then it gives an opportunity for other sponsors to look
at the package and provide reviews or uploading.
> Finally it can be observed many developers don't take part of the
> mentoring programme at all. This is something which can maybe be
> improved, once it is disclosed what discourages them from
> sponsoring. Several people confirmed they would start to sponsor
> packages again, once the metrics idea is implemented, as they are not
> willing to sponsor everything.
I would encourage developers interested in starting to do sponsoring
to join teams. Several teams use a tool called PET that mostly
eliminates the need for RFS mails since it provides a status/todo
board for all packages maintained by the team. The perl modules team
is probably the best example of this.
> I (Arno) am currently staging a patch to Debexpo introducing limited
> support for sponsor metrics.