Re: rejections and reprocessing etc.
Hi, sorry for much delay. I've been much busy these days
(as well as other volunteers in this team, including FD&DAM).
In <87puedswu0.fsf@dt-jtlaptop.datatrans>,
on "15 Apr 2001 21:59:51 +0100",
with "rejections and reprocessing etc.",
James Troup <james@nocrew.org> wrote:
> Rejections
> ----------
>
> We have a couple of rejections coming up; here's what I'm proposing to
> do. There will be two types of rejection: hard and soft rejections.
Thanks for explanation of both types.
> Soft rejections is my way of saying ``Do I really have to?''; i.e. I'm
> not happy with a candidate, but I'm not unhappy enough to resort to a
> hard rejection. A soft rejected applicant will be put on hold in the
> DB and this list will be notified. I'll then leave it up to you guys
> to decide what to do. If the applicant gets enough "yes"
> votes/approvals/whatever, I'll consider my initial decision overturned
> and accept him under protest. I'm not dreadfully fussed what the
> magic number of approvals is and who can make them, so I welcome
> suggestions on that detail.
Are there already these 'On-Hold by DAM' indicator in our NM DB ?
I welcome this sort of notification, and your consideration on
'overturning of initial decision'.
Well, if the applicant plans to work on www/doc translation for
some minor languages, then he can't get much approvals, but I think
the needs are high when the translation team doesn't have enough
members. Yes, I know that anyone can be committers on main WWW cvs
without being one of Debian members, but I think that he could be
a good ambassador from Debian for people who use that language when
he is a member of Debian Project officially. He can do effort to
promote our ideal on Free Softwares even without the membership,
but the response of the audience will differ when the speaker is
an official member of the project or not.
So I propose that only one vote will be enough on translation task
when that vote comes from one of our www admins or one of translation
coordinators.
On packaging tasks, the criteria would be a little high, but I suppose
that almost all applicants can't get more than 10 votes. maybe 3 or 5
is enough, when no one cast the against on that case.
(Of course, the final decision always comes from our DAM.)
> A hard rejected applicant is rejected; no discussion. I consider this
> part of my delegated powers under the constitution and (FWIW) the new
> DPL agrees. You can overturn this decision (no power in Debian is
> absolute) by way of a GR, but that's about it. Note neither of these
> rejection methods are intended to be common place, hard rejections
> especially. If I hard reject someone I genuinely believe that I can
> not in good faith trust them to be a Debian developer.
I see.
> Reapplying from a soft rejection will require an 'ok' from the NM
> committee after an as-of-yet undetermined time span. Details on the
> length of the time span and how an 'ok' is generated are undecided and
> again suggestions/comments welcome.
How about 3 months as the time span of suspension ? More than 6 months
are too long, and will be accepted just the same as the hard rejection
by people outside the project.
And, I think the procedure to generate an 'ok' may be:
1) inform the members of NM committee about that application.
-> How this will be done ? All mails to debian-newmaint-admin
will be forwarded to debian-newmaint-discuss, and published
at the web archive. So we can't use this list if that event
have to be kept secret.
2) ask the members for 'ok'
3) If there are enough members who express the 'OK', then
the application can go into the DB.
-> How many is 'enough' ? My suggestion is one-third (1/3) of
the total members.
> Reapplying from a hard rejection will require an 'ok' from DAM (and NM
> committee too, if you want; I'm not fussed, but the DAM part is
> non-optional). The time span for hard rejections will be determined on
> a case by case basis.
I wish to have notification. How about the similar procedure to
the proposed one above for soft-rejected applicants ?
Maybe more numbers (more than a half, 1/2+1) are required to generate
'ok' for this type.
> For the benefit of the applicant more than anything else, I don't
> intend to post details publicly of either form of rejection. I'm
> unlikely to get involved in a public dispute, regardless, but the only
> way it'll happen at all is if the applicant initiates it.
Then, do we need another list for private/secret informations ?
I think the current setup for open discussion with debian-newmaint-admin
and debian-newmaint-discuss should be kept as is. IMHO, we should open
our discussion as much as possible unless the absolute reason to keep
secret exists. So if this special case needs a closed list, then we
should request the creation of another list.
Regards.
--
Taketoshi Sano: <sano@debian.org>,<sano@debian.or.jp>,<kgh12351@nifty.ne.jp>
Reply to: