[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#381267: qa.debian.org: Confuses NMU/QA and sponsor of NMU/QA



On Tue, Aug 08, 2006 at 11:42:40PM -0500, Steve Langasek wrote:
> On Mon, Aug 07, 2006 at 08:57:14PM +0200, Lionel Elie Mamane wrote:
>> On Thu, Aug 03, 2006 at 03:20:32PM -0700, Steve Langasek wrote:
>>> On Thu, Aug 03, 2006 at 11:37:17AM +0200, Lionel Elie Mamane wrote:

>>>> http://qa.debian.org/developer.php?login=lmamane&comaint=yes says I've
>>>> NMUed thinkpad, while what I've done is sponsor an NMU by
>>>> jdthood@yahoo.co.uk.

>>> [You] (among other things) should consider yourself responsible for
>>> any bugs introduced in that upload.

>> That is correct. But because I *sponsored* the upload doesn't mean the
>> person that prepared is "off the hook".

> True, but that didn't seem to be the issue you were raising in your
> bug report; you seemed to be objecting to listing the package on
> your own QA page, not to its absence from jdthood's page.

I was objecting to the *category* it is listed under on my QA page. I
*want* it on my QA page, but the category it is listed under seems to
me to be factually wrong. So, I am not asking for it to be removed
from my QA page, but categorised differently. Sorry for the
misunderstanding.

I indeed hadn't noticed its absence from jdthood's page when I
originally filed the bug.

> You're welcome to credit anyone you want, but if you sign an NMU,
> it's you that I intend to hold responsible for the contents.  Your
> responsibility as someone uploading (or authorizing the upload of)
> changes to another maintainer's package is to *personally* verify
> that these changes are appropriate and correct, and to remedy them
> if you are mistaken.

I completely agree with this. I would also say:

 If you sign a MU, it's you that I intend to hold responsible for the
 contents.  Your responsibility as someone uploading (or authorizing
 the upload of) changes to a package is to *personally* verify that
 these changes are appropriate and correct, and to have them remedied
 in some way or another if you are mistaken.

> This is not a responsibility that can be delegated, and so for all
> intents and purposes, this is your NMU at least as much as it is the
> NMU of the sponsoree.

It seems we are stuck in a definitional disagreement (how to _call_
things), but that we agree on all other practical consequences. I
agree that I'm responsible for it, but I wouldn't call it "my NMU",
only "the NMU I sponsored". The same way a MU I sponsor is not my MU,
but ... "the MU I sponsored". My responsibility when I sponsor an MY
is to *personally* verify that these changes are appropriate and
correct. But neither me (nor the QA page) calls it my upload; we both
call it "sponsored uploads by lmamane@debian.org". I just fail to see
what would make this any different with an NMU.

For me:

 Person in "Changed-By:" field of .changes file of the upload is
 responsible for the contents of the package (unless hir name is there
 without hir authorisation - then someone in the chain has committed
 fraud). That person is called "NMUer" if she is not in the Maintainer
 or Uploader fields of the package.

 The person that signed the .changes and .dsc file is responsible for
 the contents of the package. If that person is not the person in the
 "Changed-By:" field of the package, that person is called "sponsor".

>> To take it from the other side, do you think it is a bug that the
>> PTS shows Thomas Hood's name, and not mine?

> I'm not sure what you're referring to here.

The news item dated 2006-05-03 of
http://packages.qa.debian.org/t/thinkpad.html, with subject "Accepted
5.9-2.1 in unstable" says "Thomas Hood", not "Lionel Elie Mamane". Do
you think this is a bug? I ask because you insist to call it my NMU,
so in your opinion, shouldn't it by my name there?

-- 
Lionel



Reply to: