Re: sponsor rules
On Mon, 16 Jul 2001, Susan G. Kleinmann wrote:
> I guess I think that somebody who sponsors an upload under any of the
> conditions which related to this package:
>
> No lintian check.
> Not installable.
> No NM progress -- not even an ID check.
>
> should lose the privilege to sponsor packages until all of the above
> (and any other serious shortcomings) are resolved.
Having done a fair amount of sponsoring, I agree with the first two, but
not the third. If someone takes the initiative to volunteer for Debian
and package up some software, I don't care or need to know who they are.
Nor should I trust them just because I know *do* who they are. In many
aspects, it's the same sort of relationship a maintainer has with the
upstream author. (When's the last time a maintainer was required to check
the ID of upstream?)
The way I see it, any sponsored upload should be checked out as if it were
my own package. When I sign it with my key, I'm saying that it's
presentable to the archive, not generally dangerous, etc. This means that
the sponsor has to take a really close look at the software and the
packaging job (if not every time, then at least the first time, and then
check out the diff for any future uploads). All of us make mistakes,
forget things, etc., so I'd propose that you deal with the sponsor of bad
packages the same way you'd deal with the maintainer of bad packages
(however that works). (Aside: lintian is a godsend for sponsoring.)
* voivod dons an asbestos suit...
tony@mancill.com | An ounce of perception,
http://www.debian.org | a pound of obscure...
| (Peart)
Reply to: