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

Re: Thoughts on Debian quality, including automated testing



On Thu, Dec 22, 2005 at 10:43:34AM +0100, Thijs Kinkhorst wrote:
> On Thu, 2005-12-22 at 08:38 +0000, Andrew Suffield wrote:
> > On the other hand, I think there might be some benefit to requiring
> > that the Maintainer field must always denote one single Debian
> > developer, who would be the "buck stops here" guy for that
> > package. Not an applicant, not a mailing list, and not a group of
> > people.
> 
> This "not an applicant" thing is a bad idea. As you might know, the
> NM-process is designed around the idea that someone has to prove they're
> up to the task they want to do. That's why for packagers it's required
> to have packaging activitity. Disallowing them to have the final
> responsibility over a package disables you to evaluate whether they're
> actually fit for this responsibility.

Actually it doesn't, you could work it out something like this:

The maintainer is a developer who is ultimately responsible for the
package. The applicant does most of the work. One of the primary
criteria for judging the applicant would be how much work the
maintainer has to do - the question put to them would be of the form
"Would you be comfortable with handing this package over to this
person, after watching them work for N weeks?".

The issues with the current system are that we end up with a lot of
packages being non-maintained by failing applicants, and we get a lot
of useless packages added to the archive just because an applicant
needed to find something of their own. This scheme should fix the
former and reduce the latter.

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ |
 `. `'                          |
   `-             -><-          |

Attachment: signature.asc
Description: Digital signature


Reply to: