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

Christian Schwarz <schwarz@monet.m.isar.de>



Christian proposes:
...
> [The last sentence is completely new: Currently, a few people don't have a
> working forward file on master or don't check their mail box there.]
> 
> Usually, a package has exactly _one_ maintainer.
> 
> Only in rare situations, a package will be allowed to have several
> maintainers. This is a special policy exception for a single package and
> that exception has to be approved by a discussion on debian-devel. The
> `Maintainer:' field of such a package would have to use the following
> format:

You're being (a) unclear and (b) overly restrictive.  You imply some
kind of permission is required for having several maintainers for a
single package.  This is not / should not be the case.

>    `MN1, MN2, MN3, ... <email@host>'
> 
> The maintainer MN1 is called `coordinator' of the package. (Note, that the
> exact syntax with the commas `,' is important since such maintainer fields
> need to be parsed by scripts.)

This doesn't allow well for largeish groups.  What if they want to put
the group name in the `phrase' part of the email address ?  Will it
not cause confusion if different users of the address put in different
comments ?

I think you need an exception mechanism, for things that otherwise
don't fit.  Computer-based systems with no good manual override are
often a bad thing.

Furthermore, commas are no good because they're already a separator
for separate addresses in a single field.  (Admittedly we already
allow a syntax like   John F. Bloggs <email@host>  which is not
permitted by RFC822.)

I suggest that in cases where a package is maintained by several
people the list of people _not_ necessarily be kept in the developer
DB.  If this causes some maintainers to appear not to be doing
anything we can add them specially, or something.

Ian.


Reply to: