[53] in athena10
Re: Athena 10 project infrastructure status
daemon@ATHENA.MIT.EDU (Greg Hudson)
Fri Jan 25 13:15:16 2008
From: Greg Hudson <ghudson@MIT.EDU>
To: Tim Abbott <tabbott@mit.edu>
Cc: athena10@mit.edu
In-Reply-To: <Pine.LNX.4.64L.0801231648290.24893@phoenix.csail.mit.edu>
Content-Type: text/plain
Date: Fri, 25 Jan 2008 13:14:40 -0500
Message-Id: <1201284880.5853.26.camel@error-messages.mit.edu>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
On Thu, 2008-01-24 at 17:30 -0500, Tim Abbott wrote:
> We use effectively "debuild -sa -us -uc" instead of "debuild"
Okay, I modified dasource to specify those options.
> Changing the contact email on the packages to a list not containing Anders
> and I would result in the world having to maintain versions of the package
> sources with two different contact addresses, which would be annoying.
Agreed.
> In the long run, we'll probably have to do this, so that people contact
> the right support source
I don't think that's really necessary. IS&T developers don't need to be
completely insulated from support requests which should be handled by
SIPB.
So in the long run I'd like to see the maintainer address include both
the SIPB and IS&T developers. That could be debathena with some IS&T
people added, release-team with andersk and tabbott added, athena10 (not
sure I want the implied version dependency there), or a new list.
For now it can remain as debathena, assuming I don't wind up needing the
signing key in order to create our apt repository. (I haven't yet
really learned how signing of binary packages going into an apt
repository works.)
> If one were to increment the orig version precisely when one changed the
> package outside the "debian/" subdirectory, the orig tarball obtained via
> "tar --exclude=debian" would be a valid orig tarball that's always the
> same.
For now I've modified dasource to create such an orig tarball when there
is a Debian version component.
> Given that it seems MacAthena will be probably be using the orig versions
> as their upstream, there may be advantages to having a well-defined
> non-Debianized version of the sources.
I don't feel like the presence of the debian sbdir in the upstream
sources really harms the Macathena work, but I'm happy to defer this
issue for now. Not making sweeping changes to the imported Debathena
materials is good.
