usrmerge testing in our CI

[Chris Lamb]

Holger Levsen wrote:

> <helmut> as soon as buildds are merged, varying trixie no longer makes 
>          sense to me in either case
[…]
> so should we stop testing usrmerge variations at all now?

Thanks for taking this to the list. For 100% clarity, I understand
Holger's "at all" suffix to mean, "shall we disable the usrmerge
variation for bookworm as well as trixie?"

Assuming that's the correct interpretation, from the perspective of
someone patching packages to make them reproducible, it no longer
makes sense to prepare patches that address any usrmerge issues. At
the very least, they will not be prioritised by package maintainers.

I'm less clear about whether we should cease testing bookworm. It
doesn't seem right for the CI to claim that various [bookworm]
packages are "reproducible in bookworm", when the presence of usrmerge
(or lack thereof) in bookworm means that they can still vary.


Regards,

-- 
      o
    ⬋   ⬊      Chris Lamb
   o     o     reproducible-builds.org 💠
    ⬊   ⬋
      o