What "openmaintainer" means

[Perry E. Metzger]

On Wed, 25 Apr 2018 04:43:12 +1000 Joshua Root <jmr at macports.org>
wrote:
> On 2018-4-25 03:56 , Ken Cunningham wrote:
> > Waiting for the maintainer to review the ticket submission
> > someday often resulted in months of nothing happening, or years.  
> 
> The maintainer timeout was 72 hours all along, so that's not really
> relevant to a discussion about the limits of openmaintainer.

I think if you don't feel a clean version update falls in the limits
of openmaintainer (that is, just bumping the version and the
checksums), then I'm not sure what does fall under "openmaintainer"
for you.

This is not a criticism, I meant that literally! I was just noting
that (1) to many people, that seems to be what is meant (and I was
told by a number of maintainers in the last four months that they had
their port set "openmaintainer" so they wouldn't be bothered with such
questions) and (2) we need clear rules, because without them, one
just decides to do nothing on the basis that it is safer.

As I see it, there are several options here:

1) Joshua can remove "openmaintainer" from his ports because he
   doesn't feel comfortable with such changes, and we agree on what
   changes do and don't fall under "openmaintainer" (but presumably
   simple version bumps that build correctly would fall under it).

2) We can adopt a method where maintainers can put a comment in their
   Portfiles explaining what can and can't be done by other
   members of the project without explicit consultation. Then Joshua
   would put in a comment explaining what he was giving permission
   for by putting in "openmaintainer".

   Such comments could tell us all sorts of individualized policies.
   For example, a comment could say:

   # Please consult the maintainer on anything other than revision
   # bumps.

   or

   # You do not need to consult the maintainer for upgrades to this
   # port provided you run the test suite and it passes.

   If we adopt this, it would be best if such comments were placed
   near the top of the Portfile where they are unmistakable.

3) We can adopt a set of designations beyond "openmaintainer", rather
   than just having the two levels we have now ("openmaintainer" vs.
   not openmaintainer), that have more nuance. I'm not sure what the
   levels would be, but we could discuss that and come to some
   agreement which we document.

Some combination of 1, 2, and 3 could also be adopted. (I note that
there are sensitive "infrastructure ports" that might benefit from a
deviation from the 72 hour maintainer timeout as well.)

Any other options people see? Thoughts about what "openmaintainer"
should generally mean?

I'm quite serious about saying that clear rules and certainty helps.
If I know what the rules are, it's much easier to quickly handle
things that can be handled quickly and to refer things up the chain
if they need more explicit examination from experts, but it requires
that we be pretty clear on rules.

Perry
-- 
Perry E. Metzger		pmetzger at macports.org