CFV: New OpenJDK Members Group Member: Simon Tooke

[Andrew Hughes]

On 09/06/2020 12:18, Kim Barrett wrote:

snip...

>> My suggestion would be that, if we are considering Reviewer status as a
>> requirement for Member status, then the latter could be awarded when a
>> successful vote takes place for the former. The registrar has to update
>> the person to Reviewer status anyway. It would be quite simple to add
>> them to the Members group at this time and forego the additional vote
>> for many cases.
> 
> I don't think anyone has proposed that Reviewer status is a
> requirement for becoming a Member. I also think Reviewer status need
> not be a sufficient criteria. Reviewer status provides a partial
> indication of the level of involvement and contribution that is
> considered necessary to be granted Member status. I think there are
> other ways to make "significant contributions" that don't lead to
> Reviewer status.

I don't disagree with any of that, but I guess you can also see why this
inevitably fuzzy criteria also makes it unclear when it is appropriate
to propose someone for membership of the OpenJDK members group.

> 
> And the two should not be conflated. Groups have members. Projects
> have Reviewers. Those organizational categories and their associated
> roles are distinct according to the Bylaws.  (I can't comment on the
> rationale behind that separation; I'm a relative newcomer and don't
> have the relevant historical information.)
> 

I've been here since the start of OpenJDK and it's no clearer to me. I
suspect it has its roots in whatever the setup was at Sun back then.

> Even if there was an agreement to link them that way, there is the
> question of which Group(s) the newly minted Project Reviewer should be
> added to.

I was thinking it would make one also a member of the OpenJDK Members
group, but that others could still be proposed for membership as they
are now (which I think deals with some of your concerns above).

In other words, the additional voting process to become a member of the
OpenJDK Members group is unnecessary for anyone who has successfully
been voted in as a reviewer. It's basically an optimisation to avoid two
voting rounds (the second often being forgotten) for someone the text
you referred to suggests would be a candidate for membership of the
OpenJDK members group. Votes for such membership would then only be
necessary for candidates who we feel should have the status, but aren't
active as reviewers.

As it stands, becoming a reviewer does not grant any group membership.

> 
>> It's not the onerous nature that was the issue so as much as the bizarre
>> situation of having Reviewer status, but not Member status by virtue of
>> not being in a group.
> 
> It's only bizarre if one considers the organizational structure
> described by the Bylaws, including the distinction between Groups and
> Projects, to be problematic.

Not so much problematic as unclear.

My reading of the bylaws suggests group membership in general is not
required in most circumstances.  This is probably why we have quite a
few people who are not members of any group.

The OpenJDK Members group is the exception to this, as it exists solely
to ascribe voting rights to a more general group not bound by a
particular project.

> 
>> Equally, there were people given Member status without even authorship,
>> never mind Reviewer status.
> 
> The Bylaws describe the process for initial creation of a Group,
> including the bootstrapping of the initial Members list by the Group's
> Lead. Presumably the Lead selects people who seem likely to be
> significant contributors, since the Group didn't previously exist to
> contribute toward.
> 
> 

Right, that's the case for groups in general.

What I'm referring to is the creation of the Members group at the time
the bylaws were put in place (the transition described in section B).
Anyone who was a member of a pre-bylaws group was made a member of the
new post-bylaws OpenJDK Members group.
-- 
Andrew :)

Senior Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

PGP Key: ed25519/0xCFDA0F9B35964222 (hkp://keys.gnupg.net)
Fingerprint = 5132 579D D154 0ED2 3E04  C5A0 CFDA 0F9B 3596 4222