[sakai2-tcc] Override Votes

[John A. Lewis]

I also have some concern about the Override Votes, mostly in that I
think we may be doing them too quickly, and therefore too often. (I also
think some folks may be voting -1 on things a bit too quickly and
strenuously, which may also be causing the large number of Override
Votes, but I'm getting ahead of myself.)

In other similar groups that I've worked with using Apache-style voting
roles, the goal has been to work toward getting a -1 voter to change
their vote to at least a 0. The main way to do this has either been to
address their specific concerns (e.g. change the code so that it
addresses a specific objection that caused their -1 vote) or to convince
them -- hopefully with logic :) -- that their concern is
incorrect/unfounded/exaggerated/etc. These would sometimes turn into
multi-week discussions with multiple revisions of the thing under
consideration, all in order to get the entire group to a point of
consensus where the -1 votes would all turn into at least 0 votes, if
not into +1 votes.

We should regard the need for an Override Vote as a failure of the
group. One way or the other, we should be working the issue at hand so
that we can get everyone onto the same page -- either eliminating all
the -1 votes or getting a large number of -1 votes. Going to an Override
Vote and having it pass means we are telling the dissenters that they
are being unreasonable and we are not interested in their objections to
the issue anymore. Doing this once or twice a year might be fine if
we've truly exhausted the attempt at consensus -- doing this repeatedly
will mean we have a dysfunctional meritocracy and need to revisit the
makeup of the voting members.

I am as guilty as anyone (and probably more than most) at bowing to
expediency in these matters. In the case of the hybrid-mode code
changes, I scanned the objections that JF documented so well, I scanned
Lance's response to them, and it looked to me like there was good faith
work going on to address them. So I was willing to vote +1 for the code,
and I was willing to vote YES on the Override. But if someone else, who
is paying way more attention to the details than I am, thinks there are
still some issues to be addressed, then we ought to keep working the
issue until their concerns are assuaged. In hindsight, I think I should
have noted NO on this Override, and perhaps on the others as well. Not
because I don't want the changes (I very much want Sakai 2 to be
completely supportive of Hybrid mode), but because I should support the
right of someone else (who is watching the details more closely than I
am) to block the issue with their -1 until they are able to get on board.

This does mean that voting -1 is a powerful statement, and (as Uncle Ben
says) with great power comes great responsibility. We do need to make
sure we are only using a -1 vote for things we genuinely find
objectionable. If the rest of the group seems supportive of something
and one finds that thing merely distasteful, the correct vote is 0 with
appropriate concerns stated (which the group should still take to heart
and try to address).

All that said, we are a new group and these are the learning experiences
that we have to go through. So ultimately, I think this is healthy as
long as we continue to self-evaluate and grow and adapt to provide the
best stewardship of this project that we can.

What do others think?


On 11/15/2010 01:07 PM, Aaron Zeckoski wrote:
> Does no one else here see a problem with the fact that 2 of the people
> most involved with the core of the Sakai 2 codebase voted against this
> code change and yet it still went in?
>
> It seems like the override thing is being used far too much for my
> comfort. Whatever happened to actually addressing concerns? Instead we
> are just ignoring them and proceeding like they don't matter. I mean,
> if I am just going to be overridden whenever I express concerns then I
> am thinking my time is better spent elsewhere.