On May 6, 2007, at 6:03 AM, Dan Connolly wrote:
Maciej Stachowiak wrote:
Hello HTML WG Chairs,
I think the decision process for our first formal resolution has
not turned out so well. Counting every "no" vote as a Formal
Objection means that to register any kind of dissent, one must
automatically lie across the metaphorical railroad tracks,
requiring either work stoppage or escalation of the dispute. I
think this creates two problems:
1) People who disagree with the majority feel pressured not to
vote "no", since it would potentially impede the progress of the
group. In a straw poll, people should feel free to vote their
conscience without worrying they may put the work of the group at
risk.
This isn't a straw poll.
People who just want to express disagreement
without formally objecting are supposed to just express that
disagreement in email or in the rationale field and abstain.
Yes, this is what I think is a problem. If someone dislikes the plan,
but not at a "lie across the railroad tracks" level, then they can't
say "no", they have to say "abstain", which I think is unfair.
2) People who do want to register their disagreement automatically
escalate it to the most serious level just by virtue of their "no"
vote. For example, I don't think the voter who said he liked the
name "HTML 5.01" better than "HTML 5" necessarily would have
chosen to escalate it to the Director after the vote.
True, he probably didn't understand what I meant the "no" option
to mean when he chose it. That makes for a little extra "paperwork,"
but it doesn't seriously impact things.
Well, there's people in between where it's hard for me to tell if
they would have registered an FO as a follow-on to a "no" vote. In
other working groups I've been in, dissenters in a voice vote after
long discussion rarely chose to turn their dissent into a Formal
Objection. Would the person who proposed that the new standard be
named not "HTML 5" but "Canonical HTML" have chosen to escalate the
issue after a vote? He seems to feel strongly, but it's hard to tell.
I'd say the same may even apply to Terje's no votes, even though he
is fully aware that they constitute a Formal Objection.
In many W3C Working Groups, resolutions often carry just by
majority vote, with dissentors given the opportunity to lodge a
Formal Objection afterwards if they still strongly disagree.
Often the Formal Objection
is only done at LC time so there is plenty time to work things out
before the FO has to be recorded for the Director. I suggest the
chairs adopt this process for future WG decisions.
I expect to give participants various chances to withdraw their
objections.
OK, but that's hardly the same thing as giving them the chance to not
register Formal Objections in the first place while still expressing
disagreement in a clear way.
Furthermore, ordinary comments on a Working Draft or Last Call
draft are not normally automatically considered as Formal
Objections, even if the resolution is "Disagree". Generally the
disputant has to explicitly say it is a Formal Objection.
I propose that for future resolutions, "no" votes be treated only
as ordinary statements of disagreement. Only if the dissentor
explicitly chooses to escalate his dissent to a Formal Objection
should it be treated as such.
WBS doesn't currently support that, as far as I know. Perhaps
I'll look into getting it enhanced.
I'm not sure what WBS has to do with it. Couldn't you just not count
the "no" votes as Formal Objections at all, and strong dissentors get
a chance to lodge a Formal Objection afterwards "by hand", just as a
commentor at Last Call would? I asked participants in other working
groups and they generally reported that telecon or f2f decisions
allowed disagreement without it necessarily counting as an FO.
This way, Formal Objections will be the rare special appeals they
were meant to be and not commonplace features of the decision
process.
Regards,
Maciej
--
Dan Connolly, W3C http://www.w3.org/People/Connolly/
