W3C home > Mailing lists > Public > public-html@w3.org > January 2009

Re: Moratorium on the spec-splitting discussion

From: Maciej Stachowiak <mjs@apple.com>
Date: Wed, 28 Jan 2009 16:28:09 -0800
Cc: public-html@w3.org
Message-id: <97799637-60B5-49F3-953C-CD44478F8F9A@apple.com>
To: Sam Ruby <rubys@intertwingly.net>


On Jan 28, 2009, at 8:24 AM, Sam Ruby wrote:

> Maciej Stachowiak wrote:
> [please forgive the excessive snippage]
>> I'm asking about "group decision", not necessarily consensus. Per  
>> the W3C Process, when consensus cannot be achieved, a group  
>> decision should nontheless be made by other means, such as voting.
>
> I've made a relatively concrete proposal as to how we should  
> evaluate requests for advancement to FPWD.  My current preference is  
> that we tweak that if necessary and then execute on it.
>
> Meanwhile, you've suggested an alternative, namely voting.  I'd like  
> to see that alternative fleshed out before deciding to go that way.   
> I see a number of issues that need to be worked.
>
> 1) Votes tend to be polarizing.  I've base this on my experience  
> with a number of different standards bodies.  The IETF even has a  
> rather amusing way of solving this via humming.  Unfortunately it is  
> not immediately obvious how such an approach could be applied here,  
> but it would be a fun bikeshed project if anybody wishes to pursue  
> it. ;-)

Votes can be polarizing, that is true. But I believe they can also  
provide a degree of clarity that lets the side that lost the vote move  
on. For example, the vote to adopt Web Apps 1.0 as the initial review  
version of HTML5 was polarizing, but I believe that in the end it was  
for the best and actually reduced the heatedness of the discussion,  
relative to the period before we had a clear decision. My expectation  
would be somewhat greater polarization leading up to the vote,  
followed by the ability to move on.

> 2) Results of votes can be engineered based on how the question is  
> phrased.  Do you favor "curtailing creativity" or "creating  
> confusion"?  Um, neither, thank you.  But I assure you that no  
> matter how well intentioned you try to create the poll, somebody  
> will object because they perceive the question itself as being  
> biased along one of those two lines.

That is true. But a process that involves not having a vote can also  
be engineered. For example, your proposed process is by design  
strongly biased towards publishing. While perfect neutrality may not  
be achievable, we should do the best we can.

> 3) Votes, by themselves, aren't prescriptive enough.  The vote was  
> 47-53.  Now what?  This is the weakest issue I have, as it can  
> straightforwardly addressed by requiring a justification.

If the level of opposition is high enough that the vote is even close,  
then perhaps it is wiser to make changes to the proposal to achieve  
wider agreement. This could even be formalized by having a "yes, but  
only with some changes" option. FPWD doesn't require consensus, but if  
voting uncovers changes that could bring it much closer, then I say  
that is a great side effect of the voting process.

> But it still leaves a few loose ends: when to hold a vote?  Only if  
> you are sure of the results?  Early and often?  Can somebody just  
> keep retrying until a vote passes?

I would say:

1) Anyone can propose that a draft be published as FPWD whenever they  
feel it is ready for that step, and ask for a vote.
2) If the chairs judge that the proposal is in good faith and has some  
level of support, they schedule a vote, ensuring that there has been  
enough time for discussion.
3) The chairs use their best judgment to prevent abuse such as  
repeatedly calling for a vote on the same text.

> That should be enough to get started.  If you or anybody else would  
> be willing to work to produce a concrete proposal that addresses  
> some or all of these issues, I would very much like to discuss the  
> outcome of such.

We've held votes before, and while there were issues, I don't think we  
need to fear for the workability of a voting based process.

I would add that I expect that Mike's document would probably pass an  
FPWD vote with some minor changes (possibly just a disclaimer  
indicating that its normativity is an open issue). I'm not sure if  
that is my preferred outcome, but I would accept it and move on. As it  
is, I feel like the group is being strong-armed into publishing  
without the ability to voice objections and be heard.

Regards,
Maciej
Received on Thursday, 29 January 2009 00:28:50 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:39:00 UTC