Re: minutes: HTML WG Weekly 21 May 2009 [draft]

On May 24, 2009, at 8:06 AM, Sam Ruby wrote:

> Maciej Stachowiak wrote:
>> On May 24, 2009, at 3:50 AM, Sam Ruby wrote:
>>>
>>> I agree that it was a lopsided vote.
>>>
>>> I agree that there was an intent to publish as a Note, but  
>>> disagree with any implication that it constituted a commitment or  
>>> a decision to ultimately publish as a Note, in particular I  
>>> disagree that it was a decision that would need to be reversed.
>>>
>>> I disagree that the conference call is "informal", but I agree  
>>> that further discussion is warranted.
>> Let me put it this way. I think if we want to make a decision as a  
>> WG not to publish any further Working Drafts, and not to aim to  
>> publish as a Note, I think that decision should be taken as  
>> seriously as the decision to publish in the first place.
>> I think discussion on a single conference call, where abandoning  
>> the Design Principles document was not even an agenda item (though  
>> other Design Principles discussion was), and when there had been no  
>> mailing list discussion of doing so, does not constitute an  
>> adequate process for assessing consensus.
>
> No assertion was made that consensus was determined on that  
> conference call.  If you want to take exception to what actually was  
> said or done, feel free to do so.

I don't object to what was said, I objected to Chaals saying "the  
group has no plan to take this any further" and therefore I shouldn't  
do anything, and then Laura agreeing with him. I don't think the group  
has made such a decision.

>
>> After writing the above, I checked what the W3C Process had to say  
>> about stopping work on a document. I was surprised to learn that,  
>> apparently, the  W3C Process does not allow stopping work without  
>> publishing either as a Recommendation or a Working Group Note: <http://www.w3.org/2005/10/Process-20051014/tr.html#tr-end 
>> >. These are the only allowed end states, and the proper way to  
>> abandon a document is to publish it as a Note. This means that any  
>> time we agree to publish a Working Draft, we are committing to at  
>> least publishing a WG Note at some point, though of course  
>> publishing as a Note does not imply any endorsement.
>
> Fair enough.  I once thought there was the possibility that a few  
> small changes might improve the chances that consensus might form.   
> If that isn't the case, then I would suggest that those who have  
> opposing views be presented with the opportunity to prepare brief,  
> factual statements about the areas of disagreements.  The intent  
> would be that such statements would be included in the front matter  
> of the Note.

I don't know if it's the case or not that small changes could bring  
about consensus. Perhaps anyone who still has problems with the  
document could make their case for changes.

 From my point of view, the survey revealed that feelings on the  
Design Principles were like this: over 90% of the group agreed with  
every principle largely as-is; and on the other hand, a few people  
found something to object to in almost every principle. That was two  
years ago, perhaps results would be different if we took a new survey.  
But if the sense of the group is similar, then indeed increasing the  
degree of consensus may be hard, since the objectors may not be  
satisfied with small changes, and large changes (like removing most of  
the principles) would be unreasonable in light of the widespread  
general support.

That being said, I don't think the correct response to > 90% support  
with some persistent objectors is to stop work and publish the  
document with a disclaimer. Per the Process, our responsibility is to  
hear out objections and report any that go unaddressed, but not  
necessarily to record them in our work output itself. And if we cannot  
find a way to satisfy some objectors, then we are certainly not  
required or even encouraged by the Process to stop. Otherwise we are  
holding ourselves to a 100% agreement standard, which is not what the  
Process means by consensus. Indeed, I would be inclined to say that  
 >90% agreement with good faith efforts to address serious objections  
would be consensus, even if some objectors were still unhappy.

Regards,
Maciej

Received on Sunday, 24 May 2009 22:40:44 UTC