Re: HTML Working Group Decision Policy - for discussion

On Oct 8, 2009, at 5:17 AM, Philippe Le Hegaret wrote:

> On Wed, 2009-10-07 at 17:43 -0700, Maciej Stachowiak wrote:
>> On Oct 7, 2009, at 12:32 PM, Philippe Le Hegaret wrote:
>>
>>>
>>> [[
>>> The group SHOULD set a time limit for acknowledgment by a reviewer  
>>> of
>>> the group's substantive response; a reviewer cannot block a group's
>>> progress. It is common for a reviewer to require a week or more to
>>> acknowledge and comment on a substantive response. The group's
>>> responsibility to respond to reviewers does not end once a  
>>> reasonable
>>> amount of time has elapsed. However, reviewers SHOULD realize that
>>> their
>>> comments will carry less weight if not sent to the group in a timely
>>> manner.
>>> ]]
>>>
>>> This is left undefined in the current draft. Step 2 doesn't have the
>>> boilerplate statement that will advise the commenter. The  
>>> expectation
>>> here would be to have a timeout (2 weeks is acceptable in general).
>>
>> I've also had feedback from several people outside the mailing list
>> that there should be a time limit after which we assume we're not
>> going to get a response. Does everyone agree that this is reasonable?
>> If so, I'll add a 2 week waiting period (or longer, if people want)  
>> to
>> the policy.
>
> Sounds good to me. An other related external feedback that I've got
> regarding time limits is the duration between the responses sent  
> back to
> commenters and the transition request out of LC. It might be good to
> indicate that we'll leave two weeks in between as well (unless we get
> positive responses from all those commenters).

How about minimum of two weeks from the time we get every bug into a  
final state? That would mean at least two weeks for commenters to see  
their replies.

Note: Given the size, scope and importance of the spec, I expect our  
first Last Call for HTML5 will result in some significant changes, so  
we'll probably have to go back to Working Draft and have at least a  
second Last Call before finally transition to CR. Smaller specs may be  
able to get away with a single Last Call.

Regards,
Maciej

Received on Thursday, 8 October 2009 12:31:42 UTC