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

Re: Withdrawal of HTML5-warnings draft

From: Sam Ruby <rubys@intertwingly.net>
Date: Tue, 18 Aug 2009 07:46:48 -0400
Message-ID: <4A8A94A8.9080901@intertwingly.net>
To: Maciej Stachowiak <mjs@apple.com>
CC: Manu Sporny <msporny@digitalbazaar.com>, HTMLWG WG <public-html@w3.org>
Maciej Stachowiak wrote:
> 
> On Aug 17, 2009, at 10:59 PM, Manu Sporny wrote:
> 
>> Noting that the current poll demonstrates that consensus would be
>> difficult to achieve[1] regarding the publishing of the HTML5-warnings
>> draft, I am withdrawing it as a candidate for publishing as a heartbeat
>> WD for HTML WG.
> 
> I think it's up to Sam whether to withdraw the poll. Here are a few 
> comments on some of your coments.

There is nothing to withdraw... it was just a poll.  But it is time to 
make a decision and move on:

http://lists.w3.org/Archives/Public/public-html-wg-announce/2009JulSep/0024.html

>> There are a number of interesting data points that came out of the poll
>> data:
>>
>> 1) The biggest problem seemed to be a lack of consistency with which
>>   warnings were provided as well as the warning language. Even though
>>   more consistent rules were applied to the second draft, the changes
>>   were not enough. The next draft that I author will apply a consistent
>>   set of rules for warnings and status markers. For example, a
>>   combination of WHAT WG status markers and all ISSUES from the HTML WG
>>   issue tracker older than X months).
> 
> I would support a draft incorporating status markers and HTML WG ISSUEs 
> older than a certain age (and even new ISSUEs that have old email 
> discussion). I believe it might be possible to add these markers in an 
> automated way. I think James showed some prototype work towards that 
> end, at least on adding the section maturity markers. Combined with 
> cleaning out stale issues, this should put us in a good place for 
> tracking our status as we approach Last Call.
> 
> I believe a draft along these lines could gain consensus.

More importantly, it would have a nice side benefit of cleaning up the 
issues list.  I was disappointed to the RDFa issue expressed for the 
first time in the context of a draft.  I was pleased to see Shelley 
raise issue-76.  I was also pleased to see the work that you (Maciej) 
have done towards cleaning out old issues.  And I want to thank Manu for 
his draft as I believe that it was an important catalyst prompting both 
of these actions.

Which reminds me, Maciej (or anybody else) care to propose a second wave 
of issues for closure for the upcoming week?

>> 2) Publishing more than one WD at a time seems to be a very unpopular
>>   way forward. It may be that publishing non-normative differences
>>   as WDs are problematic, or it may be that publishing more than
>>   one WD is seen as confusing... but the issue is not as
>>   uncontroversial as Ian, Sam and I had hoped it would be.
> 
> I was surprised by this myself. Sam has been promoting the idea of 
> multiple independent drafts, and there hasn't been a lot of opposition 
> in principle. But it seems like most of the Working Group is not on 
> board with this idea, at least in the case of such minor differences. To 
> me this says: we need to work harder on getting one draft that better 
> reflects broader consensus.

Multiple drafts never has been a goal in itself, getting to the point 
where we have a single draft that represents broader consensus is.

Borrowing your words below, I also think part of the split was due to an 
artifact in the poll construction.  Ian's response, in particular, made 
it clear that it was for him.

>> 3) The issue seemed to be divisive, with many people voting for one
>>   WD and not the other.
>> 4) Since respondents could see the poll results before the poll was
>>   over, the results may have been subject to block voting (for both
>>   drafts).
> 
> I think part of the split was due to an artifact in the poll 
> construction. In particular, for people who were happy to publish both 
> drafts, there was no way to express a preference in case only one ended 
> up being published. Since the "only one draft" option took an early 
> lead, at least some people chose to vote for only one draft to express 
> their preference.

WBS has the option of not allowing the results to be seen until the poll 
is closed.

>  - Maciej
> 
>> 5) I had not seen any opposition or support from a number of the voters
>>   that voted, prior to voting. It's concerning because they provided no
>>   feedback, have made very few posts to HTML WG (if any) over the past
>>   several months, but still voted on the spec. Both drafts received
>>   these votes, and we have no idea if the people that voted had read
>>   the thread in detail or truly understood what they were voting for.
>>   There are several examples where it is clear that the voters were not
>>   tracking the discussion on this mailing list.
>> 6) The poll was fairly close, so there is a non-trivial number of people
>>   that care about warning/status language being published with the
>>   HTML5 spec.
>>
>> The next step is to iterate on the current HTML5-warnings draft and
>> create a mechanism that is programmatic that inserts the proper warning
>> text, from both the WHAT WG status tracker and HTML WG issue tracker (as
>> an overlay) to the W3C published spec. Once the "issue consistency"
>> problem is addressed, HTML5 community feedback will be encouraged in
>> order to determine if there are more improvements that will be required
>> when publishing warning/stability text along with the HTML WG draft.

I will note that while the document links to emails and bug reports in 
the status section, neither draft contains a link to the open issues 
list itself.

>> -- manu
>>
>> [1] http://www.w3.org/2002/09/wbs/40318/wd08/results
>>
>> -- 
>> Manu Sporny (skype: msporny, twitter: manusporny)
>> President/CEO - Digital Bazaar, Inc.
>> blog: Bitmunk 3.1 Released - Browser-based P2P Commerce
>> http://blog.digitalbazaar.com/2009/06/29/browser-based-p2p-commerce/

- Sam Ruby
Received on Tuesday, 18 August 2009 11:47:32 UTC

This archive was generated by hypermail 2.3.1 : Friday, 10 October 2014 16:24:51 UTC