W3C home > Mailing lists > Public > public-owl-wg@w3.org > October 2007

(resolution status/documents)Re: minutes for 17 October

From: Jim Hendler <hendler@cs.rpi.edu>
Date: Fri, 19 Oct 2007 12:47:37 -0400
Message-Id: <73B8B6C8-92A4-4026-AB54-AD863EC18818@cs.rpi.edu>
Cc: "Peter F. Patel-Schneider" <pfps@research.bell-labs.com>, team-owl-chairs@w3.org, public-owl-wg@w3.org
To: Bijan Parsia <bparsia@cs.man.ac.uk>

(This note has two parts - the first is a pedantic and boring  
discussion of WG and W3C policy - I suggest all who don't care about  
such minutia skip it -Sandro, you probably need to read it.   The  
second is more germane to the WG, as it makes clearer what the issue is)

<:pedantic>
Bijan - the dispute seems to be whether "publication schedule" is a  
specific enough agendum item for a discussion of moving specific  
publications to WD by a specific time.  I will be happy to say  
"charter violation" instead of "process violation" (in fact, the  
process Issue I would point out is to follow the charter).  I claim  
that the agenda did NOT provide enough warning of a specific nature  
on something that members of the WG had not discussed first in email  
or anywhere else.  If there had been a specific agenda item posing  
the resolution, I would have had time to state my issues before,  
rather than after, the telecon - which is necessary so that those of  
us who cannot make the telecons can be allowed to participate.\
   I'm sorry that you and i disagree on this, but it is early in the  
WG and it is an important time to make sure we are sticking to  
process - it is much easier in the W3C to avoid issues early on then  
to fix them later.
  Further, we should probably get a chair/W3C rep resolution on  
something - if an agenda is emailed out, and then it is posted to the  
wiki and changed, I think only the agenda AS MAILED should count -  
because many of us can access email when traveling, but may have more  
trouble getting to a wiki - further, the email has a verifiable date  
that is easily ascertained, where the Wiki takes detective work to  
make sure we can tell whether someone changed something substantive  
within the 24 hour window
</:pedantic>

Anyway, I'm not opposed to a decision that we consider moving the  
documents forward - but the way the resolution is written it says we  
WILL move THESE three documents forward BY A CERTAIN DATE - I would  
like to believe we will work on understanding these documents, making  
sure we have consensus on the key issues, and publishing as ready.  I  
do not believe we have had anywhere near the amount of discussion  
necessary to decide at this point whether we are ready or not.  The  
current documents represent the consensus reached in an informal  
process outside the W3C.  Our WDs represent the consensus within the  
WG.  I do not believe these are currently the same thing, and would  
like us to do due diligence so that we end up producing the best new  
version of OWL while minimizing impact on current implementations  
(across the Semantic Web community, not just in the academic SW  
groups).  This doesn't mean I don't think the features in the  
documents are useful, or that I oppose them, frankly I just feel I  
haven't had time to consider them fully - I raised a large number of  
issues with early drafts (in public) and I need a chance to go back  
and evaluate where we stand.  So none of this is meant to be  
negative, rather, I think we need to be more deliberative - which is  
very frustrating to those who've been working on these for a long  
time, but remember the rest of us are just getting up to speed, and  
you need to give us time to reach appropriate levels of understanding
  -JH



On Oct 19, 2007, at 11:56 AM, Bijan Parsia wrote:

> On 19 Oct 2007, at 14:28, Jim Hendler wrote:
>
>> i see the following in the log:
>> RESOLVED: Our first working drafts, to be published before the 3- 
>> month heartbeat, will be one: (1) Structural Specification, (2)  
>> Semantics. We may include (3) RDF Mapping in this list. These are  
>> based on the text for each of these at http://www.webont.org/owl/1.1/
>>
>> we should be clear - this resolution is not an actual resolution  
>> to publish
>
> It *is* an actual resolution to publish and was understood that way  
> by everyone on the call, afaict.
>
>> as such a thing
>> would be a process violation at this time
>
> It would be helpful when raising process issues to point to the  
> part of the process document that substantiated your claim.
>
> It's quite common to make decisions...including decisions to  
> publish...at a meeting where some people didn't attend. In some  
> groups, e.g., the HTML WG, whose size is too big to expect everyone  
> to participate in telecons, they have --- and perhaps this is in  
> their charter --- agreed to always have asynch decision making  
> (i.e., by web survey). That's not in our charter.
>
>> - i can point to the charter issues,
>
> The charter reads:
>
> """When deciding a substantive technical issue, the Chair may put a  
> question before the group. The Chair must only do so during a group  
> meeting for which the agenda indicated the possibility that a  
> decision on that particular issue might be made. """
>
> I'll concede that this is a technical issue. So then we look at the  
> agenda, which is a bit complicated because, qua wiki page, it evolved.
>
> The mailing list reminder, which, I believe, is informative (wiki  
> page is canonical):
> 	http://lists.w3.org/Archives/Public/public-owl-wg/2007Oct/0065.html
>
> Seems to be in time for the meeting (24 hours before).
>
> It contains:
>
> Deliverables (15 min)
>    Wiki authorship policy
>    Publication schedule (Bijan)
>
> Which, I believe, sets up a reasonable expectation that some  
> decision could be made about the publication schedule. To clarify  
> this point, the Wiki agenda was evolved to make this fact more  
> salient:
>
> 	http://www.w3.org/2007/OWL/wiki/index.php?title=Teleconference. 
> 2007.10.17/Agenda&oldid=383
>
> This was done at 08:54, 17 October 2007, so the day of. I think the  
> original item indicated the possibility that there would be a  
> decision and the subsequent clarification firmly clarified it.
>
>> but suffice to say an actual decision to publish has to be  
>> announced in advance,
>
> This is garbled. What announced in advance of *what*? You mean that  
> it is announced before a meeting where a decision might be made to  
> publish that that decision might be made at that meeting? That's a  
> reasonable *request*, but I don't think it's a process issue. I  
> cannot find anything to that effect in the charter. You can, of  
> course, delegate a proxy:
> 	<http://www.w3.org/2005/10/Process-20051014/groups.html#group- 
> participation>
>
>> there's some other issues as well (usually the decision to publish  
>> is via a formal vote so as to meet all these needs)
>
> That's not my experience and I do think it's at all necessary.
>
>> however, a resolution saying we are considering a proposal to do  
>> this (that some people still object to) is fine - so that's what i  
>> assume this is - right?
>
> Nope.
>
> There were no objections to the resolution that we shall publish:
> 	Structural Specification
> 	Formal Semantics
> 	And optionally RDF Mapping
>
> By the heartbeat deadline (3 months or so from the first telecon, I  
> believe).
>
> Neither was there a request for a vote. The resolution went through  
> several iterations to meet concerns of members attending the  
> telecon. There was consensus.
>
> What was not resolved was whether we'd publish these documents  
> substantively earlier, e.g., next week, or by the first F2F.
>
> Cheers,
> Bijan.

"If we knew what we were doing, it wouldn't be called research, would  
it?." - Albert Einstein

Prof James Hendler				http://www.cs.rpi.edu/~hendler
Tetherless World Constellation Chair
Computer Science Dept
Rensselaer Polytechnic Institute, Troy NY 12180
Received on Friday, 19 October 2007 16:48:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:13:26 GMT