W3C home > Mailing lists > Public > public-sws-ig@w3.org > November 2005

Re: Options we have with respect to the draft charters (i.e., RE: [fwd] Draft charters for work on Semantics for WS)

From: David Martin <martin@AI.SRI.COM>
Date: Sat, 19 Nov 2005 16:29:15 -0800
Message-ID: <437FC35B.5050600@ai.sri.com>
To: David Martin <martin@AI.SRI.COM>
CC: "Amit Sheth @ LSDIS" <amit@cs.uga.edu>, public-sws-ig@w3.org

Hi Amit -

Please allow me to add the following to my previous message - You are 
right of course that we should focus on meeting the needs of the vendor 
and user community, and avoid counterproductive feuding, and I 
appreciate that you have emphasized this point of view.  I agree that we 
should take attribution issues offline, and I also agree with your 
emphasis on collaboration and on building useful tools.  My main point 
in recent messages has been that we could and should work together in 
this regard.  In fact, I think (something like) WSDL-S together with 
(something like) OWL-S (lite) could provide a very nice bridge between 
the Web service and Semantic Web efforts at W3C.  (And this is not to 
suggest that this would be the only path towards adding semantics to Web 
services.)

Cheers,
David


David Martin wrote:
> 
> Hi Amit -
> 
> I assume you are referring to messages posted recently by Steve, 
> Massimo, and me, which have included observations about overlapping 
> technical ideas in OWL-S and WSDL-S.  With all respect, I don't think 
> it's fair to characterize or dismiss these messages as mere academic 
> debates.  In fact, none of these messages is primarily concerned with 
> the question of attribution.  Most of them are concerned with the 
> question of whether OWL-S should be considered as an input to the 
> working group.  Surely that is a legitimate question for discussion with 
> regards to a proposed WG charter.  In these messages, the primary point 
> is simply this: since OWL-S proposed some of the same central approaches 
> as are in WSDL-S, why should it not also be considered as an input? 
> (There could of course be some qualifications as to which parts are 
> relevant to the WG.)
> 
> One recent message from me:
> http://lists.w3.org/Archives/Public/public-sws-ig/2005Nov/0008.html
> is focused on clearing up some possible points of confusion about the 
> relationship between OWL-S and WSDL-S.  My primary concern there is to 
> show that (parts of) OWL-S and WSDL-S can in fact be used together.  My 
> intention is to set the stage for fruitful collaboration in the context 
> of this WG, if it happens.
> 
> Note that the last third or so of the abstract of the WSDL-S submission 
> is devoted to an attempt to argue that WSDL-S is superior to OWL-S.  In 
> such a case, some response in a public forum may well be expected!  The 
> abstract is the most visible part of a document.  I would ask you to be 
> sensitive to the fact that the WSDL-S submission, especially if it 
> becomes an input to a W3C WG, will be visited many times by many people.
> 
> Best regards,
> David
> 
> 
> Amit Sheth @ LSDIS wrote:
> 
>> We need to decide how we want to focus our energy--
>>  
>> (a) positively to come up with recommendations that vendors and W3 
>> community would likely embrace
>>  
>> or
>>  
>> (b) in academic debates related to  who put up something in a draft 
>> document posted on the web first, who discussed the idea in a stable 
>> version first, who did in an invited talk first, or who did so in a 
>> refereed publication first.
>>  
>> My suggestion is to deal with (b) separately, preferably one-on-one 
>> first,
>> as side meetings at conferences/workshops next, and as the last 
>> resort, in writing review to submitted papers.
>>  
>> Then I can point out that in our refereed ICWS03 paper  Adding 
>> Semantics to Web Services Standards (2003) 
>> <http://lsdis.cs.uga.edu/lib/download/SVSM03-ICWS-final.pdf>,
>> we had cited v0.6 draft of OWL-S, giving it due credit, while 
>> discussing clear distinction wrt our WSDL-S
>> approach. And I can share a long list of examples of ignoring of 
>> METEOR-S's
>> contribution to the area, such as proposing functional 
>> semantics/ontology (with
>> RosettaNet use case and ontology) or non-functional semantics and
>> QoS ontology.
>>
>> I suggest we get back to (a), and whatever the outcome, one of us
>> (I can volunteer my students) make an annotated bibliography of all 
>> work on
>> either or both of the charters Carine has outlined.  An example of use 
>> of positive energy
>> is in an on-going WSMO-WSDL-S collaboration that would show how more 
>> comprehensive models
>> such as WSMO can use WSDL-S as grounding (see 
>> http://www.wsmo.org/TR/d30/v0.1/
>> or 
>> http://lsdis.cs.uga.edu/projects/meteor-s/wsdl-s/WSMX-Meteor-S-interoperability-final.pdf). 
>>
>> While talking about languages/representations/features, let us not 
>> forget the
>> prototyping/tooling/use cases (has that been done for a given idea?)
>> and all those things that make anything real to potential
>> technology adopters.
>>
>> In my analysis, the key driving factor for the two proposed charters 
>> is to limit their scope
>> such that tangible results can be reached in a reasonable period.  
>> Hence the identification of WSDL-S
>> as a starting point may be seen in this context, rather than an attempt
>> to attribute to WSDL-S all the scientific/research credit for various 
>> SWS features and capabilities,
>> many of which are shared with (and some built upon) other illustrious 
>> submissions.
>>
>> Regards,
>> Amit Sheth  (speaking personally, rather than for the WSDL-S team)
>> http://lsdis.cs.uga.edu/~amit
Received on Sunday, 20 November 2005 00:29:36 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 16 March 2008 00:11:02 GMT