W3C home > Mailing lists > Public > public-sws-ig@w3.org > January 2007

Re: OWL-S questions: status, Input/Output classes, and SPARQL

From: David Martin <martin@AI.SRI.COM>
Date: Thu, 18 Jan 2007 14:55:30 -0800
Message-ID: <45AFFAE2.3010500@ai.sri.com>
To: Marco Sbodio <marco.sbodio@gmail.com>
CC: public-sws-ig@w3.org

Hi Marco -

Marco Sbodio wrote:
> 
> Hi all,
> 
> I use OWL-S in some projects, and I would like to ask some questions.
> 
> 1. What is the current status of OWL-S? version 1.2 is in Pre-Release
> [1] since several months: what are the plans?

The technical work is essentially done, but there is some polishing that 
is needed before we make the release official.  Mainly polishing on 
documentation, but also a final check on the ontologies to make sure 
they pass validation tests, etc.  The main thing that has caused the 
delay is that we are significantly revamping the "walk through" so that 
it is more helpful to a novice.  Apologies for the long delay on this.

> 
> 2. In OWL-S version 1.2, Process.owl [2] defines Input and Output as
> subClassOf Parameter:
> 
> 2.1. it is asserted that Input is disjointWith Output, but not that
> Output is disjointWith Input: isn't this necessary?

No.  disjointWith is inherently symmetric; any good reasoner should 
automatically take this into account.

> 
> 2.2. I think that Parameter should be a subClassOf ProcessVar, but
> Process.owl does not state this: why? Indeed the comments say that
> Parameter is a subtype of ProcessVar, but the ontology doesn't; I
> think it's necessary, otherwise Input/Output would not have a
> parameterType (whose domain is ProcessVar).

You are right that Parameter is meant to be a subClassOf ProcessVar. 
Technically, it is not necessary to state that because it can be 
inferred from the declaration that ProcessVar is the unionOf Parameter 
and some other things.  However, there's no harm in adding the explicit 
subclass statement and I will do so.

> 
> 3. In OWL-S version 1.2, Expression.owl [3] allows the usage of
> several expression language to define conditions, among which SPARQL
> (there is in fact the class SPARQL-Expression, defined as a
> QuotedExpression whose expressionLanguage is SPARQL). I could not find
> any example of how to use SPARQL in expressing conditions within
> OWL-S. Since I'm working on this topic, I would be happy to read
> ideas/papers/examples on this topic. How do you plan to use SPARQL in
> expressing conditions (preconditions and results conditions)?

You are right; there are no examples of this, but we are planning to add 
some additional documentation about it.  In the meantime, perhaps 
someone else has an example they can contribute to this list?

Regards,
David Martin

> 
> Thank you in advance.
> 
> Best Regards
> marco
> 
> [1] http://www.ai.sri.com/daml/services/owl-s/1.2/
> [2] http://www.ai.sri.com/daml/services/owl-s/1.2/Process.owl
> [3] http://www.ai.sri.com/daml/services/owl-s/1.2/generic/Expression.owl
> 
Received on Thursday, 18 January 2007 22:53:34 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 16 March 2008 00:11:07 GMT