W3C home > Mailing lists > Public > public-sws-ig@w3.org > November 2004

Re: OWL-S version 1.1 now available

From: Daniel Elenius <daele@ida.liu.se>
Date: Sun, 21 Nov 2004 14:55:45 -0800
Message-ID: <41A11CF1.4090104@ida.liu.se>
To: Ian Dickinson <ian.dickinson@hp.com>, public-sws-ig@w3.org

Ian Dickinson wrote:

> Daniel Elenius wrote:
>
>> Of course, but that happens all over the place in OWL-S. An OWL-S 
>> document is not much use without an OWL-S-aware processor. Looking 
>> only at the RDF level, there is also no difference between Split, 
>> Split+Join, Choice, and Any-Order, except their URIs. So, using the 
>> same reasoning, would you want to turn them all into one class?
>
> I didn't say turn them into one class.  I suggested, and it's *just a 
> suggestion* that one could be made a sub-class of the other, wlog.
>
>> So, how would you do it? 
>
> Well, in a sense that's not up to me. I'm just giving you some 
> feedback from someone not involved in the day-to-day design and 
> development of the standard.  The OWL-S committee has range of 
> possible, valid, responses to such feedback, a range which includes, 
> among others, ignoring it.
>
> Fwiw, I guess I was thinking that cleanly separating utility 
> encodings, like "this is a closed bag", "this is a closed list" into 
> one package, and then re-using those in the main profile would 
> potentially make life easier for developers (particularly OWL-S tool 
> developers).
>
> For example (again *just a suggestion*), using N3 for compactness and 
> assuming a namespace "ds" for owl-s data-structures:
>
> ds:List a owl:Class .
> ds:first a own:ObjectProperty
>          ; rdfs:domain ds:List .
> ds:rest a owl:ObjectProperty
>         ; rdfs:domain ds:List
>         ; rdfs:range ds:List .
> ds:nil a ds:List .
>
> ds:ClosedBag a owl:Class
>              ; rdfs:subClassOf ds:List .
>
And also

ds:ClosedList a owl:Class
             ; rdfs:subClassOf ds:List .

right?

> service:ConrolConstructs

Do you mean service:ControlConstruct (i.e. the same as in the 
allValuesFrom restriction below)? Or service:ControlConstructs (plural, 
i.e. something different)? In any case, I guess it should be the process 
prefix, but that's a minor point.

>     a owl:Class
>     ; rdfs:subClassOf ds:ClosedBag
>     ; rdfs:subClassOf
>          [ a owl:Restriction
>            ; owl:onProperty ds:first
>            ; owl:allValuesFrom service:ControlConstruct
>          ]
>     .
>
I don't quite see how you want to use this class. Is it a superclass of 
all the control construct classes? Or does it replace the 
ControlConstructBag and ControlConstructList classes as the range of the 
process:components property? If so, it still looks like you would need 
two different classes to separate between the ordered and unordered 
cases, right?

> This is, to be fair, almost the same as the current proposal. The only 
> difference, on reflection, is that "Bag" or "List" moves out of the 
> URI of the class and into an rdf:type declaration.
>
> Ian
Received on Saturday, 20 November 2004 22:55:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 16 March 2008 00:10:58 GMT