W3C home > Mailing lists > Public > www-ws@w3.org > August 2002

Re: DAML-S subclassing, restrictions

From: Sheila McIlraith <sam@KSL.Stanford.EDU>
Date: Sat, 17 Aug 2002 08:32:33 -0700 (PDT)
Message-Id: <200208171532.IAA05767@poco.Stanford.EDU>
To: edward.buckley@bt.com, www-ws@w3.org


> Hello,
>
> Am I correct in thinking that there is nothing to stop people subclassing
> properties.

You are correct.  Defining a subproperty of a property to specialize it
is exactly the intention.  Subclassing and subpropertying allows you
to exploit the richness of the language to define and reason about
taxonomies of concepts.

> For example, with the degreeOfQuality Profile attribute it is expected that
> subclassing will occur (due to the documentation explaining it). However
> with e.g. geographicRadius it isn't expected but is it forbidden?

No it is not forbidden.

> If this is the case, will something similar be introduced for DAML-S,
> similar to the 'final' keywork in Java?

> Also, for the attributes that are commented saying that "...descriptions
> will specialiase this..." (e.g. communicationThru), do service descriptions
> HAVE to specialise or can they just use the property as it is?

Can just use the property as it is.

> Also on this point (!), do the range restrictions apply to specialised
> properties/classes? So for communicationThru, could you have multiple
> specialisations even if the maxCardinality was 1?

Correct.

Sheila McIlraith
Knowledge Systems Lab
Stanford University
Received on Saturday, 17 August 2002 11:32:38 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2007 12:25:40 GMT