W3C home > Mailing lists > Public > www-rdf-logic@w3.org > July 2001

Re: DAML-S expressiveness challenge #2

From: tim finin <finin@cs.umbc.edu>
Date: Mon, 16 Jul 2001 09:12:40 -0400
Message-ID: <3B52E848.21447F13@cs.umbc.edu>
To: David Martin <martin@ai.sri.com>
CC: www-rdf-logic@w3.org, Mark Burstein <burstein@bbn.com>, narayana@ai.sri.com
David Martin wrote: 
> In a DAML-S process description, we need to be able to express the
> following sort of restriction:
>   the set of things whose property P has values in [C1, C2,...CN]
> that is,
>     the set of things x such that, for each property instance P(x, y), y
> is a daml:list whose first element is an instance of C1, second
> element an instance of C2, ..., and nth element is an instance of CN.

Does it have to be a daml:list and is N fixed?  If it doesn't have
to be a list and if we know what N is for each use, an easy solution would
be to restrict values to a new class that has a named, restricted
property for each of the list elements.  To capture the ordering that
is inherent in a list, the properties could be named 1st, 2nd,
3rd, 4th, 5th, 6th ... with the obvious interpretation. We'd also need 
to assume (probably) that N is never greater, in practice, than some 
upper limit (e.g. 20).

I guess this would be like creating special classes to represent the lists
of lengths from 1 to 20.  We could call them 1List, 2List, 3List, ... 20List.
Each of these could then be specialized, as needed, by attaching restrictions 
to the ranges of their named elements.

--
 Tim Finin, Prof Computer Science & Elect Eng, Director Inst. for Global
 Electronic Commerce, Univ of Maryland Baltimore Cty, 1000 Hilltop, Baltimore
 MD 21250. 410-455-3522 fax:-3969 finin@umbc.edu http://umbc.edu/~finin/
Received on Monday, 16 July 2001 09:06:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 08:34:47 GMT