W3C home > Mailing lists > Public > public-owl-wg@w3.org > October 2008

Review of OWL 2 Reference Card

From: Uli Sattler <sattler@cs.man.ac.uk>
Date: Thu, 2 Oct 2008 17:49:23 +0100
Message-Id: <665F7922-F2A5-4D8A-B870-BF3D5657FFC7@cs.man.ac.uk>
To: OWL Working Group WG <public-owl-wg@w3.org>

Hi, I had a look at the reference card and found it very confusing:

- it seems to be a structured  list of OWL (and RDFS) terms which also  
tries to describe (some aspects) of the syntax of OWL (i.e., how terms  
can be combined) and (some aspects) of the meaning of OWL. I found the  
way the second and third aspect was carried out very confusing:  (1)  
the usage of "->" is incoherent (compare the entry for  rdfs:label  
with the entry for owl:intersectionOf  (2) for a few terms, e.g.  
owl:FunctionalProperty, no 'signature' via "->" is given at all, but  
an attempt to specify its semantics...this is incoherent with the  
rest. I would suggest to stick to a single aspect that is feasible (I  
don't know, perhaps an example would be?).

- the structuring and headers are confusing: (1) what kind of  
"Language Elements" are we looking at? (2) restrictions are classes,  
but a datatype is not...so why "Classes, Datatype and  
restriction" (and all 3 would need to be plural) (3) why single out  
"individual axiom" from "assertion"?

- the grouping is confusing: whereas each of the entries following  
"owl:class" is a class expression, the entries following  
"owl:Restriction" need to be combined in the correct way in order to  
form a restriction...

minor comment:

- it seems as if "class" is sometimes (in the profiles) used for  
"class name" and sometimes for "class name or expression". this should  
be unified

Cheers, Uli
Received on Thursday, 2 October 2008 16:49:12 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 2 October 2008 16:49:12 GMT