RE: Guide comments

Bernard,

Thanks for your great comments.  I have incorporated most of them into the
next draft.  Details below.

- Mike

Michael K. Smith, Ph.D., P.E.
EDS - Austin Innovation Centre
98 San Jacinto, #500
Austin, TX  78701

* phone: +01-512-404-6683
* mailto:michael.smith@eds.com

----------------------------------------------------------------------------

> line 365 and 366, "instances of classes", "instances of properties"
> There's still some inconsistency about the use of instance vs 
> individual in this document.

DONE

> line 375, para begins "A set of OWL assertions loaded into..." --
> The sentence that begins "These assertions..." is IMO a little 
> ambiguous due to the commas, suggest rewording as:
> "These assertions may include facts
> about individuals that are members of classes, as well as various
> <em>derived</em> facts, i.e. those facts not literally present in the 
> original
> textual representation of the ontology, but <em>entailed</em>
> (logically implied) by the semantics of OWL. "

DONE with mods
 
> line 423, para begins "One advantage of OWL ontologies..."
> The "They" in the second sentence is ambiguous, suggest replacing with 
> "Tools".

DONE

> line 488, para begins "Ontology developers adopting OWL"
> First introduction of term "dialect" which contradicts earlier comment 
> that OWL is a set of _languages_.
> Also, typo(?) "predicatable" in penultimate sentence.

We have changed the terminology to 'sublanguages'.  I think I will
leave 'dialect'.  

> line 728, para begins "Importing another ontology"
> You indicate that importing brings the definitions into the "knowledge 
> base"; however, you've defined a knowledge base earlier as a container 
> for individuals, not classes.

DONE.  We have repeated problems with this distinction, since both an
ontology and a knowledge base are just collections of OWL assertions.

> line 1155, example on the use of ObjectProperty
> Can the domain and range of an ObjectProperty be modified? For example, 
> is the following valid:
> 
> <owl:ObjectProperty rdf:ID="madeFromGrape">
>    <rdfs:domain rdf:resource="#Wine"/>
>    <rdfs:range rdf:resource="#WineGrape"/>
>    <rdfs:range rdf:resource="#AntiFreeze"/>
> </owl:ObjectProperty>
> 
> <owl:ObjectProperty rdf:ID="madeFromGrape">
>    <rdfs:domain rdf:resource="#Grappa"/>
> </owl:ObjectProperty>

Yes.  Though your example would need to be modified to 

<owl:ObjectProperty rdf:about="madeFromGrape">
   <rdfs:domain rdf:resource="#Grappa"/>
</owl:ObjectProperty>

And this may not be what you intend?  It says that madeFromGrape is a
mapping from the INTERSECTION of Wine and Grappa to the INTERSECTION
of WineGrape and AntiFreeze.

> I.e. two questions: what's the cardinality of rdfs:range (and 
> rdfs:domain), and if it's more than 1, can the range of a property be 
> extended in another ontology?
>
> line 1261, Vintage class example
> Why is this class defined to be a subclass of #Thing? (Whereas none of 
> the other classes in this document make that definition, even though 
> it's assumed.)

No reason.  I think I was cutting and pasting from something sent by
one of my co-editors or another member of the working group.

> line 1307, example vintage years
> Why is xsd:decimal used, rather than, say, integer, or positiveInteger?

There were unresolved issues within the WG related to datatypes and I
was not very clear on how they would ultimately work out.  At this
point you are absolutely correct, integer is best.  Since the datatype
is associated with years, we should probably allow for -1000 (1000
BC).  CHANGED to integer.

> line 1342, para begins "The <tt>yearValue</tt> property"
> I don't think you mean _decimal_ in "The <tt>yearValue</tt> property 
> relates <tt>WineYear</tt>s to decimal numbers
> greater than or equal to 1700."

I did until you noted that integer would be better.

> cheers
> 
> Bernard
> 
> Bernard Horan
> Senior Staff Engineer, SunLabs Europe
> Sun Microsystems Ltd, M/S ECBG01
> Sun House, 306 Cambridge Science Park
> Milton Road, Cambridge, CB4 0WG, UK
> +44 (0)1223 418925
> +44 (0)1223 420058 fax
> bernard.horan@sun.com

-----Original Message-----
From: Bernard Horan [mailto:Bernard.Horan@Sun.COM]
Sent: Friday, December 06, 2002 7:10 AM
To: michael.smith@eds.com; dlm@ksl.stanford.edu; volz@fzi.de;
welty@us.ibm.com
Cc: www-webont-wg@w3.org
Subject: Guide comments


Guide editors,

Just wanted to express my appreciation for a great job on the guide.

Here's some comments and requests for claficiation on the latest 
version (http://www.w3.org/2001/sw/WebOnt/guide-src/Guide.html).

As there's no page or para numbers I'll use the line numbers from the 
HTML source:

line 365 and 366, "instances of classes", "instances of properties"
There's still some inconsistency about the use of instance vs 
individual in this document.

line 375, para begins "A set of OWL assertions loaded into..." --
The sentence that begins "These assertions..." is IMO a little 
ambiguous due to the commas, suggest rewording as:
"These assertions may include facts
about individuals that are members of classes, as well as various
<em>derived</em> facts, i.e. those facts not literally present in the 
original
textual representation of the ontology, but <em>entailed</em>
(logically implied) by the semantics of OWL. "

line 423, para begins "One advantage of OWL ontologies..."
The "They" in the second sentence is ambiguous, suggest replacing with 
"Tools".

line 488, para begins "Ontology developers adopting OWL"
First introduction of term "dialect" which contradicts earlier comment 
that OWL is a set of _languages_.
Also, typo(?) "predicatable" in penultimate sentence.

line 728, para begins "Importing another ontology"
You indicate that importing brings the definitions into the "knowledge 
base"; however, you've defined a knowledge base earlier as a container 
for individuals, not classes.

line 1155, example on the use of ObjectProperty
Can the domain and range of an ObjectProperty be modified? For example, 
is the following valid:

<owl:ObjectProperty rdf:ID="madeFromGrape">
   <rdfs:domain rdf:resource="#Wine"/>
   <rdfs:range rdf:resource="#WineGrape"/>
   <rdfs:range rdf:resource="#AntiFreeze"/>
</owl:ObjectProperty>

<owl:ObjectProperty rdf:ID="madeFromGrape">
   <rdfs:domain rdf:resource="#Grappa"/>
</owl:ObjectProperty>

I.e. two questions: what's the cardinality of rdfs:range (and 
rdfs:domain), and if it's more than 1, can the range of a property be 
extended in another ontology?

line 1261, Vintage class example
Why is this class defined to be a subclass of #Thing? (Whereas none of 
the other classes in this document make that definition, even though 
it's assumed.)

line 1307, example vintage years
Why is xsd:decimal used, rather than, say, integer, or positiveInteger?

line 1342, para begins "The <tt>yearValue</tt> property"
I don't think you mean _decimal_ in "The <tt>yearValue</tt> property 
relates <tt>WineYear</tt>s to decimal numbers
greater than or equal to 1700."

cheers

Bernard

Bernard Horan
Senior Staff Engineer, SunLabs Europe
Sun Microsystems Ltd, M/S ECBG01
Sun House, 306 Cambridge Science Park
Milton Road, Cambridge, CB4 0WG, UK
+44 (0)1223 418925
+44 (0)1223 420058 fax
bernard.horan@sun.com

Received on Friday, 6 December 2002 09:23:08 UTC