W3C home > Mailing lists > Public > public-webont-comments@w3.org > October 2003

RE: Minor comments on 18 Aug 2003 OWL Guide [attempt 2]

From: Smith, Michael K <michael.smith@eds.com>
Date: Tue, 21 Oct 2003 13:03:39 -0500
Message-ID: <2928D6FE55D5584390909578CAC8397D03AD0027@usplm216.txpln.us.eds.com>
To: "Ian B. Jacobs" <ij@w3.org>, public-webont-comments@w3.org

Ian,

Thank you for your suggestions.  Please reply to the mailing list 
as to whether the proposed changes below adequately address your 
comments with respect to the Guide.

- Mike

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

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

> 1) Subsections of 3.4 are numbered; subsections of 3.3
>    are not. I think those of 3.3 should also be
>    <h3> numbered headers.

I agree, not sure what historical artifact led to that.  Will fix.

> 2) Under 3.3, inverseOf, "z" is mentioned in the first
>    sentence (before the colon) but is not used in 
>        P1(x,y) iff P2(y,x)

Will fix.
 
> 3) Under 3.3, inverseFuntionalProperty, on the other hand,
>    "z" is not used in the first sentence but does appear
>     in 
>        P(y,x) and P(z,x) implies y = z.

Will fix.
 
> 4) The first few paragraphs of section 4 (Ontology mapping)
>    seemed to cry out for at least a mention of namespaces.
>    This comment is not that important, but if there is
>    more information in another document about ontology
>    sharing, a reference to it would be helpful.

I see what you are saying.  The intent here was to be fairly 
abstract.  Perhaps I should be more explicit in the example. I
talk about the food and wine ontologies and use the &vin 
construct but without explicit reference to &food.  How about
modifying the second paragraph of 4.1 from

 One way to do this is by defining a class in the food ontology
 and ...

to read

 One way to do this is by defining a class in the food ontology 
 (&food;Wine) and ...

> 5) I think the second example in section 5.1.2 deserves
>    more explanation. Section 5.1.1 is about intersections.
>    The example in 5.1.2 refers to the "intersection of
>    sweet and non-sweet fruit," but the intersectionOf
>    operator that was just introduced is not present. 
>    I see from section 3.1.1 that "Instances of the class 
>    belong to the intersection of the restrictions [that
>    appear in the class definition]." That subtle point is
>    lost in the wake of 5.1.1. 

I thought the point was fairly clear.  But I have been staring 
at this document for a long time.  5.1.1 states that

 The construction above states that WhiteWine is exactly the
 intersection of the class Wine and the set of things that are white in
 color.

5.1.2 states

 This says that the instances of Fruit are a subset of the intersection
 of sweet and non-sweet fruit, which we would expect to be the empty
 set.

>    In fact, would the example in 5.1.2 be better off
>    in section 5.1.1?

The reason to pull this example into the unionOf discussion is that 
some people have misinterpreted the implicit combination of the 
two subClassOf's.  The other point is that this second example is not 
definitional, and the text is an explicit reminder to the reader of
the difference.  As well as the difference between this construct
and that in 5.1.1. 
 
>    Also, the example revealed to me that I do not
>    understand why rdf:resource is used in some cases
>    and why rdf:about is used in others. The topic
>    is introduced in 3.1.1, but I don't get a good
>    understanding of the difference there. The Guide says
>    that rdf:about is used to "extend the definition of
>    a resource." However, I don't see any extensions
>    going on in the example in 5.1.2. 

There is some confusion in my text.  The "extend the definition" text
is meant to differentiate rdf:ID and rdf:about.  And rereading it, I
don't think that is clear, given how rdf:resource is used in the 
middle of the text.  I suggest changing:

 Within this document, the Region class can now be referred to 
 using rdf:resource="#Region". 

to 

 Within this document, the Region class can now be referred to 
 using #Region, e.g. rdf:resource="#Region". 

The id/about distinction is that rdf:ID is meant to be used in the
initial introduction of an element and rdf:about is used to elaborate
on it in a further description.  

rdf:resource is used in a property instance, typically to identify 
the range element.

-----Original Message-----
From: Ian B. Jacobs [mailto:ij@w3.org] 
Sent: Wednesday, October 15, 2003 11:40 PM
To: public-webont-comments@w3.org
Cc: ij@w3.org
Subject: Minor comments on 18 Aug 2003 OWL Guide [attempt 2]


Hello,

[I just hit the send button by mistake on an incomplete
 mail....let's try again...]

I have just read most of the 18 Aug 2003 draft of 
the OWL Guide [1], which I found very readable and
helpful. I have a couple of questions on minor
editorial points.

1) Subsections of 3.4 are numbered; subsections of 3.3
   are not. I think those of 3.3 should also be
   <h3> numbered headers.

2) Under 3.3, inverseOf, "z" is mentioned in the first
   sentence (before the colon) but is not used in 
       P1(x,y) iff P2(y,x)

3) Under 3.3, inverseFuntionalProperty, on the other hand,
   "z" is not used in the first sentence but does appear
    in 
       P(y,x) and P(z,x) implies y = z.

   It looks as though the first sentences were switched...

4) The first few paragraphs of section 4 (Ontology mapping)
   seemed to cry out for at least a mention of namespaces.
   This comment is not that important, but if there is
   more information in another document about ontology
   sharing, a reference to it would be helpful.

5) I think the second example in section 5.1.2 deserves
   more explanation. Section 5.1.1 is about intersections.
   The example in 5.1.2 refers to the "intersection of
   sweet and non-sweet fruit," but the intersectionOf
   operator that was just introduced is not present. 
   I see from section 3.1.1 that "Instances of the class 
   belong to the intersection of the restrictions [that
   appear in the class definition]." That subtle point is
   lost in the wake of 5.1.1. 

   In fact, would the example in 5.1.2 be better off
   in section 5.1.1?

   Also, the example revealed to me that I do not
   understand why rdf:resource is used in some cases
   and why rdf:about is used in others. The topic
   is introduced in 3.1.1, but I don't get a good
   understanding of the difference there. The Guide says
   that rdf:about is used to "extend the definition of
   a resource." However, I don't see any extensions
   going on in the example in 5.1.2. 

Thank you,

 _ Ian
    
[1] http://www.w3.org/TR/2003/CR-owl-guide-20030818/
-- 
Ian Jacobs (ij@w3.org)   http://www.w3.org/People/Jacobs
Tel:                     +1 718 260-9447
Received on Tuesday, 21 October 2003 14:04:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:43:29 GMT