W3C home > Mailing lists > Public > public-webont-comments@w3.org > March 2002

on Use cases/ Web portals

From: Jim Farrugia <jim@spatial.maine.edu>
Date: Wed, 20 Mar 2002 10:10:24 -0500 (EST)
To: public-webont-comments@w3.org
cc: Jim Farrugia <jim@spatial.maine.edu>
Message-ID: <Pine.OSF.3.91.1020320100300.7994A-100000@nasa.spatial.maine.edu>
Hi,

Here are some comments that I hope may be helpful.

Jim

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

2 Use cases 
2.1 Web portals
Paragraph 3, beginning with the third sentence.

Original:

This ontology can provide an expressive terminology for describing content, 
and inferences sanctioned by the ontology can be used to improve the 
quality of search on the portal. For example an ontology can include knowledge 
about the topic of the portal such as "academic papers are written by 
one or more authors, which are people; people have surnames and given 
names and affiliations, which are organizations" and so on. These rules 
might say that the surname, given name, and name of affiliated organization is 
sufficient to unambiguously identify a person in the community. That is the 
sort of inference that an ontology can enable.

Comments:

1. The first sentence above talks about "expressive terminology for describing 
content." The second sentence gives the example of "knowledge about
the topic of the portal such as academic papers are written by one or 
more authors, which are people;..." This example seems to be talking about
 more than just "expressive terminology." That is, I sense a mismatch 
between what the first sentence promises and what the second sentence 
delivers by way of example.

2. The third sentence starts with "These rules," but it is not clear what
rules are being referred to. Are the rules being referred to the "knowledge" 
mentioned in the previous sentence? If so, then maybe the second 
sentence could mention (at least parenthetically) that knowledge is 
expressed in rules, or something to that effect. But even with such a 
change in the second sentence, if there is no change in the first sentence, 
there would still be mismatch between the "expressive terminology" of the 
first sentence and the "knowledge" of the second sentence.

3. The "These rules" of the third sentence presents another problem. Let's
suppose that the rules being referred to are the ones expressed in the quoted 
statements in the second sentence (i.e., "academic papers are written by..."). 
The third sentence says that these rules "might say that the surname, 
given name, and name of affiliated organization is sufficient to 
unambiguously identify a person in the community." The rules might say 
this, but they don't. So, there is a compounding confusion that builds in 
the reader's mind: the uncertainty about what "These rules" refers to 
is increased when the reader has to try to understand what these rules 
"might" do. 

4. The last sentence says "That is the sort of inference that an ontology
can enable." The antecedent for "That" is unclear. It seems that the 
"That" refers to the what the rules might say, namely, that "the surname, 
given name, and name of affiliated organization is sufficient 
to unambiguously identify a person in the community."  But, if so, aren't 
you talking about a rule, not an inference? Can you clear up for me the
difference between a rule and an inference? Do you see any difference? Also,
perhaps some additional explanation could be given about "the sort of 
inference that an ontology can enable."  

5. None of the sentences after the first one above shows directly how 
"inferences sanctioned by the ontology can be used to improve the quality
of search on the portal."

6. I would suggest a rewording of my own that addresses my comments, 
but I can't until I better understand the following: (1) what you mean 
by expressive terminology, knowledge, rules, and inferences; and (2) how 
inferences can be used to improve the quality of a search.
-----------------------------------------------------------------------------
Received on Wednesday, 20 March 2002 10:15:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:43:27 GMT