[xtm-wg] Andrius' comments

[Steve Pepper <pepper@ontopia.net>]

Some responses to:

At 20:59 07/02/01 -0800, Andrius Kulikauskas wrote:
>I want to thank you for your tremendous work.  I have only read a
>fraction, but I've read it very carefully.  I focused on the
>Terminology, and I've actually made a diagram of it all.  Your
>descriptions are quite good, some of them excellent.  In particular, I
>liked your definition of Subject and that inspired me to try to unravel
>the whole thing from there.

Thank you for those kind words!

>Here are some nitpicking points, regarding the Terminology section.  The
>things that I don't understand, maybe others won't either. Here are the
>few that stand out:
>
>Topic Occurrences
>I don't understand the distinction between 'topic occurrence' and
>'resource' (addressable information resource).  I think the definition
>for addressable information resource is very clear, simple and central. 
>The definition for topic occurrence is unfocused, indirect and suspect. 
>What does 'information' 
>mean?  Is it in the mind, or in the computer?  What is the difference
>between 'information' and 'information resource'?  Why is the concept of
>'occurrence' 
>necessary?  If it is just for legacy issues (which is legitimate), I
>think it would be fair and noble to say so.  The word "occurrence" is
>itself used here in a very abstract and unhelpfully nonintuitive way. 
>Most importantly, what is the relationship between 'topic occurrence'
>and 'resource'?  If there is a distinction, it would be most helpful to
>make it clear.

An occurrence is a resource that is special because it is known to
be in some way pertinent to a certain subject. The definition has
been adjusted slightly to reflect that. I hope this meets your
concerns.

>Topic Name and Base Name
>I don't understand if there is any real distinction between 'topic name'
>and 'base name'.  It seems very subtle.  My reading is that it's
>ambiguous.  They can mean the same thing (the "topic characteristic"
>which is the name, whatever that means).  Or they can have slightly
>different meanings: 'topic name' is the string of characters 
>within the <baseNameString> element, whereas 'base name' is the entire
><baseNameString> element (including the tags).  In other words, I don't
>understand.  It would be helpful to make the distinction very clear, or
>to say very clearly that there is no distinction.

Again, you are right. There has been some lack of clarity. The
definition now makes clear that "topic name" includes any variants,
whereas "base name" does not. (Ideally I would like to change the
DTD to reflect this, but I think it's too late for that. As it
stands the DTD is not wrong; it is just a little less intuitive
than it could be.)

>Topic Type #3
>In the definition of Topic Type, #3, I think there is ambiguity (or at
>least real complexity) in the word "is".  It says, Topic Type: "A topic
>whose subject is a class of topic."  I first read that to mean: Topic
>Type X is a topic X whose subject S is a class Y of topic X.  And then I
>wondered, what does the "is" mean?  I started thinking confusing things
>about the subject indicators.  (I came to the conclusion that, given a
><TOPICREF>, you don't know if it's intended as a topic or topic type
>(until you look for <INSTANCEOF>, which lets you know - but that might
>be made more explicit). I read it again, and then I thought that the
>second instance of "topic" means "topics" in the plural.  I guess it
>looks like your using topic in that short sentence in two very different
>ways topic=X, but also as Topic(s).  Can this be simpler?

Yes and no. It is in the nature of the power of the paradigm that
it gets complex. (There are not many modelling formalisms in which
a thing can be both a class and an individual.) We use the exact
same formulation for "association type" and "occurrence type".
Those seem to us to be reasonably easy to understand, and that helps
in interpreting the more self-referential "topic type".

>Topic Type #2
>Definition #2 seems ambiguous (the word "the").  Are there, or are there
>not, other kinds of classes, not specified by <instanceof>?  It could be
>supposed either way, which is confusing.  In other words, do topic type
>and class mean the same thing?

Agree about the ambiguity. Changed "the" to "a".

Steve

--
Steve Pepper, Chief Technology Officer <pepper@ontopia.net>
Convenor, ISO/IEC JTC1/SC34/WG3  Editor, XTM (XML Topic Maps)
Ontopia AS, Maridalsveien 99B, N-0461 Oslo, Norway.
http://www.ontopia.net/  phone: +47-22805465  GSM: +47-90827246


------------------------ Yahoo! Groups Sponsor ---------------------~-~>
eGroups is now Yahoo! Groups
Click here for more details
http://click.egroups.com/1/11231/0/_/337252/_/981642727/
---------------------------------------------------------------------_->

To Post a message, send it to:   xtm-wg@eGroups.com

To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@eGroups.com