[xtm-wg] RE: RE: Questions on topic maps

["Ivan Uemlianin" <ivan@jurakm.com>]

Thanks for all replies

 

(nb small typo in original post: when discussing ISO 13250, "scope
element cannot occur in topic elements" should have been "scope
*attribute* cannot occur in topic elements")

 

 

1. topic.scope

 

> | I notice that in ISO 13250 topic elements can have scope attributes,
> | but that in XTM the scope attribute cannot occur in topic elements.
> | Why the change?
>
> In ISO 13250 the scope on <topic> was just a shorthand and the topics
> specified in that attribute were intended to be 'inherited' to the
> scopes of all the characteristics of that topic.

 

OK, thanks.

> Scope only applies to the characteristics of topics, not to the topics
> themselves. Names, associations, and occurrences may all have limited
> validity, but not the topic itself.

 

How would you represent a topic with limited validity, or a topic
which only had an interpretation in certain contexts?  How would you
represent the difference between a topic which just happened only/not to have
any characteristics within a certain scope, and a topic which in
principle could only/not have characteristics within a certain scope?

 

Some examples:

 

- A TM being used as a front-end into an encyclopaedia or a
  text-book.  Here it might be appropriate to mark topics (as well as
  occurrences, etc) as 'beginner', 'intermediate', 'advanced'.  This
  would allow toc or index generation to skip topics not within the
  current scope; topics contingently without characteristics but
  within scope would still be listed (allowing users to add links,
  to note the 'empty' topic's relationships with other topics, &c).

 

- More generally, there are concepts which have an interpretation only
  within a certain context.  For example, the concepts of function and
  dysfunction in biology have no interpretation (i.e. are meaningless)
  in physics or maths (e.g. one can have a dysfunctional liver, but 'a
  dysfunctional lever, and 'a dysfunctional number 3' are literally
  nonsense).  How would this be represented in a TM of the natural
  sciences?

 

2. role.class

 

> There is actually a many-to-many relation here, in the sense that a
> topic may play multiple roles, while the same role may be played by
> multiple topics.

 

You're right.

 

> I chose the simpler route, where an association role
> essentially represents

 

[a single aspect of]?

 

> a topic's involvement in an association.

 

The TMDM allows each association role to have one rolePlayer and one
class, and there do not seem to be any restrictions on topics occurring in
more than one association role per association.  So would the suicide of
Tosca example work out as something like the following?

 

association: death-of-character
    associationRole[i].rolePlayer: Floria Tosca
                       .class:      perpetrator
    associationRole[j].rolePlayer: Floria Tosca
                       .class:      victim

 

If this is about right, I agree that it is a good solution.  I can tend to
obsess over terseness (too much Perl; I'm on Python now, and I'm
getting better.  Please no flames).

 

> | We are
>  Jura Technology

 

Thank you for your interest.  Pending a press release and relaunch of our web-site, here are some initial remarks:

 

Jura has developed AutoCat, a suite of tools for text-based information management.

 

Those tools include:

 

1. a highly scaleable XML indexing engine with XQL support;
2. a standards-based Topic Map engine plus associated web navigation and building tool;
3. a Rule-Based categorization engine for classifying input documents;
4. a 95% compliant XSL engine.

 

All tools are C++ based and will soon have CORBA bindings.  They have been ported to Linux and Windows NT and will build on any platform with a GCC compiler.  Web-navigation product works under Apache or IIS.

 

We hope to relaunch our web-site soon, where demos will be available.

 

Best Regards

 

Ivan Uemlianin

 

 

Ivan Uemlianin, PhD
Head of Topic Map Development
Jura Technology Ltd

 

6 Tai Seion
Llanddeiniolen
Caernarfon
Gwynedd LL55 3AF

 

Head Office:
35 Norroy Road
London SW15 1PQ

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

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

Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.