[topicmapmail] Understanding the Reference Model

[Jan Algermissen]

Hi Alexander--


Alexander Johannesen wrote:
> 
> Hi all,
> 
> I'm really trying to grasp the ins and outs of the RM, 

That is really good to hear!

> and having just
> fought myself through the RM Editor's Draft 2.32, I have a few
> questions. First of all, is that the latest revision?

Yes. Regarding any future revisions: the concepts are now very stable,
though the propse (the way to explain them) might change in the future.
. 
> 
> 1. Are the terms RM and TMM used interchangably?

Yes,sorry if that causes confusion. The whole thing started out as the
'Reference Model' but since it is really describing the nature of Topic
Maps, we thought we'd better call it the "Topic Maps Model".
 
> 
> 2. Is my assetion :) that the RM is what others might call a pattern
>     correct?

I don't really understand what you mean by 'pattern', but let me say
what the RM intends to be: The goal of the RM is to provide a standardized
means to define Topic Map Applications. A Topic Map Application (TMA) is
a set of semantics consisting of the definitions of properties and
assertion types, and merging rules. An example for this is the Standard
Application Model (SAM) that somehow reflects what people commonly seem
to understand "Topic Maps" to be.

The RM enables topic map authors to package their topic maps with any
(additional) number of TMAs, thus enabling them to be interchanged
between peers while preserving their semantics.

Very simply expressed, the idea is this: If you use, for example, an
assertion type (or association type in XTM-speak) expressing the
mother-child relationship type in your topic maps and only hand them
to someone else without additional information - how is the other
party supposed to figure out what your association type means?

By supplying a Topic Map Application that (propably among others) defines
the semantics of your association type, you provide a standardized
definition that the recipient can use to figure out what knowledge
your topic map actually represents.

It is important to stress the idea of publicizing TMAs, providing
TM authors with a means to re-use well-known sets of semantics. This 
works a bit like RDF namespaces for that matter.


> 
> 3. The specs seems to be totally chemically free of examples? And
>     will there be some, either in UML and / or XTM format?

All the practical stuff there is about the RM is propably in my
implementation (TmTk). It comes with a small number of TMA definitions
in an XML syntax so maybe looking at them and working with the
software would help you. Although TmTk is already quite functional,
I am still reluctant of releasing it due to the notorious lack of
documentation. (I can't really document the software without
writing an RM commentary...)
I think you would profit from playing around with it in your course
of getting your head into 'RM mode'. Let me know if you want the latest
version.
 
> 
> I'll confine myself to this as a start-off.

What do you mean? Doing examples? That would be great. What do you have in
mind?


Thanks for trying to understand the RM. I *know* it takes some effort and
patience! But the concept is really very simple.


Jan



> 
> Kind regards,
> 
> Alexander
> --
> ___________________ ______________________ _____________________________
>                     |                      |
> http://shelter.nu/ | alex at shelter . nu | http://shelter.nu/xsiteable/
> ___________________|______________________|_____________________________
> 
> _______________________________________________
> topicmapmail mailing list
> topicmapmail@infoloom.com
> http://www.infoloom.com/mailman/listinfo/topicmapmail

-- 
Jan Algermissen                           http://www.topicmapping.com
Consultant & Programmer	                  http://www.gooseworks.org