[sc34wg3] How Two Syntaxes Can Make One Standard

Martin Bryan sc34wg3@isotopicmaps.org
Wed, 25 Jul 2001 08:11:45 +0100


Firstly let me thank Michel and Steven for their paper, and state publically
that they have my support in acheiving the stated goals.

As you will have guessed from my first word, there is a "secondly", or more
correctly a "however" :-)

To me the key paragraph of the document is:

> From an economic standpoint, there are significant
> advantages in using a distinct software module that
> implements this generic processing, commonly called a
> "topic map engine" or a "topic map parser".  We urge
> that the term "topic map parsing" be reserved to mean
> all of the aspects of "topic map processing" that are
> required to be done by all topic map software that
> takes, as input, interchangeable topic maps that
> conform to either the HyTime-based or XTM-based
> syntaxes.

The key word in this paragraph is a small one that must never be forgotten:
"all"

Unless all features of the ISO 13250 model can be separately identified as
part of the information set created by the topic map parser for use by topic
map applications this exercise will be simply that - an exercise.

To date I have seen no model proposed that identifies the source of all
aspects of an ISO 13250 conformant topic map. Until I do so I will remain
sceptical of existing efforts.

I have one other request to make to those working on this effort, which is
highlighted by the following paragraph:

> Four rules must be applied by all topic map parsers:
>
> -- the subject-based merging rule
> -- the name-based merging rule
> -- the node-demander rule
> -- the no-redundancy rule

Please so not introduce terminology into the model which is not supported by
definitions in 13250 without a formal definition of the meaning of the term.
For example, there is no concept of node-demander or redundancy in 13250.
There is a concept of merging, but this is clearly identified as being
"scope-dependent" which the above names do not make clear. Any "rules" must
be clearly related to the clause in 13250 which states them. If there is no
such clause than a clear case for the "rule" must be agreed by all members
of the user community, and not just by those involved in the discussions
relation to one of the proposed syntaxes.

Martin Bryan

Martin Bryan