[sc34wg3] Alignment of N0396 with N0393

Lars Marius Garshol sc34wg3@isotopicmaps.org
29 Apr 2003 10:00:31 +0200

* Lars Marius Garshol
| My first question about this document is how it is intended to be
| used. Is it just a rough draft showing a first attempt at a SAM-RM
| mapping that is intended to live together with SAM and RM, or is it
| intended as a replacement for N0396? Clarification on this would be
| welcome.

* Jan Algermissen
| Well, it shows the approach to be taken for expressing the ontological
| commitments of N0396 in terms of N0393.

I meant in terms of the ISO process.
* Lars Marius Garshol
| The subtle problem is that the merging rules are wrong. Subject
| identifiers and source locators share a namespace,
* Jan Algermissen
| Oh, what does that mean?

See section 3.4.5, equality rule, fourth bullet point. I'm also pretty
sure you don't implement the fifth bullet point there.
* Lars Marius Garshol
| Further, base names are here merged if they have the same string
| value,
* Jan Algermissen
| They are not. Why do you think so.
| What do you mean by 'base name'? The name(-string) or the naming
| characteristic?

The characteristic. 

Maybe I am wrong here, but if so I really need help on figuring out
what is going on. Let's say we have a topic with three topic names:
"Norway" in the scope "English", "Norge" in the scope "Norwegian", and
"Norway" in the scope "English" (again). Could you walk us through the
RM representation of this and merging that would happen to it, and show
how the RM would know what to merge at each step?

| Well, it surely can but in our conversation we figured out already
| that the prose of N0393 is not clear enough for a number of people
| to understand how.

The latter part of your sentence is certainly true. Being walked
through an example might help me understand it, however.
* Lars Marius Garshol
| SAM associations are also modelled as RM assertions, despite the fact
| that these are structured differently, 
* Jan Algermissen
| Why are N0396 associations structured differently from N0393
| assertions?

You know that perfectly well, having just explained the rationale for
it to Dmitry. See your own posting at <URL:
http://isotopicmaps.org/pipermail/sc34wg3/2003-April/001767.html >
* Lars Marius Garshol
| Another subtle problem is that a number of things that are handled by
| the infoset formalism in the SAM is done by prose in the RM, 
* Jan Algermissen
| Which ones?

Come on, Jan. Here's one example, from your description of PR_Scope:

  "Topics that exhibit a value for this property must be surrogates
   for relationships (that is, they must exhibit a value for the
   property IS13250::a-sidp)."

I'm not sure if it does so correctly, but obviously it does try to say
that scope only applies to topic names, occurrences, and associations.
(When you add variants it will apply to those as well.) The SAM
handles this through the formalism, by specifying a property on
certain item types.
| The prose of N0393 may be unclear, but I don't see how the infoset
| formalism is more clear that a property based model?

Compare the documents, then. How hard do you think it is to understand
N0396, compared to understanding N0393 and N0406?

| So do you think that a 15+ pages document is less complicated than a
| 3 pages XML doocument (that can evetually even be processed by
| software)?

Yes. Also, you are not comparing apples and apples here. The 3-page
thing does not implement all of SAM, nor is it self-contained.

As for "processed by software" I think you will have to wait a while
for software that can read a sentence like

  "Values are equal if the first strings of the pairs are byte
  identical and the second strings of the pairs are byte identical."
and implement it correctly. The same applies throughout. N0393 and
N0406 are both very far away from being implementable, and automated
implementation of N0406 you can forget about so long as it remains in
anything like its present form.

| The draft alignment contains a suggestion for a slight modification
| of the model in N0396 in order to reduce the complexity of that
| model. I'd be curious what you think about this proposed change.

It has been considered many times by the authors of N0396 over the
past two years, and we have discussed it on this mailing list several
times, but we have always decided against it for many different
reasons. The costs of supporting reification for roles in name and
occurrence assignments and obscuring the conceptual structure of the
model were among the chief arguments against it.

I think you can find more discussion on this in the "occurrence
variants" thread.

Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
GSM: +47 98 21 55 50                  <URL: http://www.garshol.priv.no >