[sc34wg3] Comment on N0328 - identical association members
Lars Marius Garshol
23 Nov 2002 12:35:27 +0100
* Lars Marius Garshol
| The question is whether you think that would be enough. Personally,
| I would think that once this is established a glance at the equality
| rule for association roles would be enough to tell people how it
| works. Do you disagree?
* Marc de Graauw
| No. It's just that XTM allows it, so I think it would be good to
| make it very explicit.
Well, "allows" may not be the happiest choice of term. It sounds as if
having two roles with the same type and player is not allowed in the
SAM. It is allowed, but it's equivalent to have just one.
If you look at F.3.3 in XTM 1.0 you'll see that the same rule appears
| Why allow it in the interchange syntax if neither SAM nor RM4TM does
| allow it?
Simply because expressing this sort of constraints on the syntax is
too difficult. Therefore the deserialization specification describes
what information is redundant and how it is lost.
| The advantage of the way SAM models it (contrary to XTM) is that it
| becomes very simple to explain associations in an intuitive way: an
| association is a set of <topic, role> pairs (stricty speaking a set
| of <topic-reference, role-reference> pairs). Which would be good for
| tutorial purposes.
I agree, and also for implementation purposes, query languages,
mapping to the relational model, and whatnot.
| That lead me to thinking whether we could say an association without
| roles is simply a set of topics .=20
Actually, the decision we made on assoc-role-player-type in
Montr=E9al means that this can never happen. We may want to change our
minds on it, but so far it doesn't seem that people know what was
decided nor have an opinion on it.
| After all, if someone makes an association without roles this person
| does seem to assert the topics in this association form some kind of
| collection without further saying anything about the collection
| (except maybe through a provided association type). I am not very
| certain about the soundness or value of this idea yet.
I agree that it would be possible to decide that that is the
interpretation, but I am not at all sure that we want to. It seems to
me that requiring the roles is much more straightforward.
| Nor of the relation with the ongoing RM4TM discussion about
| roletypes in assertions.
I'm not sure there is any relation. We've established that
associations and assertions are not isomorphic.
 <URL: http://www.ontopia.net/omnigator/models/topic_complete.jsp?tm=3Dt=
Lars Marius Garshol, Ontopian <URL: http://www.ontopia.net >
ISO SC34/WG3, OASIS GeoLang TC <URL: http://www.garshol.priv.no >