Re: [xtm-wg] Editors Alert: scopes on association roles.

[Steve Pepper <pepper@ontopia.net>]

At 22:48 08/02/01 -0500, Nikita Ogievetsky wrote:
>It came to my attention that F.3.2 says:
>(as I was revisiting F.3.3 :-))
>  
>"4. Either the Class or Instance roles are in a scope other than the unconstrained scope. "
>  
>I think this line should be removed because there is no scopes on association roles.
>It must be a leftover from the old DTD.

It isn't incorrect, but it could be slightly confusing.

It is, in fact, the *roles* that are scoped, as the conceptual
model clearly states (see B.5 and B.8, but note that there is
a typo in the current version of B.8: "membership" should read
"role").

This is also as we have always said:

   "A topic has three kinds of characteristic: names,
    occurrences, and *roles* played in associations."

   "Scope expresses the limits of the validity of a
    topic characteristic assignment.")

However, *in the syntax* there is a constraint that all roles
in the same association have to be in the same scope. This is
enforced by having the <scope> element be a child of the
<association> element rather than the <member> element. (See
the note on B.8 in Annex B.) Again, this constraint is something
that was discussed and decided in Dallas.

So as it stands F.3.2 is not wrong in the way you think.

Having said that, there *is* something wrong with that fourth
condition in F.3.2, which is that it seems to forbid the use
of scoping on class-instance associations altogether. This
was not my intent, and I plan to clear it up with Graham today.

For me, <instanceOf> is a useful -- but less powerful --
alternative to using a class-instance association. The reason
I say it is less powerful is precisely because you cannot
express scope when using <instanceOf>. We've known this all
along, and it's OK. <instanceOf> is syntactic sugar for the most
common case. But that is no reason to hamstring the mechanism
used to express the more general case.

I will be suggesting to Graham that we add the words "in the
unconstrained scope" after the word "association" in the second
para of F.3.2, and that we remove the fourth error condition.

Steve

--
Steve Pepper, Chief Technology Officer <pepper@ontopia.net>
Convenor, ISO/IEC JTC1/SC34/WG3  Editor, XTM (XML Topic Maps)
Ontopia AS, Maridalsveien 99B, N-0461 Oslo, Norway.
http://www.ontopia.net/  phone: +47-22805465  GSM: +47-90827246


------------------------ Yahoo! Groups Sponsor ---------------------~-~>
eGroups is now Yahoo! Groups
Click here for more details
http://click.egroups.com/1/11231/0/_/337252/_/981705366/
---------------------------------------------------------------------_->

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

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