RE: [xtm-wg] XTM 1.0 AG Review Specification

["Michel Biezunski" <mb@infoloom.com>]

Jason,
Since the content model of <topic> includes "instanceOf*", I'm
> assuming that
> topics can be instances of multiple classes? Why can't occurrences and
> associations? If they could then you could restrict <instanceOf> to
> appearing only once but allow its content to contain multiple
> <topicRef>s or
> <subjectIndicatorRef>s.

There are 2 ways to consider occurrences and associations.
1) As having something to do with topics.
2) As topics themselves.

In the first case, it's better they are well-defined, because otherwise
instead of being able to merge various topic maps, you would spend your
time wondering whether they actually are the same, or only look the same,
or could be the same. You can always use scope to add semantics to those.

In the latter case, when you decide to treat the association or the
occurrence
as a first class topic, then you get access to the full description of
topics,
including the fact each of them can be the an instance of various things.

> From an aesthetic point of view, <baseNameString> just looks
> funny. Why not
> <name><baseName>My Base Name</baseName></name> instead? This seems more
> applicable since <baseName> also contains <variant> elements which are
> alternate names and not really part of the base name.

Everybody has different opinions about what is the best name. Remember this
is
only for interchange between machines. If the design is correct, then it's
fine. If
you are designing a user interface you can always change the way it's
presented
to your users, provided that your application knows how to express it into
the common
interchange jargon.

> If I want a set of variant names to be applicable across different scopes,
> do they have to be repeated in the <baseName> element that
> contains each of
> these scopes? Or are the variant names declared in the unconstrained scope
> used as a fallback when no matching parameters are found in a specific
> scope?

Variant names are inside a given scope. They are really nothing more than a
way
to say: if you use a display device that can't display more than a certain
number of characters,
this is what to use. Variants should not be used to represent for example
names in different
languages, because they fall into different basenames with different scopes
instead.

> The first example in section 3.7.3 (<variant> Element) is rather
> perplexing.
> Are the two topics part of the same document?

Yes.

>If I'm not mistaken, the
> second topic (with the id of "sort") is meant to be merged with
> the official
> "sort" published subject indicator based on its sole subjectIndicatorRef.
> The offical "sort" includes an equivalent subjectIndicatorRef so the two
> topics are deemed to have the same subject, correct? Why didn't
> the example
> refer to the offical "sort" directly? Are the core PSIs
> automatically merged
> with every topic map?

Because the example doesn't use the published subject indicator, instead it
uses
a topic whose id is "sort" and that is present in the same document. You are
never
obliged to use published subject indicators. If you want to use your own,
you can.


> By the way, core.xtm, language.xtm, and country.xtm are undisplayable with
> Internet Explorer. It's complaining about an undeclared xlink namespace
> prefix. I don't know why it's not getting it from the DTD (where it's
> declared as #FIXED) but it would be nice if that could be added.
>
> Why <roleSpec> and not just <role>?

Same answer as for basenamestring. Role is overloaded. It has to do with the
difference
between the way xlink uses role and the way links are done in XTM. If we
would have put
role, then the question would have been: why your role is different from
xlink:role ?

> Typo in F.2.4: "considrer".
>
> Since you've created topics for language and country codes will
> you also be
> publishing PSIs for the various MIME media types as well?

This is a trap in which we are trying to avoid. Actually, the way country
are used is to
show what is the mechanism to refer to an existing ontology. All existing
lists of codes,
terms, etc., can be used as  published topics. This should (in my opinion)
not be the domain
of XTM, but on the contrary everybody should be able to appropriate to
represent those.

Regards,
Michel
==========================================
Michel Biezunski, InfoLoom
Tel +33 1 44 59 84 29 Cell +33 6 03 99 25 29
Email: mb@infoloom.com  Web: www.infoloom.com
==========================================


------------------------ Yahoo! Groups Sponsor ---------------------~-~>
eGroups is now Yahoo! Groups
Click here for more details
http://click.egroups.com/1/11231/0/_/337252/_/981875142/
---------------------------------------------------------------------_->

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

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