Re: [xtm-wg] XTM-UCS Analysis

[Patrick Durusau <pdurusau@emory.edu>]

-------------------------- eGroups Sponsor -------------------------~-~>
Test your WML code with our
Online WAP Testing Tool at
http://click.egroups.com/1/9112/4/_/337252/_/969366566/
---------------------------------------------------------------------_->

Ann,

As always your analysis is appreciated and I found it quite helpful.

I think you are correct that my suggestion of using a test case from biblical
studies does start down the slippery sloop of "functional analysis." And I would
agree that avoiding the "requirements bloat" that you see in XML Query activity
(I am not sure some of the participants see it quite that way ;-)) is a good
thing.

On the other hand, the summary of use cases you provide:

<snip><snip>

>  So here's a revised "XTM-UCS list" of usecase-based
> requirements on XTM maps:
>
> 1. Navigating complex content: XTM supports supplementary structure and
> information providing specific support for applications providing
> navigation/access functionality
>
> 2. Topical organization: XTM supports supplementary structure and
> information encoding ontologies, taxonomies, indexes etc.
>
> 3. Knowledge modelling: XTM supports supplementary structure and information
> providing specific support for knowledge management
>
> 4. XTM can "Interchangeably represent" (words from 13250) the structure &
> information in (1) (2) and (3).
>
> Now I'm going to be controversial, possibly:
>
> I have considerable confidence, both from my participation in the abstract
> modelling WG, and now further strengthened by the time I've spent with the
> usecase-survey material, that if XTM meets the abstract/conceptual structure
> which we already have, it will be adequate for uses 1,2,3.
> Therefore, there is no point muddying the waters with further detailed
> requirements in these areas - we have good reason to trust what has gone
> before, including the wide-ranging discussions which fed into 13250.
>

seem to me to be overly broad.

By way of illustration, consider use case #1,  Navigating complex content. I am
not sure that "supplementary structure and information providing specific
support for applications providing navigation/access functionality" provides any
constraint or boundary on the work being performed by the other two subgroups.
While I quite readily admit that my use of the term "use case" is perhaps too
tied to an actual implementation, it seems to me that it is also possible to
have use cases that are so general that they are not helpful in
modeling/implementation or judging that modeling/implementation against the use
case.

What reference material I have been able to consult (primary Schreiber, et. al.,
Knowledge Engineering and Management) appear to take a more specific view of use
cases than you advance but then they are working in the context of developing
actual software systems. Even allowing for the difference in domains (standards
vs. software systems), I am still concerned that the use cases are too broad to
be useful.

Personally (as an individual and not as a representation of any group or
organization) I don't see the use cases as driving either the Conceptual
Modeling or Syntax Interchange groups since both of them are working from the
basis of ISO 13250. To that extent, I think Ann's point to "trust what has gone
before..." is well taken. On the other hand, I would like to see use cases that
are specific enough to make people aware of how XML based topic maps could be
used without straying into what Ann refers to as "functional analysis."

<snip><snip>

What is the sense of the group in terms of deriving more specific use
requirements from the material on hand? If not from the standpoint of informing
the Conceptual Modeling group or the Syntax Interchange group but as general
illustrations of the sort of use cases that XTM could meet? Ann, would that
constitute "muddying the waters....?" I see the general parameters of our
discussions as being bounded by ISO 13250 on one side and the abilities and
limits of the XML family of specs on the other with a looming deadline of
December, 2000 straight ahead. I am not suggesting that we derive uses that are
beyond those bounds or that question those boundaries, at least as part of the
XTM 1.0 spec.

In terms of specific activities, should we use our time in London to build upon
Ann's general use cases so we can deliver that information to the other working
groups and/or as part of the spec so people will have some guidance in terms of
what is in or out of bounds for use cases? And if that is the case, we will need
an agenda so that we can discuss, draft and vote on such specific use cases or
perhaps we should use the words "use case illustrations."

I have been following the posts to the xtm-wg list fairly closely but have not
seen interim results from the Conceptual Modeling group. Would it be possible
for someone to post their latest modeling in a non-proprietary format? (Dia,
gifs, etc.)

Patrick

--
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
pdurusau@emory.edu



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

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