Re: [xtm-wg] Re: Summary of PM discussion

[Sam Hunting <sam_hunting@yahoo.com>]

[Sam Hunting]
> >I guess I see things, well, more "multidimensionally". My (CURRENT)
> >preference is for the following:
> >
> >   - graph model (the underlying formalism)
> >   - UML notation (what the model is expressed in)
> >   - infoset-style prose description

[guy]
> I think this is the right top-down approach for a well understood
> problem.

Even a blind pig finds a truffle every so often ;-) (Me, I mean.)

[guy]
> The discussion has congealed around the three basic metaphors for
> modeling information (paths, containers, and positions).  The graph 
> model, based on a connected intersection of paths, provides a high 
> level abstraction for the integration of any structure.  

For "metaphor" should I be thinking "API"? Or is that category
confusion?

[guy]
> Its abstraction is its greatest strength and its greatest weakness.

You're not kidding. "Where do I find/store X" in the graph is the first
question and the answer is not always easy.

[guy] 
> The class diagrams of UML and their inheritance and aggregation
> concepts are a formalization of the container metaphor.  While they 
> don't provide the global integration of a graph, they provide a more 
> intuitive abstraction of the more detailed components (objects) of 
> the system. This is very useful to someone designing the structure of
> an XTM engine.

Does this translate to a claim that the UML notation is not suitable
for specifying "the graph"? I think Eliot would disagree...


> Finally, the property-set uses the metaphor of a multi-dimensional
> position in space to describe the specific characteristics of a 
> system. 

But surely a node in a hypergraph is such a "position"?

[guy]
> These characteristics provide the fine grain detail required to 
> determine in an intuitive way, when two things are equal or when 
> something is valid. 

Particularly useful when doing things like establishing subject
identity?

[guy]
> Each level is grounded on the level above it.  Unfortunately, whether
> it is the right grounding can only be determined by looking at the >
> detail of what it supports.

Sorry, I'm missing the "above" here. Aren't the three metaphors --
containment, paths, and multidimensional positions three ways of
looking at the same thing? Where are the "level"s here?

[guy] 
> Each starting point can be expanded to cover all possibilities.  UML
> is much more than class diagrams, groves have containment and can
> describe graphs, the mathematics of graph theory can describe 
> partial-orders and multi-dimensional spaces.  But when you start 
> to mix and stretch metaphors, it can take a lot of effort to 
> understand what is going on.

The problem is, that not all people "grok" the same metaphor. That is
why I am trying to separate:

    - model
    - notation
    - prose

Surely describing "federated global knowledge interchange" requires
more than one metaphor?

S.

=====
<!-- "To imagine a language is to imagine a form of life."
     - Ludwig Wittgenstein, Philosophical Investigations -->

__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail. 
http://personal.mail.yahoo.com/?.refer=text

------------------------ Yahoo! Groups Sponsor ---------------------~-~>
Find software faster. Search more than 20,000
software solutions on KnowledgeStorm. Register
now and get started.
http://us.click.yahoo.com/556S5A/RNSCAA/2h4EAA/2n6YlB/TM
---------------------------------------------------------------------_->

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

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

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/