[sc34wg3] Re: Mathematical model (was SAM-issue term-scope-def)

[Lars Marius Garshol]

* Sam Hunting
| 
| One advantage of thinking in graph terms is that you get metrics --
| you can say, "so many nodes in this graph", etc -- so design
| decisions/performance tuning could presumably based on suchmetrics.

I think you shouldn't confuse mathematical models with implementation
models. That is, an implementation need not follow the mathematical
model exactly.

In any case, you can achieve precisely the same thing in both the RM
and SAM by counting nodes/information items, so I don't think this
issue need affect the choice of formalism for the mathematical model.

| A second reason to think in graph terms is that using it onemight
| hope to emulate the success of the relational model

I return to this below.

| But I think that while the RM might enable model theory as the model
| theorists might understand it, I don't think it is, itself, a model
| theory.

The RM doesn't exist yet, but what we've seen so far is not even
remotely like a model theory. I agree that it might enable one,
though.

| (Last I looked at RDF model theory, they'd punted on reification, so
| I'm not sure of its real-world utility.)

That made me wonder a bit, as well. RDF's reification does seem to be
under fire from many quarters at the moment, though, so the problem
might be the reification rather than the model theory.
 
* Lars Marius Garshol
|
| That is true. The PMTM4 proponents have made "this-is-a-graph-and-
| therefore-maths"-like noises in the past, but I'm not sure if they
| would make the same claim of the RM.
 
* Sam Hunting
|
| For the purpose of developing metrics, or the purpose of having the
| functional (and, dare I say it, the marketing) equivalent of the
| relational model, yes, the graph "is maths." 

Developing metrics is not maths in the sense that we've been talking
about here, and I don't think the RM is sufficiently formal to serve
as the foundation for something like the relational model. The
mathematical model, on the other hand, ...

-- 
Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
ISO SC34/WG3, OASIS GeoLang TC        <URL: http://www.garshol.priv.no >