[
Lists Home |
Date Index |
Thread Index
]
Didier PH Martin <martind@netfolder.com> writes:
>
> Hi Peter
>
> > Haven't explored Xlink: using Cocoon and aggregation (and
> some of the
> > newer Cocoon features for flow manipulation of multiple XML
> pipelines)
> > I suspect we've negated much of the need for it. Then
> again all our
> > XML manipulation is purely local, do you think we'd see
> more value in
> > Xlink if we where aggregating non-local XML?
> >
>
> Precisely. Also another advantage is to be able to type the
> links. For example I can link or create an association
> between to frames/objects/row. That association can even be
> typed like for instance having the association "brotherOf" or
> "allied" or "commercialPartner". Typing association is what
> UML is doing but the association types are limited. More
> power to modeling domains can be obtained by being able to
> create custom types as we can already do with class based
> programming languages.
Interesting, my inclination so far have been to restrict association
types to those defined in existing modeling tools. In particular, UML
and ER modeling. We've barely had requirements for semantics beyond
basic pointers (references) though it's clear some kind of inheritance
and composition is needed within the next year. If you're creating your
own association types how do you model them?
|
