[
Lists Home |
Date Index |
Thread Index
]
From: "Jimmy Cerra" <jc2astro@hotmail.com>
> I like your thinking... :)
>
> ---
> > It seems that we have a layered architecture emerging:
>
> Where would OpenMath stand?
It sounds like it is an "architecture" (like RDF), which is (roughly) a semantic
pattern that may overlay an element structure. Typically, this is expressed
concretely: we express our semantic pattern in a particular DTD and
then ask "can our source document be transformed into that DTD
using some simple transformation?"
ISO defined a particular set of transformations (architectural forms definitions)
and some other people have made suggestions in this area (such as
Megginson's XAF, Cowan's AFNG, and most concretely Lubell's APEX*)
> OpenMath expressions are meaningless without the context provided by
> externally linked "content dictionaries." As these dictionaries are
> standardized, does this mean the OpenMath is a collection of "information
> pools" rather than an individual language?
Information pool and information item are terms associated, in
particular, with Maler and el Andouloussi's excellent book on information
modeling in SGML, which is mainly concerned with doing large scale
analysis of document sets for large corporation. I don't know Open
Math fits in, I will look at it.
Cheers
Rick Jelliffe
* See http://www.topologi.com/public/dsdl.html for references
|
