[
Lists Home |
Date Index |
Thread Index
]
Hi
Folks,
An excellent question was
raised:
> Embedded-Grammar&Components approach = RDF?
In the two examples I showed, in fact, each
component did loosely follow the RDF Class/Property/Value design pattern, e.g.,
<Person
id="John">
<age>24</age>
<address>101 Curl Drive, Columbus,
OH</address>
</Person>
[Although, to be truly RDF I would use rdf:ID
instead of simply id, and I would need some of the other RDF
machinery]
This is only a surface resemblance to RDF. I am
making no recommendation on how each component should be designed - use RDF or
not.
The feature of the Embedded-Grammar&Components
approach that I would like for you to focus upon
is the idea of bundling a grammar with the components - that is, bundling
with the components instructions on how to interpret
them.
In a large, complex
Enterprise the set of "things" may be well-known and understood. For
example, in a military Enterprise the things may be "missile", "aircraft",
"tank", etc. In other words, the components of the Enterprise are
well-known and understood. However, how the components
are used in the Enterprise changes quickly. Thus, the problem is:
How can a collection
of well-known and understood components be deployed in a variety of settings,
across a variety of systems? More precisely, how can a collection of
well-known and understood components be deployed in a dynamic
Enterprise?
I am proposing a
solution to this problem. The solution that I am proposing is to bundle a
grammar with the components. The grammar instructs the recipient of the
message on how to interpret the components.
Comments? What are
your thoughts on this idea of bundling components with a grammar which
instructs recipients on how to use the components?
/Roger
| 