[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: ProtocolDataModel issues

On Thu, 16 Sep 2004 18:39:06 -0700, Ezra Cooper <ezra@xxxxxxxxxxxx> wrote:
> I had understood that on this team we'd be talking about issues related
> to manipulating the server's objects, including objects that are not
> normally represented in a "feed" (list of entries). In order to do that
> we should set a scope: which objects do we want to be able to
> manipulate using Atom? Here's my proposal:
> * Entries
> * Authors (Users, Principals)
> * Feeds/Blogs/Channels/Streams/...
> * Categories
> * Templates

The format spec provides a reasonable prose definition of Entries,
Authors, Feeds and their associated slots of data (using Content
Constructs, Person Constructs, Date Constructs, Link Constructs).

> A Category has properties: name, URI, and (optionally) a type. The type
> could identify a categorization system in which this category
> participates, if it's helpful to know that.
> A Template has properties: title, content, URI, MIME Type.

It would be useful to see these defined in a similar way to the model
defined in the format spec. Could you give examples of how these would
be used - what kind of transactions? How they would appear in XML

Various folks are working on other schemas and formal modelling. Some
of us are attempting to express Atom in RDF/OWL (see [1], [2]), so
anything that may help putting Categories and Templates into that
model would be appreciated.


[1] http://bblfish.net/work/atom-owl/2004-08-12/blogexample.html
[2] http://semtext.org/atom/