RE: structured documents [draft-hopmann-collection-props-00.txt]

I agree that taking on compound documents would be a significant amount of
work that probably requires either an addition to our charter or the
formation of a new working group.

Just agreeing on what a compound document is could be challenging.  I was
recently involved in discussions where the following were just some of the
characteristics proposed:

Composed of multiple files
Composed of elements that might not be files (might be database objects,
cells in a relational database, etc.)
Composed of other documents
Component parts can be independently manipulated
Contains multiple renditions
Contains content elements in a variety of formats
Is a configuration of specific versions of component documents
Component parts can be shared by multiple compound documents

Lots of models exist that could be the basis for work in this area.  DMA
provides a fairly simple model where documents are composed of renditions,
which in turn are composed of content elements.  They are struggling with
trying to extend this model to allow documents to contain other documents.
Documentum has a model where documents can contain other documents.  Etc.

--Judy 

Judith A. Slein
Xerox Corporation
jslein@crt.xerox.com
(716)422-5169
800 Phillips Road 105/50C
Webster, NY 14580


> -----Original Message-----
> From: Larry Masinter [mailto:masinter@parc.xerox.com]
> Sent: Tuesday, January 12, 1999 6:44 PM
> To: James.Anderson@mecomnet.de; Lisa Lippert (Dusseault) (Exchange)
> Cc: w3c-dist-auth@w3.org
> Subject: RE: structured documents
> [draft-hopmann-collection-props-00.txt]
> 
> 
> I think if we want interoperable standards here, that it would
> be useful to take on the job of designing a set of attributes
> that could successfully carry a useful document model.
> 
> If support for structured documents is needed for interoperable
> clients, then it should only be "optional" in the sense of another
> set of collective features which form yet another standard for
> which 'conformance implies interoperability'.
> 
> I don't know if this requires a charter addition, or the formation
> of a separate working group.
> 
> Larry
> -- 
> http://www.parc.xerox.com/masinter
>

Received on Wednesday, 13 January 1999 10:49:28 UTC