Re: Announcement from Technical Architecture

["Martin Bryan" <mtbryan@sgml.u-net.com>]

Duanne/Anders
>
> There are several known issues documented within.  We are calling for an
> initial comment period of Two weeks from todays date.

You have left the date of 26th May in the document

> To post a comment,  please send it to the Technical Architecture List

Re comment in 1. Introduction that "Someone should be able to store DTD’s
without going through modelling processes"  I wholeheartedly endorse this.
Writing a DTD is a modelling process. It should not be necessary to
duplicate this effort using UML.

Re Fig 2.2 you might need to state explicitly whether or not a Query Deamon
is required to concatenate the results from all relevant repositories before
returning the results, or whether, for efficiency reasons, it can choose to
return results on a repository-by-repository basis.

There should be a means whereby registered "users" of a repository are
notified that it has  been updated.

Re 2.3.1b): It should be possible to use the URI to request all information
relating to the data element, or just part of it (e.g. just the definition,
or just the name, or just the associated datatype). In addition the URI
should be able to be extended to point to the relevant importable object to
allow the data object to be imported into an XML DTD (as an external
parameter entity) or into an XML Schema (as an import element).

Re 3.5a), why must processing instructions be used. (A DTD is not expressed
as a processing instruction but as a form of markup declaration.)

Martin Bryan
Chair, CEN/ISSS DAMSAD project group