Proof of Concept

["Paul Spencer" <paul.spencer@boynings.co.uk>]

This project has now been running for a while. In order to make sure we
allow for all possibilities, we have ranged quite widely in what we are
looking at. I feel the time has come to refocus and ensure we deliver the
main requirements in a short period.

Three important points from the eGov TC purpose:

* Provide a forum for Governments internationally to voice their needs and
requirements with respect to XML-based standards which can be handed off to
relevant OASIS TCs

* Provide a mechanism for the creation of best practice documents

* Promote the adoption of OASIS specs/standards within Governments

This project supports all three. We are allowing Governments to state their
needs, we are promoting the ebXML registry and possibly other OASIS
specifications, and we will deliver best practice for the use of registries
for storing schemas and components. So we are on target there.

The aims of the project are:

* To show that the ebXML registry/repository is a suitable platform for
e-Government schema management

* To build a demonstration registry

* To create best practice for its use

Note that the aim is schema management. Although a registry can be used for
many other purposes, the aims of this project have always been around the
registering, storage and discovery of schema documents, schema components
and associated metadata, and the assembly of schema components into
documents.

We have identified several activities required to achieve these aims:

* Provide guidance on the mapping of metadata to the ebRIM. This includes
(possibly as a second phase) deciding between direct mapping and mapping via
the CCRIM. So far, it has resulted in a requested change to the ebRIM. Once
the decision is made on that, the work can be completed quite easily.

* The ebXML registry TC is working on the CCRIM to ebRIM mapping, so we will
rely on their work.

* We need a mapping of schema components to the ebRIM.

* We need to look at schema assembly. I am not yet convinced that CAM is the
right approach to this, but I am equally not convinced that it is not.

* We need a working registry.

Have I left anything out?

The third of these bullets requires the writing of a new standard. Should we
draft this, keeping the ebXML Registry TC informed, or should we stick to
the aims of our TC and pass it over to them to work on? What sort of
priority would they give it? I suspect the best compromise is for us to lead
it, but work as a joint team.

Currently, the document we have drafted is a combination of other documents.
I propose to re-write this based on the content of this email. This will
mean cutting some of the current material, but I think that is necessary to
achieve the focus we need.

I would also like to kick off the assembly part of the project.

I look forward to some lively discussion over the next few days, then moving
the project forward in an agreed direction.

Regards

Paul Spencer
Director
Boynings Consulting Ltd
http://www.boynings.co.uk