Re: [egov] "Semantic Interoperability" - a new TC?

[Farrukh Najmi <Farrukh.Najmi@Sun.COM>]

peter@justbrown.net wrote:

>John, as promised:
>
>
>
>Below is a very draft outline for a possible new OASIS TC, or at least a distinct piece of work. It has already been posted and discussed with a number of people and OASIS members, but not yet published as a formal proposal. It can also be seen online at:
>
>
>
>http://www1.europarl.eu.int/forum/interop/dispatch.cgi/sibig/docProfile/100004/d20031216125609/No/t100004.htm
>
>
>
>Since drafting this (after discussion with and at the request of a number of participants at the XML Conference in Philadelphia in December) I have noted other discussions and initiatives that touch the same/similar areas including:
>
>- The Danish Government ITST's proposal to the eGov TC list on "Naming and Design Rules for eGovernment;
>
>- a proposal within the ebXML Registry TC to set up a new sub-committee on Semantic Content Management;
>  
>
Above mentioned Semantic Content Management is already established an SC 
within ebXML Registry TC. Carl Mattocks and I chair that SC and Duane  
Nickul and Joe Chiusano are members. All four of us are members of the 
egov TC.

A key goal of this SC is to enable semantic interop using ontologies and 
semantic equivalence between ontology classes.
We would very much welcome folks with experience in this area to join us 
in this effort. We would also welocme liaisoning
with relevant groups working on related problems.

>- some of the work advancing on the Business Centric Methodology;
>  
>
>- the OASIS Symposium on "Reliable Infrastructures for XML" on 26-27 April.
>  
>
We will be submitting a paper titled "Semantic Content Management using 
ebXML Registry" for this forum. We hope that it is accepted.

>Although the emphasis to date has been on creating technical specifications that certainly advance the cause of semantic interoperability, I think that we need also to look at business-level issues.
>
>
>
>I am looking to secure funding through the European Commission for a "work package" managed through the European Standards Organisation, CEN, to develop a set of guidelines for public administrations that would cover, at a policy level, issues relating to:
>
>- object (class) identification;
>
>- naming conventions;
>
>- (core component) vocabulary development;
>
>- registration and infrastructure;
>
>- governance
>
>
>
>I have tried to persuade the Commission to do this work through OASIS, but thus far, to no avail: not through lack of willingness, but because of a rather tortuous set of financial regulations.
>  
>
Thank you for making this case. OASIS is well suited for this activity 
given its ongoing TCs and SCs.
I am not sure however that a new TC may be the most effective path 
forward. It seems that the existing
Semantic Content Management SC, ebXML Registry TC, BCM are defining the 
specs that provide
the foundation. egov TC is defining the requirements and best practices 
to bring form and structure
based upon teh foundation.

A new TC could have the potential of being too much of a good thing (as 
in "we love standards so much we need 2 of each" ;-) )

>
>
>The issues for the eGov TC as I see it are:
>
>- is there a distinct job of work to be done, in line with the attached draft mission?
>
>- could this be done within an existing OASIS TC, Which one? By whom?
>  
>
The best fit in my not so unbiased opinion is the Semantic Content 
Management SC (SCM SC) of the ebXML Registry TC.

>- how does the proposal relate to/overlap with the other (draft/proposed) initiatives?
>  
>
The SCM SC would address only the core foundation of vocabulary semantic 
and syntax based upon OWL/RDF etc.
We would still need things like Naming and Design Rules to be addressed 
elsewhere.

>- if not, should a new TC proposal be launched?
>  
>
I think a new TC does not make sense under the title of Semantic 
Interop. Maybe we should have one or more SCs of egov TC looking at NDR 
etc.?

>- is there "added value" to be had from a CEN input (with possible EC support)?
>  
>
Absolutely.

>
>
>The text of the strawman proposal is below:
>
>
>
><citation DocID="eu.ep:doc:gri.gen.0059" VerN°="0-1" xml:lang="en-gb">
>
>
>
>Title: Semantic Interoperability Business Implementation Guidelines (SIBIG or "See Big")
>
>
>
>XML has clearly proven itself as a standard that enables syntactic interoperability between systems, but it has a long way to go to improve and enable semantic interoperability. 
>
> 
>
>Vertical industries and communities of interest are forging ahead with particular markup vocabularies within their specific domains, and in response to their domain-specific needs.
>
> 
>
>Experience in many organisations has shown, however, that many business issues concerning the use of XML need to be addressed very early on, if the full benefits are to be reaped of an XML-centred and semantically interoperable information architecture. This was most clearly and most recently demonstrated during the XML2003 Conference in Philadelphia in a number of (unintentionally) related presentations from the public sector [1].
>
> 
>
>Properly considered within any enterprise of community of interest, these issues will provide a clear framework for semantic interoperability, centred on the use of key XML-based standards and specifications.
>
> 
>
>The purpose of the SIBIG TC is therefore to:
>
>1. Consider the commonly raised business issues surrounding the introduction of XML standards, including the importance of:
>
>- a common business vocabulary;
>
>- standardised naming conventions for document, content and other digital object types;
>
>- a proven methodology for developing the above;
>
>- mapping mechanism between the business vocabulary and specific XML-based constructs, including:
>
>  - element and attribute names and values;
>
>  - metadata elements and values;
>
>  - business processing terminology;
>
>  - RDF and Topic Maps subjects, identifiers and indicators
>
>- establishing a framework within which these business issues are developed;
>
>- a system of governance that oversees and maintains this framework;
>
>- handling information assets as any other type of business asset,
>
> 
>
>2. Demonstrate the business value in such an approach;
>
>3. Demonstrate the best use of existing specifications, in particular those of ebXML and UBL;
>
>4. Work together with existing agencies, particularly in the public sector in North America and the European Union, to encourage use of such standards and frameworks, to complement their own work in these areas;
>
>5. Propose a standard approach for the development of such frameworks, through the creation of a series of best practice documents and framework "templates";
>
>6. Vest ownership of the TC's deliverables in OASIS, and to make them freely available to anyone without licensing or other fees;
>
>7. Offer a cost-effective solution to organisations and enterprises wishing to improve semantic interoperability.
>
> 
>
>The TC shall take as its starting point the work on semantic interoperability as presented by the European Parliament, the Danish Government, the Naming and Design Rules work of the UBL TC and the ebXML Core Components specifications. In addition, it should work in close conjunction with the BCM and eGov TCs.
>
> 
>
>The deliverables of the SIBIG TC will be a set of guidelines and best practices that establish a framework for developing semantic interoperability upstream from particular XML implementations. These deliverables should be drafted in such a way that they are as directly usable as possible with commercial off-the-shelf XML-enabled applications and software or, where appropriate, can be used as requirements documents to direct the work of bespoke systems developments.
>
> 
>
>The work will be carried out in a number of phases:
>
>1. A face-to-face forum to collect experiences and exchange initial ideas on the scope of the deliverables. This could be envisaged alongside the XML Europe 2004 Conference (April 2004, Amsterdam) and/or the OASIS Symposium on "Reliable Infratsructure for XML" (April 2004, New Orleans)
>
> 
>
>2. Initial research to identify those aspects of current practices that could be included in the proposed deliverables (2Q2004 and 3Q2004)
>
> 
>
>3. Finalisation by the TC of a set of formal deliverables (4Q2004)
>
> 
>
>The TC will seek to raise reseach funding from public agencies to underwrite the costs of the reasearch, project coordination and involvement of non-profit organisations that are not able to support full OASIS membership.
>
> 
>
>[1] See in particular: "Information Architecture with XML: From Lingua Ubiquita to Lingua Franca", "How the US Federal Government is Using XML", and "Naming and Design Rules for E-Government - The Danish Approach"
>
> 
>
></citation>
>
> 
>
>I hope this helps for the TC meeting tomorrow
>
> 
>
>Peter
>
>
>
>To unsubscribe from this mailing list (and be removed from the roster of the OASIS TC), go to http://www.oasis-open.org/apps/org/workgroup/egov/members/leave_workgroup.php.
>
>  
>


-- 
Regards,
Farrukh