W3C home > Mailing lists > Public > public-gld-wg@w3.org > March 2012

Please consider some additional ISA Programme vocabularies

From: Bernadette Hyland <bhyland@3roundstones.com>
Date: Thu, 8 Mar 2012 06:16:49 -0500
Message-Id: <025F5848-CED8-4E03-B71C-88DA86138DBA@3roundstones.com>
To: W3C public GLD WG WG <public-gld-wg@w3.org>
In the last 3 months in our WG meetings (telecons & F2F in January 2012), we've heard that a couple people from the W3C team are working with the European Commission's ISA Programme to create a number of vocabularies designed to help improve public sector data interoperability.  Phil has been our WG liaison for this work.

Per Phil, that work is due to be complete by the end of March by which time there should be 4 stable vocabularies created by working groups comprising several European data publishers and public administrations plus other experts and one or two folk from outside the EU as well.

The question asked in Brussels is: what happens next?

In terms of implementation, one of them, ADMS, is already the subject of an active implementation program by both publishers and relevant developers. Similar activities are envisaged for the other three. But the vocabularies themselves need to be taken to a global audience.  

Phil identified the opportunity to get these vocabularies considered for publication as a W3 Rec along with ORG, RDF Statistical Cube and DCAT that the GLD WG is working on.  

Having these additional vocabularies go through the W3C process is more work for us however, the heavy lifting has in large part been handled by EC ISA programme participants (hopefully).  However, the additional of 4 vocabs still requires our focus, review, consideration & feedback.  It would be a good way to test our vocab checklist that we're documenting in the GLD BP doc.

The upside to this additional work for the W3C GLD WG?

Adding these 4 vocabs takes what might consider to be regional work and to international standing,  providing a stable, geographically neutral w3.org namespace (the current drafts all use the example.org place holder).

If we do it right, initiatives like vocab.data.gov and Joinup might leverage the same terms.

The 4 vocabularies are:

1) ADMS (Asset Description Metadata Schema), a vocabulary to describe code lists, standards, controlled vocabularies and other 'semantic infrastructure'. The aim is that by using this vocabulary, portals such as the EU's Joinup Platform can act as a go to location for such 'semantic assets' wherever they may be hosted. As well as Joinup, two other consumer implementations are under way as well as many more promises of publishing - getting to Rec is therefore highly possible.

2) Person - a set of properties needed to describe an individual. The person vocab that Michael is working on will go further than the person core vocabulary but the basic properties of things like name, date of birth and so on are all relevant and can be built upon.

3) Business - specifically legal entities - companies registered with some sort of authority - the company you sue if you need to. This is a complement to the org ontology, not a repeat. An organization can be linked to any number of Legal Entity classes. Those that know his work will be pleased to know that Chris Taggart of Open Corporates was a significant voice in that working group.

4) Location - possibly the most tricky of the 4, this one includes an encoding of the INSPIRE Directive's Address model which EU Member States are legally obliged to use in some circumstances and that is not directly compatible with vCard and schema.org. However, as well as various address components, INSPIRE includes an address ID attribute. You can just write an address as a single string and give it an id and that's a conformant INSPIRE address that can then be linked to an address in any other format.

You an see all listed as Semantic Assets on https://joinup.ec.europa.eu/asset/all

What the GLD WG is being asked to do specifically:

1. Review the 4 vocabularies and look for any obvious problems. This needs to be done in time for comments to be acted upon during March (remember they will have reached significant maturity by then).

2. From April, take them on and, by early May, have published the specs as GLD WG First Public Working Drafts. We're asked to do this without making changes from the originals - hence the need for review soon.

From then on the specs are fully under GLD WG change control. No one expects them to go from there to W3C Rec without any changes (although we should be mindful of the existing implementations).

3. In mid-May, the European Commission will seek formal Member State endorsement of the vocabularies in their W3C FPWD form. In the EU's eyes, that's version 1.0.

4. The EU endorsement process will be repeated when the vocabs get to Rec in what they will see as version 1.1.

Known issues:

A) The ISA Programme vocabularies come as UML conceptual models that are then encoded as RDFS and XSD. We will need to consider how we can maintain XSDs of the vocabularies without the WG getting bogged down in religious debates which would serve no one's interests.


Bernadette Hyland, co-chair 
W3C Government Linked Data Working Group
Charter: http://www.w3.org/2011/gld/
Received on Thursday, 8 March 2012 11:17:26 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:32:35 UTC