Print

Print


>The museum community has created a high-level framework - the Conceptual
>Reference Model (http://cidoc.ics.forth.gr/) which describes the types
>of classes which we might be interested in, and their properties, in a
>high-level abstract model.  Could be useful.

According to the "Introduction" on the CIDOC web-site, it is planed to turn
the Conceptual Reference Model into an ISO standard.  Recalling the recent
thread about ISO's policy regarding language identifiers, I wonder what
legal and practical implications this has for adopting and using (parts of)
their ontology for TEI.  I have the impression that such issues are getting
more and more important, as creating ontologies and markup schemes becomes
a business that makes money.

The first fundamental principle of the TEI Consortium, according to its
charter, is: "The TEI guidelines, other documentation, and DTD should be
free to users".  As encouraging this statement in principle is, it is very
vague, though.  And I wonder whether there exists really a consensus among
the TEI Consortium members what this statement implies in detail.

One important lesson to be learned from the open source movement may be
that it is essential to get the legal framework right and clear.  Applied
to the TEI, this means that further efforts are needed to clarify the TEI
Consortium's policy for areas such as:

1. Use of an ontology or markup scheme for data encoding
2. Implementation of an ontology or markup scheme into software
3. Adoption of ideas by other ontologies or markup schemes
4. Accessibility of the specification
5. Republication of the specification (completely or in part)

These issues are not only vital for the TEI DTDs and the guidelines
themselves, but of similar importance for the standards referred to in the
guidelines as "normative", or only as "best practice" or
"recommended".  For example, the current guidelines recommend ISO 8601 as
the format for the "value" attribute of tags from the temporal expression
class (see P4, sec. 20.4).  However, the latest version of this 30 pages
specification, ISO 8601:2000, is not freely available on the Internet, but
only for CHF 110.00 from the ISO shop
("http://www.iso.ch/iso/en/prods-services/CatalogueDetailPage.CatalogueDetail?CSNUMBER=26780&ICS1=1&ICS2=140&ICS3=30").

Personally I would prefer that best practice for academic text encoding
discourages the use of such proprietary standards wherever possible, even
if it is then necessary to develop alternatives for already existing
standards.  The cumulated costs of proprietary standards and their negative
effect on free information exchange are, in my opinion, too high.

Dieter Köhler