>> Nevertheless, declarative markup format and processing heuristics
>> are inherently different.
The idea here is not to code an heuristic, it is to code a parameter which
would be used by such mecanisms.
>> What the TEI can do, IMHO, is only to provide a mechanism for the
>> application to ask the user for information. For instance, a
>> mechanism thanks to which the application can define a way for user
>> to tell which element should be used as sentence boundary.
But suppose that we are at the moment when the user has already answered
the questions you mention to the application. Wouldn't it be possible to
encode the answers in the document itself for later reuse ? isn't it the role
of a "processing parameter" ?
>> I don't see why tagUsage cannot be the natural place for storing more
>> formalised structures an user could use to tell an application the
>> usage it should do of tag.
The question here is not the one of formality of description, it is only that
tag names alone are not sufficient to be the unique entry points for an
application to be told how it should process a document. This comes
from the fact that the data model of an application for a text is often
not the same as the TEI data model for a text.
>> Rendition is not really a precisely defined element. It is a
>> placeholder (a "hook" say the Guildelines) for expressing thinks the
>> TEI is not responsible of.
This is a rather strange statement. If we expect applications to be "responsible"
for particular treatments of documents, based on specific encodings, it would
be nice if those encodings could be a minimum "precise" or expressed with
>> The discussion is about coupling a format intended to human-oriented
>> annotation with processing tools which need a serialisation format.
I think this is the point where we disagree. For me, encoding represents a
contract between humans about the interpretation of a text AND represents
a contract between humans and machines about the interpretation they
should share for the things they have to do together : that is text interchange
between humans AND machines. If TEI is only for humans, I am afraid
we will not be able to built useful software for TEI texts.
Serge Heiden, [log in to unmask], https://weblex.ens-lsh.fr
ENS-LSH/CNRS - ICAR UMR5191, Institut de Linguistique Franšaise
15, parvis RenÚ Descartes 69342 Lyon BP7000 Cedex, tÚl. +33(0)622003883