2010/8/25 Sebastian Rahtz <[log in to unmask]>
On 25 Aug 2010, at 16:22, Adriaan van der Weel PB wrote:
> For our book trade correspondence project, transcribing C19 correspondence between the Dutch publisher De Erven F. Bohn and their authors, suppliers and booksellers, we have so far used the P4 and more recently P5 DTD:
> <!DOCTYPE TEI PUBLIC "-//TEI P5//DTD Main Document Type//EN"
>    "http://www.tei-c.org/release/xml/tei/schema/dtd//tei.dtd" [
>    <!ENTITY % TEI.header "INCLUDE">
>    <!ENTITY % TEI.core "INCLUDE">
>    <!ENTITY % TEI.textstructure "INCLUDE">
>    <!ENTITY % TEI.transcr "INCLUDE">
>    <!ENTITY % TEI.linking "INCLUDE">
>    <!ENTITY % TEI.namesdates "INCLUDE"> ]>

coo. I am glad someone is using this for real. I sometimes wonder whether we need to keep
generating these fragment DTDs, so its nice to know someone wants it.
> We are now thinking of turning to a schema instead, and wonder whether there is a similarly canonical way of invoking a schema.
See http://www.w3.org/TR/xml-model/#d0e662 . I am not sure about support from editors like oXygen, though.

Sort of. If you use RELAX NG, then no. The closest is the notation oXYgen supports.
 <?oxygen RNGSchema="testmav.rnc" type="compact"?>
but thats just a convention they made up. The RELAXNG massive
don't believe in the idea of a 1:1 correspondence between
XML file and schema, as they think you want to validate against
different schemas at different times.

If you use XSD schemas, its something like

<TEI xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
 xsi:schemaLocation="http://www.tei-c.org/ns/1.0 testminimal.xsd"

but that makes the file invalid against the RELAXNG schema. Frankly, I think
the notation  was a dumb idea from the W3C.

It is all slightly unsatisfactory.
Sebastian Rahtz
Information Manager, Oxford University Computing Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

Sólo le pido a Dios
que el futuro no me sea indiferente