Raf Prinke writes:
>I felt confident and expanded the "magic line" to:
> <!ENTITY % x.data "persName|placeName|person">
>only to find out that <person> does not allow #PCDATA, so that I cannot
> Lord Jim and his wife
>as two <person>'s. Moreover, <person> and <personGrp> are only allowed
>within the document profile in <teiHeader> for listing participants
>in a linguistic event.
This is correct. A <person> is not the same as a name. If I remember
rightly there was considerable opposition to the idea that the DTD should
allow you to pretend that they are the same kind of thing, however
convenient it might occasionally be to do so.
>The definition of <name> in P3TAG.DOC (ie. Chapter 35) says in part:
> <name> (i.e. name, proper noun)
> Description: contains a proper noun or noun phrase.
> Remarks: Proper nouns referring to people, places, and
> may be tagged instead with <person>, <place>, or <org>.
I think this counts as a mistake. The remarks, as you suggest, date from
an earlier period of TEI thinking: they should be revised to read, "may
be tagged instead with <persName>, <placeName>, or <orgName>"
>So it seems that the original intention was to use <name> and <rs> for
>proper nouns and descriptive strings respectively, with a possibility
>of using <rs> only for both. Thus their functional equivalents in a more
>detailed encoding would be the pairs: <person> and <persName>, <place> and
><placeName>, <org> and <orgName>. But even though <place> and <org> are
>used in examples and recommended, they are not available in the DTD's
>nor discussed elsewhere in the Guidelines. Thus
There should be no examples of <place> or <org> in P3. I think we've
cleared them all out of the revised text (yet to be published) -- if you
have a list of places where they subsist, please let me know (page
references to P3 would be most convenient) and I'll check again.
> <rs type=person>
> <name type=personal>
>is a less precise (or indirect) way of saying:
No. They are different. The first is a reference to some person, which
contains a name within it. The second is data about a person, which
includes a personal name within it. <persName> is syntactic sugar for
<name type=person>. <rs> is a more general kind of <name>. Either may
appear within <person>, reflecting the fact that a person can have many
>Another difference in behaviour is that while <placeName> may appear
>inside itself, <persName> cannot. So while it is possible to have
> <placeName>Greenwich near <placeName>London</placeName></placeName>
>doing the same with "Kenneth mac Alpin" or "Elizabeth wife of Philip"
>is not allowed by the DTD.
It should be. If this doesnt work for you, then your copy of the dtd is
wrong. There are several examples in P3 which do precisely this, and they
all work for me!
As for Panorama stylesheets: yes they do a lot, but no, they don't do
everything. I am told that you can do even more, and even more wonderful,
things with Multidoc Pro, which seems to be a safer bet than Panorama for
other reasons too, but I haven't tried it myself. (You can download a free
30 day demo copy of Multidoc pro from the Citech website if you want to
check it out) And of course the *proper* way of doing it is to convert your
stylesheet to DSSSL, and use a DSSSL renderer, If you can find one
(actually, you can: try Fujitsu's new HyBrick browser -- also freely
available over the web).
Some time ago, someone announced availability of a Perl script for
converting Panorama stylesheets to DSSSL, but by the time I came to look
for it, it had gone. Anyone seen it?
>Does anyone know any tricks with Panorama+TEI which would impress
>knowledgable but traditionalist scholarly editors?
Hoyt Duggan has quite a few. Check out his Piers Plowman project.
>All my best wishes for the New Year to all TEIsts (and aTEIsts, too) :-)
From The Desktop at Burnard Towers