From lou@ermine.ox.ac.uk Thu Sep 20 11:49:51 2001
Date: Thu, 20 Sep 2001 10:38:52 +0100 (BST)
From: Lou Burnard <lou@ermine.ox.ac.uk>
Reply-To: lou.burnard@oucs.ox.ac.uk
To: Laurent Romary <Laurent.Romary@loria.fr>
Subject: Re: [Review] Terminology chapter

Dear Laurent

Many thanks for your comments on the terminology chapter. Some of them are
easily dealt with (the absence of typographic indication for element names
was an error in the formatting of P4 which we've now fixed, for example)
and some are simple slips simple to correct (e.g. unquoted attribute
values). However, you also raise an important question about the validity
of retaining the chapter at all in the light of subsequent standardization
efforts, which suggests that if we do retain it at the very least it will
need to be hedged around with some kind of warning note redirecting
readers elsewhere. 

More significantly, I think, we should take the opportunity to re-think
the role of this chapter completely, perhaps recasting it at a higher
level as a chapter about "terminological data", which points to existing
standards and significant practice in this general area. Important topics
not covered elsewhere in the Guidelines which might be addressed here
include ontologies such as wordnet and conventional thesauri such as the
Getty AAT. The TEI community needs to know about such resources, and how
to interact with them, but we don't need to re-do their work, any more
than we need to re-do MathML.

However, all of that is just my personal opinion, as is my personal
agreement with several of the criticisms you advance of the current
chapter. To make progress on this, I need to consult with other people
more expert in this domain than I, and you need to come along to the TEI
members meeting in Pisa this November and make the case for substantial
revision or removal to the TEI membership. I will of course do my best to  
make the case for you if you cannot attend, but I think you would be far
more persuasive!  

So, could you let me know as soon as possible whether or not you plan to
attend that meeting? I believe Julia has already asked you to send in a
brief statement for circulation to the TEI voters, so I won't nag you
about that as well!

with best wishes

and many thanks for your effort

Lou



 On Thu, 20 Sep 2001, Laurent Romary wrote:

> Dear Lou and Steve,
> Here below you will find my general and specific comments concerning the 
> Terminology chapter, as you will see, this is a candidate for major revision.
> Tell me if this is OK for the time being or whether you need additional 
> precision.
> Best wishes
> Laurent
> 
> Chapter 13 - Terminological Databases
> 
> * General comments
> 
> This chapter is clearly obsolete in the light of recent standards adopted at 
> ISO TC37/SC3 for terminological data representation (ISO 12200, a recent 
> variant of it, known as TBX, has been recently adopted by the Localisation 
> Industry Standard Association - Lisa) and current work at ISO on the 
> definition of a generic plateform for terminological markup (ISO CD 16642, 
> TMF - Terminological Markup Framework). The underlying principles expressed 
> in the document (in particular the opposition between nested and flat 
> structures) are quite far from existing practices in the terminological field 
> and may be misleading for someone starting a TEI based project that would 
> contain a terminological component.
> 
> There is one paragraph (p. 298 "In terms of the extended...") exemplifying 
> the use of the extended pointer natation for internal reference within a 
> terminological entry. Should this be kept or modify using the XPath syntax?
> 
> Should we not replace all character entities (especially those corresponding 
> to iso8859-1 characters) by their full code, in order to make the examples 
> more readable. There are several of these in the German and French examples 
> provided in the document. This probably a general editorial policy decision.
> 
> I do not know what is planned to clearly identify element names in the text 
> and whther reviewers are supposed to identify them.
> 
> * Proposal for revision process
> 
> The first step is to identify if there currently exists markup projects 
> relying on the current version of Chapter 13 and evaluate the consequences of 
> completely redefining the format.
> The revision should be based on the principles expressed in ISO CD 16642, so 
> as to ensure both a compatibility with the TBX proposal (which is an 
> application of ISO CD 16642) and provide a way to incorporate the general 
> mechanisms provided by the TEI. This work could probably be a testbed for a 
> similar activity in the domain of lexicography (TEI "Print dictionary" 
> chapter) and perhaps other sections of the TEI (linguistic annotations).
> 
> * Corrections to be considered for the current version of the document
> 
> p.291
> 
> Delete sentence "In this guise ... (E-TIF)."
> Delete sentecne "When used for marking up... (P-TIF)."
> 
> Section 13.1
> 
> Replace "A sample terminological entry consists of a series of entries like 
> the following:" by "A sample terminological entry consists of a series of 
> components like the following:"
> 
> Replace all occurances of "ISO WD 12 620" by "ISO 12 620:1999", the first 
> reference of which should be followed by "(Computer applications in 
> terminology  Data Categories)"
> 
> p.293
> 
> Delete the two sentences from "At the time of publication, work is under 
> way..." until "will appear as ISO 12 620."
> 
> Replace "that will be defined in" by "that are defined in"
> 
> p. 298 A number or reference to attribute-value pairs are missing their 
> quotation marks, namely:
> n=1 >> n='1'
> group=1 >> group='1'
> group=2 >> etc.
> n=endes1
> n=dedes1
> n=frdes1
> 
> p. 302
> 
> Replace "this section shows the how" by "this section shows how"
> 
> Delete second paragraph of 13.5 "The interchange of terminological data..."
> 
> 
> 

