TEI TRC Core Subcommittee

Notes for telephone conference of 1997-01-08

TC Core M03

Dominic Dunlop

8 January 1997

Table of Contents

1.1 Attendance

1.2 Administrivia

The conference call scheduled for Monday 1997-01-13 is to be rescheduled to Tuesday 1997-01-14 at the same time (14:00-15:00 UTC). As LB will be in the US at that time, he will join the conference by calling 1 (800) 457-0261, rather than needing to be called by the AT&T operator.

Further conference calls will be scheduled for the Mondays of 1997-01-27 and 1997-02-03 at 14:00-15:00 UTC as usual.

1.3 Item 2.7 Syd Bauman on various items (1 certainty, 4 errata)

1.3.1 Item 2.7.1 <certainty>

The feeling of the group was that more problems would be created than solved if <certainty> were allowed to have content describing the basis of a judgement of certainty. Existing mechanisms, such as links, feature structures, and notes can be used to embellish <certainty> where necessary. In particular, if <certainty> elements have IDs, a descriptive <note> may target one or more of them.


STATUS: Closed

1.3.2 Item 2.7.4 Errata sheets

The group agreed that errata sheets were a topic of sufficiently general interest that users of the TEI Guidelines would benefit from examples. Syd's examples are reasonable, although the first has a more presentational bias than might be desired, and assumes that the errata sheet and the text that it corrects constitute a single document instance.

ACTION: Editors to include discussion of means of handling errata sheets in future revision of P3.

STATUS: Closed

1.4 Item 2.8 Dominic Dunlop on MS markup (item 2)

This is another take on the "anonymous chunk-level element" issue. LB pointed out that, at phrase level, one has a choice of two elements -- <s> (non self-nesting), and <seg> (self-nesting). <seg> can also be used at the chunk level, but there's nothing analogous to <s> -- that is, something that behaves like a <p>, in that it doesn't self-nest, but that doesn't carry the semantic baggage of <p>. <s> itself cannot be used, as one might reasonably want to nest <s>s inside something that is paragraph-like. So there's a clear case (at last) for defining a new element. <block> is an obvious name, but too long; <b> means "bold" to those exposed to HTML... Some better name is needed.

ACTION: CMSMcQ: Draft a proposal. STATUS: Open

1.5 Item 2.9 Lou Burnard on seg

It should certainly be possible to use <seg> to mark sub-parts of <u>, whatever base(s) have been selected. If it's not, there's a problem with the DTD.

ACTION: LB: Draft a solution. STATUS: Open

1.6 Item 2.10 Henry Thompson on seg

Not discussed due to lack of time, but related to 2.8 and 2.9 above.


1.7 Item 2.11 Harry Gaylord on s (dup)

Already discussed. See item 2.4 in minutes of 1996-12-16.

1.8 Item 2.12 Peter Flynn on sgml documentation

Discussed only briefly. It was suggested that a revision of P4 should discuss the linkage between a TSD and corresponding document instances in more detail.


1.9 Item 2.13 Peter Flynn on Curia suggestions (separate documentation)

Not discussed due to lack of time.


Dominic Dunlop