Re: Namespace Draft

My comments re XML usage are the same as for the previous draft.
This document provokes several other thoughts.

1.1 Collection resources.  I would invert the order of the first
3 paras.  State the general description of the concept before the
specifics.

Separately, it is important to clarify whether a collection is
the entirely of a literary work or whether a literary work
may consists of multiple collections ("larger logical units").
For publishing, as distinct from WEBDAV's particular interests,
we need to be able to say that some list of resources constitutes
*all and only* a particular literary work.  In this case, I see
no reason why a book might not consist of multiple collections
(perhaps one covering chapters 1, 2, and 5, contributed by author
A, and another covering chapters 3 and 4, contributed by author B).
Please indicate which of these views you have in mind (or some other).

5.3, para 2, Depth header.  This has been tried in SGML, Hytime,
Hytime-as-revised, and simply does not work to define any but
arbitrary limits in a web of documents, because documents may be deep 
in some places and shallow in others, and the shallow parts may link to 
external documents not a part of the literary work in which the 
links reside.  Consider a collection that consists of a central
work (a collection of its own) with multiple sets of annotations
linked to from it.  Unless you manipulate the structure of the
central work so that it is evenly deep throughout, it's hard to
ask for N levels of depth and get all of the central work with
none of the annotations.  


Regards,

  Terry Allen    Electronic Publishing Consultant    tallen[at]sonic.net
                   http://www.sonic.net/~tallen/
    Davenport and DocBook:  http://www.ora.com/davenport/index.html
          T.A. at Passage Systems:  terry.allen[at]passage.com

Received on Wednesday, 18 June 1997 18:39:17 UTC