Print

Print


Dear reader,
 
Some days ago I was asked to help out in finding a good way of specifying
vertical whitespace in texts, as between paragraphs and sections.
My answer was that the rend= global attribute should be used for this
purpose, as in:
 
       <P rend="3lines, white, at end">Paragraph text....
 
I guess that this approach, or something like
       <P>Paragraph text....
       <P rend="3lines white">
 
(which is in my opinion incorrect) is being used by most encoders.
 
This is not very portable. In many circumstances the encoder likes to pass
on information on the rendition of the text and be sure that at the
receiving end a faily good 'reprint' of the source can be made
*automatically*. Such a relatively faithful reprint is sometimes not
without interest to the scholar. (Consider showing the transcribed text
in a right window, an image of the original in the left. Font changes and
whitespace give some visual alignment between the original and
transcription).
 
This means that some consensus must emerge not only on when to use
rend=, but also on the rend= attribute value. This point is also made in
the Guidelines (p. 48). Scanning for TEI-LIST and the P3 did not help
finding (a starting point for) such a consensus.
 
In the class descriptions, the following is noted on rend=:
 
  "These Guidelines make no binding recommendations for the values of the
  rend attribute; the characteristics of visual presentation vary too much
  from text to text and the decision to record or ignore individual
  characteristics varies too much from project to project. Some
  potentially useful conventions are noted from time to time at
  appropriate points in the Guidelines."
 
In my opinion such consensus would have to become part of P4. The issue
transcends rend= and reaches all informally described, potentially
formal attributes in P3.
 
I would like to know if in any project a clear encoding strategy has
been devised for rendition aspects, and of course if projects have
agreed on a single format.
 
Side remark:
 
  I can imagine the rend= to be an expression in a 'rendition
  syntax' (cf. extended pointer syntax). For example,
 
  <tag rend="P(SPACING 3LI END)">
 
  would specify paragraph level rendition, add 3 whitelines, at end.
  Here, at least three levels of rendition info can be distinguished,
  P() for paragraph level, D() for division level, C() for character
  level. All would allow for informal descriptions to be included
  when the feature is unforeseen or hard if not impossible to encode
  formally.
 
  It should be noted that reference to (standard) style specification
  languages such as FOSI or DSSSL do not exactly fit in. These describe
  generic style aspects, that e.g. apply to all paragraphs in similar
  contexts. They cannot be used when a deviant form has been encountered.
  In this respect a "rendition expression" would modify the effect of a
  predefined style for an element of the type involved, or overrule the
  style completely.
 
Anyway, let me know what you think. If you have such a standard rendition
information format, could you direct me to the place to find it and/or
publish it on TEI-LIST?
 
Thanks in advance,
Arjan.
          Arjan Loeffen, Humanities Computing, Faculty of Arts,
               Utrecht University, The Netherlands.
      ++31+302536417 (voice work), ++31+206656463 (voice home)
            http://www.let.ruu.nl/C+L/loeffen/home.htm