What I have is an electronic text of an Anglo-Saxon manuscript (Beowulf),
containing information about scribal variations in how letters were written.
This was often used by scribes to indicate things about how the text
should be read aloud. For example: stress, timing, etc.
These letter forms are represented by a number of entities. For example:
e standard e
&e; high e
&e-; tongued e
&e--; long tongued e
all represent forms of lowercase "e".
&sbr; scribal line-break
&vbr; verse line-break
Readers interested in this level of information want to see where
the scribe placed his line-breaks, his inter-word spacing, and the form
of the letters and abbreviations used (rendered as in-line x bitmaps).
However, many readers will not wish to see these variations, but will
want to see the text in standard ISO Latin 1 characters, with line-breaks
at verse-ends (for poetry), abbreviations expanded, etc.
What I would really like is to have *one* e-text which can be
viewed using different style-sheets presenting more or less information
to the reader depending on his interests.
>>ISSUE: how the reader selects the viewing style<<
Surely this is an application for style sheets rather than PI's
embedded in the text of the document. Embedded PI's would force me to
have more than one e-text source.
For those of you with Arena, a rough flavour of the two formats is given
in the following two URLs:
(the full text is available from links in my home page)
An example of how this might be coded in HTML (excluding the various
<!ENTITY ...> definitions needed) is
--- Tony Jebson --- International Computers Limited (ICL)
--- firstname.lastname@example.org --- http://www.to.icl.fi/~aj/index.html
--- All opinions expressed here (however stupid) are my own,
--- and nothing stated here is an official statement by ICL.