Hi Rob and List,
This same thing happens a lot in a manuscript I'm working with, and
I've dealt with it by creating a list of "abbreviation descriptions"
in the header, assigning each one an xml:id, and linking the
descriptions to individual <abbr> tags using @decls. If a single word
has multiple abbreviations, there will be multiple references in
<item xml:id="macron-u">-m- or -n- abbreviated by a macron over the "u"</item>
<item xml:id="que">-que abbreviated as "q:"</item>
<abbr decls="#macron-u #que">Quicûq:</abbr>
I also use <hi rend="abbr"> to highlight the specific letters that
represent the abbreviations, which helps with HTML output.
<abbr decls="#macron-u #que">Quic<hi>û</hi><hi>q:</hi></abbr>
Using this method, your markup would look something like this:
<item xml:id="yt">"that" abbreviated as "yt"</item>
<item xml:id="s">"is" abbreviated as "'s"</item>
<abbr decls="#yt #s">yt's</abbr>
This looks nicer, and you have the advantage of having space to
describe your different abbreviations. I think it opens up searching
capabilities as well. Does this do what you need?
List, is there anything terribly wrong with what I'm doing? Right now
my abbreviation list is in sourceDesc (or is it encodingDesc?) but
when I have a msDescription it'll be moved there.
I had actually been using @corresp instead of @decls to link the two,
but I have been lead to believe that @decls is the proper attribute
for linking between text and header - is this correct?
On 9/14/06, Robert Whalen <[log in to unmask]> wrote:
> Here is my solution for the doubly-abbreviated yt's (that is) in early
> modern verse:
> xpan>that is</expan></choice>
> Though this seems perfectly compatible with my P5 dtd (at least Oxygen is
> not indicating otherwise), I am concerned about nesting abbreviations within
> abbreviations, particularly as regards future processing for display of both
> original and modern spelling versions. If the content of the <expan> element
> were "that is" only-i.e., <choice><abbr>yt's</abbr><expan>that
> is</expan></choice>-then the metre in the modern-spelling version would be
> spoiled by an extra syllable.
> I suppose I could treat yt alone as a <choice><orig/><reg/></choice>
> construction and the abbreviation separately, thus:
> ><choice><orig>yt</orig><reg>that</reg></choice> is</expan><choice>
> But hitherto I've been treating yt (like ye for the, wch for which, etc.) as
> an abbreviation-and rightly so, I believe.
> Having rather poor XSLT chops, I can only assume that my current solution
> (the first example above) is susceptible to the processing needed to produce
> the desired display outcomes-both the fully abbreviated original (yt's) and
> the metre-retaining, because only partially expanded, modern spelling
> (that's)-even while it preserves the full expansion (that is) at the code
> level. Is that assumption reasonable? Or perhaps I should ask, is the
> solution really the best one?
Dot Porter, University of Kentucky
Collaboratory for Research in Computing for Humanities
[log in to unmask] 859-257-9549
Editorial Assistant, REVEAL Project
Center for Visualization and Virtual Environments
[log in to unmask]