[tei-council] Further update on PH

Syd Bauman Syd_Bauman at Brown.edu
Mon Sep 24 16:59:33 EDT 2007


LB> but i am still not persuaded that these things should be separate
LB> attributes anyway. We didnt have this argument when we were
LB> specifying how to give <geo> map coordinate values. Why are we
LB> having it now?

SR> * because 4 numbers is more than 2
SR> * because geo can be in various forms, and datatyping is harder
SR> * because the geo numbers are generally passed as is; as Conal's code shows,
SR>   these 4 numbers will be very commonly used in calculations, individually

And perhaps we should have had this argument when <geo> was
specified. I note that <geo> is underspecified: users are left to
their own devices as to the syntax of the content. (See footnote to
http://lists.village.virginia.edu/pipermail/tei-council/2007/007937.html.)


Personally I kinda prefer 4 attrs over 1 attr that has to be parsed,
but I don't think it's a big deal either way. (And this is the very
kind of stuff that folks have been saying is easier in XSLT 2 on
TEI-L just recently :-)

As for the names, the PostScriptish urx=, ury=, llx=, and lly= are
pretty geeky and opaque, but are short and do make sense once you've
read the glosses. As someone pointed out, these attrs are mostly
going to be written by software, not people.


Do I grok correctly that the proposal to replace box= with xMax= and
yMax= means that we presume xMin=0 and yMin=0?



More information about the tei-council mailing list