🚀 go-pugleaf

From: emv@math.lsa.umich.edu (Edward Vielmetti)
>I can't imagine anyone putting byte offsets into the file as cross
>references; perhaps they are (page no, line no) offsets, but that
>would be problematic for electronic text.  Better would be some sort
>of unambigous reference to where the reference is (however you would
>say "paragraph 3 of the 'Publishing' reference in the Macropedia").

Actually, it's less unimaginable than you might think. Remember that
texts are a lot less changeable than other types of documents you
might be accustomed to. I doubt anyone is editing the bible tho (and
if they do something like that then they better make a *copy* and call
it something else!)

Better would probably be to insert unique-reference tags into the text
and then compile a look-aside table of byte offsets for quick
reference which can be rebuilt easily if needed.

Meaning, you'd insert things like this: <refid 123347 "fnords"> into
the text (the number just being an increasing integer to ensure
uniqueness) and then compile a table (with a program) that might have
entries like:

	ref	as appears in text	filename	byte-offset

	fnords	<refid 123347 "fnord">	foo.txt.1	2366
	
Something like that.
-- 
        -Barry Shein

Software Tool & Die    | {xylogics,uunet}!world!bzs | bzs@world.std.com
Purveyors to the Trade | Voice: 617-739-0202        | Login: 617-739-WRLD