| I'm not absolutely positive, but I'm almost certain that this is
the way <FOOTNOTE>,<FOOTREF> are designed to work. (I thought I saw
it documented somewhere, but couldn't find the description.)
|
| re: .1
I don't think so. When I associate a callout with a footnote, I
expect the association to remain in effect throughout the file.
Subsequent footnote references (using the <FOOTREF> tag) should
generate an appropriate footnote at the bottom of the page where the
footnote reference occurs (whenever I use the callout).
What you're saying is that footnotes can appear only on the page
where the reference first occurs. The <footref> tag takes 9 arguments.
I haven't seen many documents where there are 9 footnotes on the same
page. (Although I know, sooner or later, I'll see one.)
Besides, how would you know where one page ended and the next page
began? Only the text formatting stage can tell you that.
In any case, if this is normal behavior, it's not consistent with
the behavior of footnotes in multipage tables. Footnote references
in multipage tables generate a footnote at the end of the table
(on the page where the reference occurs) no matter how many pages
the table appears on.
I may be entirely wrong.
|
| I believe that the <footnote> tag in text places the footnote
text at the bottom of the same page, but does not again repeat the
text on subsequent pages. So a <footref> tag on a later page
forces the reader to flip back to the earlier page on which the
footnote occurred. (I suspect that books in which multiple
references to the same footnote appear will often gather all the
footnotes at the end of the chapter or the end of the book.)
This is inconsistent with the way footnote text is repeated in tables,
as you indicated. But in tables, you specify the <footnote> tag
"up front", after the <table_setup> tag. Then in each table row,
when you want a callout to appear, you use the <footref> tag.
So you might want the sequence of callouts to go down the first
column, then down the second column, then down the third column.
You control that, because you give the actual callout in the arg
to the <footref> tag.
In terms of "being able to implement this stuff", it means that
the tag translator (and TeX) need keep the footnote text around
only for the duration of the table, and then it can be flushed.
If we did the same kind of repetition of the footnote text in
text (outside of tables, that is), we'd have to keep all the footnote
texts around from the start of the book or chapter to the end of
the book or chapter, 'cause we'd never know when you were going
to slip in another <footref> on us.
When you think about it, the whole business of footnoting and
calling out of the footnote from another place (footref) should be
done with symbols. It is kind of ridiculous to have to go
back to the front of a chapter to insert a new footnote and then
have to march forward through the chapter (book?) and renumber
all the old footnotes and footrefs...
Another item, for which we had working code, pressing needs and
not enough time to do it with all the bells and whistles.
|
| After talking to a few editors, I realized that my confusion resulted
from my own use of footnotes. Back in the age of the typewriter,
one avoided using footnote references in text because you always had to
remember to leave enough space at the bottom of the page for one
or more footnotes. The writer had two options: place the footnote
at the bottom of the page once (and then use ibid and so forth) or create
a list of footnotes at the end of the document.
Because my documents aren't term papers or research papers, I have
never used a footnote in text. Occasionally, I use footnotes in
tables. (Although, I knew DOCUMENT made a distinction between
footnotes in text and in tables, I thought it was for coding purposes
only.)
I haven't written a term paper in seven years, so I forgot about
all the canon law for footnotes set forth in Kate Turabian's A Manual
for Writers and the MLA Handbook.
|