I'm new to Latex (using xelatex) and I am trying to lay out a document with lots of short footnotes. I'm using para
and perpage
.
I've tried para
with both bigfoot
and footmisc
. I've tried perpage
from footmisc
, perpage
and bigfoot
. In all combinations of these (when I finally get something compiling), though, I end up with a sort of race condition where the a page will begin with footnotes 27 and 28, for example. When I rerun xetex, the footnotes end up on the previous page but are numbered 1 and 2 and the following page begins on footnote 3.
In other words, it seems as though the footnote number(s) being two digits are at the threshold that wraps the word(s) to the next page. When xetex runs again (with the words and footnotes on the next page), the footnote number is lowered by perpage which causes the word(s) to wrap back onto the previous page.
I can rerun xetex infinitely but the numbering/spacing doesn't stabilize. I can give a working example but it won't be "minimal" because the documents are often long. They're also mostly Greek.
[EDIT:] Like I said, I'm new to Latex and it seems as though the whole system is kind of fragile. If there's a Right Way™ to do it, I'm all ears. One thing I thought might help, though, was if there were a way to set a minimum footnote width. I don't think I have 100 footnotes on any page so it would just be a matter of fixing the width to something wide enough to hold the two widest digits. I don't know if that's possible but that would make footnote numbers predictable for wrapping.
para
, but never-ending oscillations as you describe are quite possible. Can't you stabilize the document using some rewording or geometry, font, etc. changes? If so, you can probably leave the problem as is until the final touches to your document, at which point you can do the required visual formatting to get rid of such problems (and add prominent XXX comments where you did some visual formatting, in case you decide to further edit the document later :-).\raggedbottom
to no avail. From my understanding\pagebreak
suggests a pagebreak at a point. The problem with this is that it's often a sequence of words, not only the ones with the footnotes, that are moving to the next page. Predicting good pagebreak locations would be impossible.