Adventures with org-footnote-auto-adjust

[tsd@tsdye.com (Thomas S. Dye)]

Aloha all,

I keep footnotes in their own section and appreciate having them out of
the way, where I don't have to think about them.

Recently, I set #+STARTUP: fnadjust because I thought it would be nice
to have the footnotes sorted and renumbered.  This was a mistake in my
case that lost data.  I think I know how it happened, though I wasn't
really paying attention to the footnotes as I worked.

In the document I'm editing, I have sentences like this:

  If you want a list to start with a different value (e.g., 20),[fn:17]
  start the text of the item with ~[@20]~.

As a matter of style, I prefer the footnote (which contains qualifying
text, rather than a reference) be at the end of the sentence, and that
it immediately follow the period.  So, I cut and paste to get this:

  If you want a list to start with a different value (e.g., 20),
  start the text of the item with ~[@20]~.[fn:17]

Now, the next time I insert a footnote, with C-c C-x f, I get something
like this:

  If you want a list to start with a different value (e.g., 20),[fn:17]
  start the text of the item with ~[@20]~.[fn:17]

The text of the original footnote, [fn:17], is lost, though the mark
remains in the text.  If the new [fn:17] is some distance away, then the
problem of duplicate numbers isn't readily apparent in the midst of
other work.  Of course, I subsequently discovered that `~.[fn:17]'
wasn't working and put the space back in.  Now, the footnote refers to
the wrong text.

I've learned that there are certain conditions (I don't know how many)
where the space after a sentence won't accept a footnote insertion. The
example sentence is one of these. Apparently, it is the `~.' combination
that triggers the condition. Org is good enough to prohibit inserting a
new footnote into one of these "black holes" (which is how I discovered
them), but it doesn't mind if I cut and paste a footnote into one. 

I'm not certain how much mischief this might have caused. I discovered
the problem when the text of *both* footnotes in a section of the
document were incorrect.

In my case, org-footnote-auto-adjust doesn't perform any crucial
function--it just makes the Org mode buffer seem more orderly.  Given
that there are "black holes" in the buffer, whose presence have the
ability to confuse org-footnote-auto-adjust so that data are lost,
should org-footnote-auto-adjust be deprecated?

All the best,
Tom

-- 
Thomas S. Dye
http://www.tsdye.com