Lars =?iso-8859-1?Q?Hellstr=F6m?= writes:
 > At 14.29 +0200 2001-06-05, Chris Rowley wrote:
 > >1.  How should LICR strings be written out to files used only by LaTeX
 > >    itself?

In the LICR, everything else will be a big mess... so UTF8 if the LICR
is Unicode.

 > Are there technically any such files today? BibTeX reads .aux files.
 > Copying and editing the contents of .toc, .lof, etc. files is an
 > established (although maybe not kosher) method for finetuning the table of
 > contents and such.

I don't think this means we have to support arbitrary encodings of
auxiliary files.  Editing such files is, after all, undocumented
practice, and using an UTF8 editor will (rather elegantly) provide
full access to these files.  Do we need more?