At least for now. This is simply a matter of continuing existing practice. My apologies to our many Japanese contributors, our growing list of Russian contributors, all others whose names cannot be correctly represented in ISO-8859-1, and Hrvoje Nikšić (who gets honorable mention here for all the effort he has gone to to increase awareness of the ISO 2022 Menace). When we change it, it will be to UTF-8. If you use Mule, please do (add-to-list 'file-coding-system-alist '("ChangeLog" . iso-8859-1)) in init.el or otherwise take steps to avoid spurious introduction of ISO 2022 DCS sequences in ChangeLogs. There is a similar problem with Windows and LF <-> CRLF for many kinds of files. This can be (and mostly has been) dealt with by use of the CVS binary file type. Please let us know if you observe similar weirdness with EOL behavior; we probably can arrange that Windows users get CRLFs and Unix users LFs as the gods of Redmond and Murray Hill, resp., intended. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN My nostalgia for Icon makes me forget about any of the bad things. I don't have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py