>>>>> "Kai" == Kai Großjohann writes: Kai> I think this is because it uses a regexp to find them, and the regexp Kai> doesn't find normal mail addresses in the body. Doubtless it should Kai> be easy to add this. Hm. Yes, there is the variable Kai> gnus-button-alist which might help. I think you mistyped "h a r d" as "easy" there. The regex to match an RFC822 email address is about 7000 characters long, and would take far too long to scan each incoming message. (I've got some friends working on a similar Perl module, and are being likewise frustrated.) For example: fred!barney@stonehenge.com is a legitimate email address. And so is that with included whitespace, as fred! barney@stonehenge.com, but how do you distinguish that from "hey, fred! barney@stonehenge.com" where only the "barney..." part is the email? Nope. The problem is a DWIM problem... there's not enough encoding in the email message to correctly identify the sequence of characters that *may* be an email address. -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Perl/Unix/security consulting, Technical writing, Comedy, etc. etc. See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!