It may not always be the poster's choice. One of the characteristics MIME was designed with was that certain transformations may be made mechanically and are not considered to change the message (in the senses that, for example, it does not need a different Message-Id:, and should be considered the same for deduping purposes). One of these transformations is a re-encoding, such as a change from Content-Transfer-Encoding: 8bit to Content-Transfer-Encoding: base64 (or the other way around, for that matter). I frequently get mail which was, in the words of the the header added to one example, X-MIME-Autoconverted: from quoted-printable to 8bit by Stone.Rodents-Montreal.ORG id DAA28998 (Stone being my house network's MTA host.) So it's entirely possible that the sender produced something reasonable, like 8bit or quoted-printable, but some intermediate MTA converted it to base64. The right way of handling this from a MIME perspective would be for the some part of the list software to do such conversions, so that (for example) the messages being packaged into a digest all get converted to 8bit before digest generation runs (and the digest itself may itself get converted, either right after generation or by some intermediate mailhost). But, I suspect that, in this case, the brain cycles to make anything like that happen are close to unavailable. I know _I_ certainly don't want to add to the time-and-energy cost the list is exacting from our honoured listowner. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B