The message I received from "F. Scott Ophof" said: ... > If addresses like <listname-REQUEST@..>, <OWNER-listname@..>, or > (gasp!) <listname-OWNER@..> (as a/the MLM at GreatCircle.com does) > are indeed implementation-dependant, then they too are effectively > not too useful as a general case. It occurred to me that none of the above should really be used as indicators of MLM-generated mail for the simple reason that they occur on the RHS of a field name. > > I can accept that MLMs came into general use after standards like > RFC822 were proposed, with those standards not taking MLMs into > consideration. But after all these years, hasn't *any* standard > been proposed and rammed through *YET*?? > > If there ain't none, then I'd like to propose that: > Listname: Name of List <listaddr@...> > be used for this purpose, and for no other purpose. And that this > header and its purpose be registered, etc. Yes, that does seem sensible to have, especially considering that X-List* fields have been floating around for a while now, waiting to be "adopted". A small semantic point here, though: just as one doesn't use a "Sendername:", "Subjectmatter:" or a "Fromname:" field, it would appear to be more appropriate and consistent (imho) to name the field "List:". > I don't care whether it would be "Listname:" or "List-Name:", but > not some "X-whatever:", please. Ie. I retract my proposal to use > "X-List:". > I wonder if the people who started using the X-List* fields (I'm a little short on history here) have anything in the works for formalizing a definitive field now. The SRI-NIC was listed in RFC822 as the keeper of registered field names. This might now be a function of rs.internic.net. > > Or is: > Precendence: group > a valid form to indicate it's a newsgroup item? Are there other > values which are valid here? > "Precedence:" as I know it is a definite sendmail-ism, and is site-dependant. The RHS of this field can be any text defined by the email administrator, and is translated by sendmail into a number (via the config file). This is then used by sendmail to determine whether to favor or penalize the delivery of this message relative to the current system load. I most certainly wouldn't rely on it as an indicator of MLM-generated mail. > > On 11 Jul 1993 22:42:16 GMT Eric Thomas said on LSTSRV-L: > >On Sun, 11 Jul 1993 13:49:22 -0400 F. Scott Ophof said: > >>Recently I've seen items from some "Revised LISTSERV"s which have > >>"X-List:" headers, and that header-line has up to now consistently > >>displayed the relevant and correct list-address itself. > >These are not from LISTSERV. At any rate I don't think this is a good > >solution, one has to think of mail sent to multiple lists, resent from > >list X to list Y, and so on. Wouldn't mail sent to multiple lists be taken care of by their respective MLMs ? And if a message was resent from list X to list Y, wouldn't it now be deemed to originate from list Y's MLM (and contain headers appropriate to list Y) ?.... just asking. The history of the message could, of course, be preserved through the use of the "Resent-" prefix defined in RFC822 (Section 4.2). ... > Regards. > $$\ __________________________________________________________________ Ravin Asar | National Science Foundation System Manager: Unix Systems | 1800 G St. NW #440 | Washington, DC 20550 Official: [log in to unmask] | Phone: (202) 357-5934 Personal: [log in to unmask] | Fax: (202) 357-7663 __________________________________|_______________________________ As always, IMHO.