jbn@glacier.STANFORD.EDU (John B. Nagle) (08/02/88)
The X mail system (is this part of X windows?) seems to indulge in heavy self-promotion. Is all the following, taken from incoming mail, necessary? It has little value to the recipient. X-To: DYNSYS-L%UNCVM1.bitnet@jade.berkeley.edu X-Mailer: Elm [version 1.5b] X-Mailer: Elm [version 1.5d] X-Origin: The Portal System (TM) X-Possible-Reply-Path: Ed_Eric_Mitchell@cup.portal.com X-Possible-Reply-Path: mslater@cup.portal.com X-Possible-Reply-Path: sun!portal!cup.portal.com!Ed_Eric_Mitchell X-Possible-Reply-Path: sun!portal!cup.portal.com!mslater X-St-Vmsmail-To: MIPL3::ST%"jbn@glacier.stanford.edu",RGD059 X-St-Vmsmail-To: ST%"jbn@glacier.STANFORD.EDU",RGD059 X-St-Vmsmail-To: ST%"jbn@glacier.stanford.edu" X-Vms-To: IN%"jbn@glacier.STANFORD.EDU" John Nagle
george@ditmela.oz (George michaelson) (08/03/88)
From article <17601@glacier.STANFORD.EDU>, by jbn@glacier.STANFORD.EDU (John B. Nagle): > > The X mail system (is this part of X windows?) seems to indulge in > heavy self-promotion. Is all the following, taken from incoming mail, > necessary? It has little value to the recipient. > > X-To: DYNSYS-L%UNCVM1.bitnet@jade.berkeley.edu Gentle flame. Read the standard, X-<field> is the RFC822 extension mechanism. it allows mail systems to add fields into the message header for end-to-end (well, I suppose other people/things could use it) purposes. -value is in the eye of the beholder. your mail system (and mine) may balk a little on these fields, but it should not barf. It is in the standard to ignore them. Elm clearly offers some enhanced functionality. apologies to mail-gurus if this is wrong/to late. -george -- George Michaelson, CSIRO Division of Information Technology ACSnet: G.Michaelson@ditmela.oz Phone: +61 3 347 8644 Postal: CSIRO, 55 Barry St, Carlton, Vic 3053 Oz Fax: +61 3 347 8987
dagg@lace.lbl.gov (Darren Griffiths) (08/03/88)
In article <17601@glacier.STANFORD.EDU> jbn@glacier.STANFORD.EDU (John B. Nagle) writes: > > The X mail system (is this part of X windows?) seems to indulge in >heavy self-promotion. Is all the following, taken from incoming mail, >necessary? It has little value to the recipient. > Things in a mail header that have an "X-" flag are mailer independant. This means that anything that starts with "X-" was put in by some mailer somewhere, either as a comment or to pass commands specifically interpretted by that mailer, and they are ignored by all other mailers. I believe that this is the only way to add lines to the header that is supported as part of the standard, all lines that don't begin with "X-" have to do with real information that is documented in RFC822. It's a little hard to tell what everything below is doing without looking at the complete header of the message, but most of the things are comments. > X-To: DYNSYS-L%UNCVM1.bitnet@jade.berkeley.edu I'm not sure where this came from, maybe some strange IBM system on BITNET. > X-Mailer: Elm [version 1.5b] > X-Mailer: Elm [version 1.5d] Says that the mail went through the Elm mailer (twice), mainly comments. > X-Origin: The Portal System (TM) Free advertisments for the Portal System. > X-Possible-Reply-Path: Ed_Eric_Mitchell@cup.portal.com > X-Possible-Reply-Path: mslater@cup.portal.com > X-Possible-Reply-Path: sun!portal!cup.portal.com!Ed_Eric_Mitchell > X-Possible-Reply-Path: sun!portal!cup.portal.com!mslater All semi useful stuff from portal listing possible return paths. I believe this is actually put in by Elm, but it is not used any mailers I know of, it has to be interpreted by a human. There are basically two return paths listed, one via a domain-ised internet and one via uucp. I believe that the two different versions of the address are just aliases for the same person. If the entire world used the domain based systems then this would certainly be unnecessary, and even with things the way they are it shouldn't be required if all sites in the intervening path builds the return address correctly. > X-St-Vmsmail-To: MIPL3::ST%"jbn@glacier.stanford.edu",RGD059 > X-St-Vmsmail-To: ST%"jbn@glacier.STANFORD.EDU",RGD059 > X-St-Vmsmail-To: ST%"jbn@glacier.stanford.edu" These three lines come from the Software Tools Mailer (yeah!!!) running under VMS, with mail sent from VMSmail (yuch) into the Software Tools system. > X-Vms-To: IN%"jbn@glacier.STANFORD.EDU" This alse comes from VMS, someone sent the mail from VMSmail into a TCP/IP based mailer. I believe this one looks like Excelans's. --darren ------------------------------------------------------------------------------- Darren Griffiths DAGG@LBL.GOV Lawrence Berkeley Labs Information and Computing Sciences Division
lamy@ai.toronto.edu (Jean-Francois Lamy) (08/03/88)
Resonable mail user interfaces allow you to filter whatever you consider to be blithering. X-whatever headers are defined in RFC-822 to be extensions to the standard set. Mailers on Symbolics lisp machines, for example, use such headers to document what the control sequences in the messages mean, so that the recipient can see the italics the sender put in. Other purported uses would be documenting the encoding format of a message (e.g. uuencoded), so that a co-operative mail user agent knew how to deal with them. As to the infamous X-Mailer: ..., consider it to be gratuitous publicity, just like having your car dealer's sticker on your car. My mail set-up zonks that one into oblivion. The X-VMS-To: header is actually useful if you have to deal with the brain-dead mailer that DEC gives with VMS: the interpretation of the % sign is radically different than the usual subsitution for "@", and is a signal that you are probably asking for trouble if you trust the mailer provided return address. All in all, no, there is absolutely no link with X. Jean-Francois Lamy lamy@ai.utoronto.ca, uunet!ai.utoronto.ca!lamy AI Group, Department of Computer Science, University of Toronto, Canada M5S 1A4
ron@topaz.rutgers.edu (Ron Natalie) (08/03/88)
You're joking, right? This has little to do with X windows. It looks like ELM turds. Remember the X- construct from RFC822.
matt@oddjob.UChicago.EDU (Ka Kahula) (08/03/88)
The point is not "What does it mean?" but rather "What the hell good is it?" I think most of the "X-" headers ever used are useless and were probably inserted for someone's own fun or for, as jbn suggests, self-promotion. The "X-Mailer" headers are the ones the really get my dander up, but even worse than these are the one or two dweebs out there who insert "Precedence: special-delivery" into every message they send! They know who they are ... ________________________________________________________ Matt Crawford matt@oddjob.uchicago.edu
jbn@glacier.STANFORD.EDU (John B. Nagle) (08/04/88)
In article <601@helios.ee.lbl.gov> dagg@lbl.gov (Darren Griffiths) writes: >Things in a mail header that have an "X-" flag are mailer independant. Shouldn't this be "mailer dependent"? Incidentally, some new examples of advertising have arrived. X-Mailer: MMDF/Ream version 4.12 (The Choice for a New Generation) X-Subliminal-Message: Send me all your money My point is that this may be getting out of hand. But it will probably get worse for a while, and then some sites will start trimming headers during forwarding. John Nagle
john@jetson.UUCP (John Owens) (08/04/88)
> The X mail system (is this part of X windows?) seems to indulge in > heavy self-promotion. Is all the following, taken from incoming mail, > necessary? It has little value to the recipient. The "X-" stuff has nothing to do with a "X mail system"; RFC822 specifies that non-standard headers should begin with an "X-" to avoid conflict with any future defined headers. The following are from a variety of mail systems: > X-To: DYNSYS-L%UNCVM1.bitnet@jade.berkeley.edu From a bitnet gateway at berkeley (UREP) > X-Mailer: Elm [version 1.5b] > X-Mailer: Elm [version 1.5d] From two different versions of Elm. > X-Origin: The Portal System (TM) > X-Possible-Reply-Path: Ed_Eric_Mitchell@cup.portal.com > X-Possible-Reply-Path: sun!portal!cup.portal.com!Ed_Eric_Mitchell From "portal", trying to let the recipient know possibile reply addresses, since addresses passing through uucp land are likely to become unrecognizable. > X-St-Vmsmail-To: MIPL3::ST%"jbn@glacier.stanford.edu",RGD059 > X-St-Vmsmail-To: ST%"jbn@glacier.STANFORD.EDU",RGD059 > X-St-Vmsmail-To: ST%"jbn@glacier.stanford.edu" One VMS mail system showing what addresses the VMS user sent mail to, before they were converted to "normal" addresses. > X-Vms-To: IN%"jbn@glacier.STANFORD.EDU" The same, from another VMS mail system. -- John Owens john@jetson.UPMA.MD.US SMART HOUSE L.P. uunet!jetson!john (old uucp) +1 301 249 6000 john%jetson.uucp@uunet.uu.net (old internet)
mlandau@bbn.com (Matt Landau) (08/04/88)
In comp.mail.headers, matt@oddjob.UChicago.EDU writes: >The point is not "What does it mean?" but rather "What the hell good >is it?" I think most of the "X-" headers ever used are useless The most useful X- header I've seen is X-Content-Type, an unofficial way of doing the Content-Type: header that Marvin Sirbu has proposed as an extension to 822 for mail systems capable of accepting and displaying different message types (text, PostScript, Andrew message format, Diamond encoded multimedia format, etc.) On the other hand, X-Zippy-Quote-of-the-Day is at least amusing once in a while, and X-Return-Path can be a useful hint to a USER saddled with a brain-dead mailer, whereas X-Mailer is just annoying.
stewarta@sco.COM (Stewart I. Alpert) (08/04/88)
In article <17601@glacier.STANFORD.EDU> jbn@glacier.STANFORD.EDU (John B. Nagle) writes: > > The X mail system (is this part of X windows?) seems to indulge in >heavy self-promotion. Is all the following, taken from incoming mail, >necessary? It has little value to the recipient. > > The X-whatever headers are not from any single system. There is an RFC-822 convention that allows [programs|sites|people|etc] to add their own headers by proceeding them with 'X-'. This assures that RFC-822 compliant mailers won't confuse these with 'officially' defined headers. These headers may or may not have any value to the recipient (or sender for that matter). :-) ***************************************************************************** Internet: stewarta@sco.COM UUCP: {uunet,ucbvax!ucscc,decvax!miscosoft}!sco!stewarta
dave@viper.Lynx.MN.Org (David Messer) (08/05/88)
In article <17601@glacier.STANFORD.EDU> jbn@glacier.STANFORD.EDU (John B. Nagle) writes: > > The X mail system (is this part of X windows?) seems to indulge in >heavy self-promotion. Is all the following, taken from incoming mail, >necessary? It has little value to the recipient. > > X-To: DYNSYS-L%UNCVM1.bitnet@jade.berkeley.edu ... > X-Vms-To: IN%"jbn@glacier.STANFORD.EDU" > > John Nagle Read RFC 822. Header fields beginning with "X-" are gauranteed to not conflict with the standard or future versions of the standard. (They are standard, non-standard fields ;-) -- If you can't convince | David Messer - (dave@Lynx.MN.Org) them, confuse them. | Lynx Data Systems -- Harry S Truman | | amdahl --!bungia!viper!dave | hpda / Copyright 1988 David Messer -- All Rights Reserved This work may be freely copied. Any restrictions on redistribution of this work are prohibited.
diamant@hpfclp.SDE.HP.COM (John Diamant) (08/05/88)
> The X mail system (is this part of X windows?) seems to indulge in > heavy self-promotion. Is all the following, taken from incoming mail, > necessary? It has little value to the recipient. > > X-To: > X-Mailer: > X-Origin: > X-Possible-Reply-Path: > X-St-Vmsmail-To: > X-Vms-To: There is no such thing as the "X mail system." There are mailers that run under the X Window System (xmh, for instance) but they have nothing to do with these headers. The "X-" prefix is specified by the mail standards (RFC822, I think) for any header which is not official. In other words, if your mailer wants to put a header on a message that isn't a supported header, it should preceed the name by "X-." You may still choose to argue that the headers are useless, but I would disagree in at least some cases (X-Mailer, for instance). However, any mail user interface worth its salt will hide these headers unless you ask to see them, so it really is a moot point. John Diamant Software Development Environments Hewlett-Packard Co. ARPA Internet: diamant@hpfclp.sde.hp.com Fort Collins, CO UUCP: {hplabs,hpfcla}!hpfclp!diamant
soley@ontenv.UUCP (Norman S. Soley) (08/08/88)
There's been a great deal of carping about the X-Mailer header that's
floating around. I'll grant that right now these things are about
as useful as a Royal Commission. But in time...
One of the things that X-headers can be used for is to implement special
features such as confirmation of reciept, encrypted mail, various content
types and so on (please don't assume that I'm suggesting that all of these
things are appropriate for the net, some, like confirmation of recipt,
most definitely aren't, but I needed some examples).
It's a reality that some of these features are going to end up being
implemented differently in different user agents, at least until the dust
settles. So the only way for a recieving ua to deal with a X-feature that could
have multiple interpretations is to know what ua generated the message.
So presto, by combining the X-Mailer and X-feature headers you're sure
to get the right interpretation. Eventually of course RFC-822 will get
supercede/updated to handle new features and then we can lose the
X-Mailer stuff.
As for the net.bad.citizenry of Portal's X-Orgin header... Well my
Mother always told me If you can't say anything nice...
--
Norman Soley - Data Communications Analyst - Ontario Ministry of the Environment
UUCP: utgpu!ontmoh!------------\ VOICE: +1 416 323 2623
{attcan,utzoo}!lsuc!ncrcan!ontenv!norm
"witty saying not available due to writers strike"