ms outlook replyers

ms outlook replyers

[Nick Papadonis]

Anyone have trouble with people using MS Outlook replying to your
emails?

For some reason (even though I have user-mail-address,
user-host-address set) when they reply, it uses my local machines
hostname and username.  This is incorrect.

Anyway to fix this, or am I doing something wrong?

-- 
- Nick

Re: ms outlook replyers

[Kai Großjohann]

On 12 Dec 2000, Nick Papadonis wrote:

> For some reason (even though I have user-mail-address,
> user-host-address set) when they reply, it uses my local machines
> hostname and username.  This is incorrect.

Maybe Outlook uses the Sender header?  Or the From_ line?

If it's the Sender header, you could try this:
(add-to-list 'message-syntax-checks '(sender . disabled))

kai
-- 
The arms should be held in a natural and unaffected way and never
be conspicuous. -- Revised Technique of Latin American Dancing

Re: ms outlook replyers

[Stainless Steel Rat]

* Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann)  on Tue, 12 Dec 2000
| Maybe Outlook uses the Sender header?  Or the From_ line?

Lookout! incorrectly uses the Sender header, that I can confirm.  Dunnow
about SMTP envelope stuff.
-- 
Rat <ratinox@peorth.gweep.net>    \ Do not use Happy Fun Ball on concrete.
Minion of Nathan - Nathan says Hi! \ 
PGP Key: at a key server near you!  \ 

Re: ms outlook replyers

[Colin Walters]

Nick Papadonis <nick@coelacanth.com> writes:

> Anyone have trouble with people using MS Outlook replying to your
> emails?
> 
> For some reason (even though I have user-mail-address,
> user-host-address set) when they reply, it uses my local machines
> hostname and username.  This is incorrect.
> 
> Anyway to fix this, or am I doing something wrong?

Yes, some applications use the Sender: address as the Reply-to field.
I believe this was discussed previously on this list.  I think what it
came down to is that there is no consensus on what the Sender: field
is really for.  

I eventually just changed mine to be my email address.

Re: ms outlook replyers

[Colin Walters]

Nick Papadonis <nick@coelacanth.com> writes:

[ modifying the Sender: header ]

> How did you do this?  What variable is it?  I was looking through
> customize but couldn't find anyting.

You could probably do it by adding an entry to
`message-default-headers', but I just added an entry to my
gnus-posting-styles.

Re: ms outlook replyers

[Nick Papadonis]

How did you do this?
What variable is it?  I was looking through customize but couldn't
find anyting.

>>>>> "Colin" == Colin Walters <walters@cis.ohio-state.edu> writes:

    > Nick Papadonis <nick@coelacanth.com> writes:
    >> Anyone have trouble with people using MS Outlook replying to
    >> your emails?
    >> 
    >> For some reason (even though I have user-mail-address,
    >> user-host-address set) when they reply, it uses my local
    >> machines hostname and username.  This is incorrect.
    >> 
    >> Anyway to fix this, or am I doing something wrong?

    > Yes, some applications use the Sender: address as the Reply-to
    > field.  I believe this was discussed previously on this list.  I
    > think what it came down to is that there is no consensus on what
    > the Sender: field is really for.

    > I eventually just changed mine to be my email address.


-- 
- Nick

Re: ms outlook replyers

[Stainless Steel Rat]

* Colin Walters <walters@cis.ohio-state.edu>  on Tue, 12 Dec 2000
| Yes, some applications use the Sender: address as the Reply-to field.
| I believe this was discussed previously on this list.  I think what it
| came down to is that there is no consensus on what the Sender: field
| is really for.

RFC 822 is describes Sender explicitly and in detail.
-- 
Rat <ratinox@peorth.gweep.net>    \ Do not taunt Happy Fun Ball.
Minion of Nathan - Nathan says Hi! \ 
PGP Key: at a key server near you!  \ 

Re: ms outlook replyers

[Colin Walters]

Stainless Steel Rat <ratinox@peorth.gweep.net> writes:

> RFC 822 is describes Sender explicitly and in detail.

Yes, but RFC 822 only covers electronic mail.  RFC 1036 describes
different semantics for the Sender header in USENET.

It looks like there have been a number of discussions on this topic
before; a search for "sender header" on the Gnus mailing list archives
seems to reveal most of them, although not the one I remember.

Re: ms outlook replyers

[Justin Sheehy]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Maybe Outlook uses the Sender header?

Yes, Outlook is broken in this regard.

Some current versions of Outlook will reply to the Sender header if it 
is present, ignoring the From header.

No one can claim that this is correct, but one isn't likely to get
Microsoft to change it very quickly in any case.

-Justin

 

Disable generation of "Sender: " by default.

[Per Abrahamsen]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> (add-to-list 'message-syntax-checks '(sender . disabled))

I think this should be default.  Anyone agree?

Let sendmail (or whatever agent the message is delivered to) create
the "Sender: " header, it can probably do a better (and more reliable)
job than Gnus.  The header is not mandatory anyway, according to RFC
822 it is "intended for use when the sender is not the author of the
message, or to indicate who among a group of authors actually sent the
message."  For News it was for some time used for authentication, but
other headers now do that job more reliably.

Re: ms outlook replyers

[Kai Großjohann]

On 13 Dec 2000, Nick Papadonis wrote:
> 
> How did you do this?
> What variable is it?  I was looking through customize but couldn't
> find anyting.

My suggestion was to omit the Sender header, hoping that Outlook would
use the From header if no Sender header is present.

(add-to-list 'message-syntax-checks '(sender . disabled))

kai
-- 
The arms should be held in a natural and unaffected way and never
be conspicuous. -- Revised Technique of Latin American Dancing

Re: Disable generation of "Sender: " by default.

[Jesper Harder]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
> 
> > (add-to-list 'message-syntax-checks '(sender . disabled))
> 
> I think this should be default.  Anyone agree?

It will break cancels and supersedes. 

If you use gnus-posting-styles the From-header doesn't match
(message-make-from) and  message-cancel-news checks the Sender-header
instead. If there is no Sender-header you won't be allowed to cancel
your own messages.

-- 
Jesper Harder

Re: ms outlook replyers

[Stainless Steel Rat]

* Colin Walters <walters@cis.ohio-state.edu>  on Wed, 13 Dec 2000
| Yes, but RFC 822 only covers electronic mail.  RFC 1036 describes
| different semantics for the Sender header in USENET.

Lookout! is an MUA, but not a Usenet client, so RFC 1036 is not applicable.
-- 
Rat <ratinox@peorth.gweep.net>    \ Happy Fun Ball may stick to certain types
Minion of Nathan - Nathan says Hi! \ of skin.
PGP Key: at a key server near you!  \ 

Re: ms outlook replyers

[Dirk Gomez]

So how can I set the sender field? Most people in my organization use
Outlook and seeing my hostname in the reply mail is simply unappealing
;)

cheers dirk

Re: ms outlook replyers

[Kai Großjohann]

On 14 Dec 2000, Dirk Gomez wrote:

> So how can I set the sender field?

Not nice, but works:

(require 'message)
(defadvice message-make-sender (around dirks-sender activate)
  "Always return Dirk's Sender header."
  "what.dirk.wants.for.sender")

Beware, untested.

kai
-- 
A large number of young women don't trust men with beards.  (BFBS Radio)

Re: Disable generation of "Sender: " by default.

[Per Abrahamsen]

Jesper Harder <harder@ifa.au.dk> writes:

> If you use gnus-posting-styles the From-header doesn't match
> (message-make-from) and  message-cancel-news checks the Sender-header
> instead. If there is no Sender-header you won't be allowed to cancel
> your own messages.

Will the cancel be obeyed by the news servers in that case?  

It seems like `gnus-posting-styles' isn't implemented fully.  I'm
certainly not impressed by its doc string.  It can't be customized
either.

I'd suggest disabling the sender generation by default, and let
someone who cares fix `gnus-posting-styles'.

Re: ms outlook replyers

[Alan Shutko]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> On 14 Dec 2000, Dirk Gomez wrote:
> 
> > So how can I set the sender field?
> 
> Not nice, but works:

Also not nice, but tested.  (And used for the last couple years or so.)

(defun message-make-sender ()
  "Return a fake sender address because I don't like it."
  (message-make-address))

-- 
Alan Shutko <ats@acm.org> - In a variety of flavors!
Teachers have class.

Re: Disable generation of "Sender: " by default.

[Jesper Harder]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> Jesper Harder <harder@ifa.au.dk> writes:
> 
> > If there is no Sender-header you won't be allowed to cancel your own
> > messages.
> 
> Will the cancel be obeyed by the news servers in that case?

Yes, because Gnus uses the From-header of the original message in the
cancel.

> I'd suggest disabling the sender generation by default, and let
> someone who cares fix `gnus-posting-styles'.

Or maybe disable it only for mail, which won't break anything.

-- 
Jesper Harder

Re: ms outlook replyers

[Nick Papadonis]

Didn't work for me.
I just had:
X-Authentication-Warning: wash-382.res.umass.edu: nick set sender to nick@coelacanth.com using -f
And no Sender field in the header.

Xemacs 21.1 GNUS v5.8.8

Anyway, why isn't this configurable in the customize menus?

- Nick

>>>>> "Alan" == Alan Shutko <ats@acm.org> writes:

    > Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
    >> On 14 Dec 2000, Dirk Gomez wrote:
    >> 
    >> > So how can I set the sender field?
    >> 
    >> Not nice, but works:

    > Also not nice, but tested.  (And used for the last couple years
    > or so.)

    > (defun message-make-sender () "Return a fake sender address
    > because I don't like it."  (message-make-address))

    > -- Alan Shutko <ats@acm.org> - In a variety of flavors!
    > Teachers have class.


-- 
- Nick

Re: ms outlook replyers

[Kai Großjohann]

On 14 Dec 2000, Nick Papadonis wrote:

> Didn't work for me.  I just had: X-Authentication-Warning:
> wash-382.res.umass.edu: nick set sender to nick@coelacanth.com using
> -f And no Sender field in the header.

Try (setq message-sendmail-f-is-evil t) -- does that help?

> Anyway, why isn't this configurable in the customize menus?

Sigh.  Lars thinks that Gnus can do a perfect job of computing the
Sender header, every time.  So no customization would be needed.  This
might be true for news (I cannot say since I don't know), but I
believe that it is not true for mail.  In fact, the computer cannot
know who is the Sender of a mail message, according to the RFC822
meaning of Sender.

kai
-- 
A large number of young women don't trust men with beards.  (BFBS Radio)

Re: ms outlook replyers

[Stainless Steel Rat]

* Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann)  on Thu, 14 Dec 2000
| Sigh.  Lars thinks that Gnus can do a perfect job of computing the
| Sender header, every time.  So no customization would be needed.  This
| might be true for news (I cannot say since I don't know),

It does, assuming that the system itself has been correctly configured.
All of the examples that I remember are really examples of changing the
system configuration out from underneath running programs, rendering
system-name invalid.

| but I believe that it is not true for mail.  In fact, the computer cannot
| know who is the Sender of a mail message, according to the RFC822 meaning
| of Sender.

That is true.  If a Sender header is required for a mail message then it
should (must) be created manually by the agent that is sending the message.
-- 
Rat <ratinox@peorth.gweep.net>    \ Caution: Happy Fun Ball may suddenly
Minion of Nathan - Nathan says Hi! \ accelerate to dangerous speeds.
PGP Key: at a key server near you!  \ 

Re: ms outlook replyers

[Nick Papadonis]

>>>>> "Kai" == Kai Großjohann <Kai.Grossjohann@CS.Uni-Dortmund.DE> writes:
    > Try (setq message-sendmail-f-is-evil t) -- does that help?

This is what I receive when I do that.  Still doesn't work.

X-From-Line: nick  Thu Dec 14 23:23:25 2000
Return-Path: <nick>
Received: (from nick@localhost)
	by wash-382.res.umass.edu (8.9.3/8.9.3) id XAA21914;
	Thu, 14 Dec 2000 23:23:25 -0500
Sender: nick@wash-382.res.umass.edu
To: nick@wash-382.res.umass.edu
Subject: tnew 
From: Nick Papadonis <nick@coelacanth.com>
Date: 14 Dec 2000 23:23:25 -0500
X-Gnus-Mail-Source: directory:~/incoming/
Message-ID: <m366km2tz6.fsf@wash-382.res.umass.edu>
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Channel Islands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-SpamBouncer: 1.3 (11/02/00)
X-SBPass: No Freemail Filtering
X-SBClass: OK
Lines: 4
Xref: wash-382.res.umass.edu mail.misc:729


-- 
- Nick

Re: ms outlook replyers

[Kai Großjohann]

On 15 Dec 2000, Nick Papadonis wrote:

>>>>>> "Kai" == Kai Großjohann <Kai.Grossjohann@CS.Uni-Dortmund.DE>
>>>>>> writes:
>     > Try (setq message-sendmail-f-is-evil t) -- does that help?
> 
> This is what I receive when I do that.  Still doesn't work.

What do you mean `receive'?  What you are showing doesn't look like an
error message.

Hm.

Oh, you're saying that the From and Sender header are different.  Hm.
What's the value of message-deletable-headers?  Check with C-h v.

kai
-- 
A large number of young women don't trust men with beards.  (BFBS Radio)

Re: Disable generation of "Sender: " by default.

[Per Abrahamsen]

Jesper Harder <harder@ifa.au.dk> writes:

> Or maybe disable it only for mail, which won't break anything.

Yep.  But as far as I can see, I don't get a "Sender: " in my mail,
and I don't understand why.  'message-syntax-checks' is nil, and I'm
posting from another machine than my mail address.  So I do not dare
touch the code.

Re: Disable generation of "Sender: " by default.

[Harry Putnam]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> Jesper Harder <harder@ifa.au.dk> writes:
> 
> > Or maybe disable it only for mail, which won't break anything.
> 
> Yep.  But as far as I can see, I don't get a "Sender: " in my mail,
> and I don't understand why.  'message-syntax-checks' is nil, and I'm
> posting from another machine than my mail address.  So I do not dare
> touch the code.

There have been long and bitter discussions about the `sender' line
here and on gnu.emacs.gnus  going back a year or two,  back then it
was generally presented as a `bad' thing to disable generation of
sender.

Respected posters advocated against hotly.  But apparently things have
changed in the way mail is processed since then and the sender line
doesn't play the same role it once did.

Not so long ago someone advocating: 
(setq message-syntax-checks '((sender . disabled))), openly would 
have been called to task...

What has changed for this attitude to have relaxed or disappeared?

Re: Disable generation of "Sender: " by default.

[Stainless Steel Rat]

* Per Abrahamsen <abraham@dina.kvl.dk>  on Fri, 15 Dec 2000
| Yep.  But as far as I can see, I don't get a "Sender: " in my mail,
| and I don't understand why.  'message-syntax-checks' is nil, and I'm
| posting from another machine than my mail address.

For mail, a Sender header is manually created by the agent that submits the
message to the network when he is not the person who originated the
message.  Ie, if I write a message and you send it for me, my mailbox goes
in the From header and your mailbox goes in the Sender header.  Gnus does
the right thing by not automatically generating a Sender header for mail.
Lookout! does the wrong thing by using Sender for anything.

News is different.  For news, the Sender header is generated when your From
address does not match your local posting address.  RFC 1036 assumes that
if you are posting from a particular host then you can receive mail at that
host.  The assumption is invalid, but it is in the standard.
-- 
Rat <ratinox@peorth.gweep.net>    \ Happy Fun Ball contains a liquid core,
Minion of Nathan - Nathan says Hi! \ which, if exposed due to rupture, should
PGP Key: at a key server near you!  \ not be touched, inhaled, or looked at.

Re: ms outlook replyers

[Paul Jarc]

Nick Papadonis <nick@coelacanth.com> writes:
> >>>>> "Alan" == Alan Shutko <ats@acm.org> writes:
>     > (defun message-make-sender () "Return a fake sender address
>     > because I don't like it."  (message-make-address))
> 
> Didn't work for me.
> I just had:
> X-Authentication-Warning: wash-382.res.umass.edu: nick set sender to nick@coelacanth.com using -f
> And no Sender field in the header.

Do you mean no Sender: was in the mail you received?  That's the
desired result, isn't it?


paul

Re: Disable generation of "Sender: " by default.

[Alan Shutko]

Harry Putnam <reader@newsguy.com> writes:

> What has changed for this attitude to have relaxed or disappeared?

Back then, the most that broke when you send out a sender address you
didn't want was some mailing list software, and the usual response was
"fix the ml software".  Now there's a huge installed base of people
using Outlook, and we know hoping for it to be fixed is well-nigh
hopeless.  (As shown by Gnus's tools to hide Outlook's nationalized
"Re"s.  Ever notice that you can put any two character code followed
by a colon starting a subject and Outlook will strip it?  Argh.)


-- 
Alan Shutko <ats@acm.org> - In a variety of flavors!
Your lucky number has been disconnected.

Re: ms outlook replyers

[Nick Papadonis]

(Message-ID Date Lines)

>>>>> "Kai" == Kai Großjohann <Kai.Grossjohann@CS.Uni-Dortmund.DE> writes:

    > Oh, you're saying that the From and Sender header are different.
    > Hm.  What's the value of message-deletable-headers?  Check with
    > C-h v.

Re: ms outlook replyers

[Kai Großjohann]

On 15 Dec 2000, Nick Papadonis wrote:
>>>>>> "Kai" == Kai Großjohann <Kai.Grossjohann@CS.Uni-Dortmund.DE>
>>>>>> writes:
> 
>     > Oh, you're saying that the From and Sender header are
>     > different.  Hm.  What's the value of
>     > message-deletable-headers?  Check with C-h v.
> 
> (Message-ID Date Lines)

So that wasn't it.  Sorry.  I'm out of ideas.

kai
-- 
A large number of young women don't trust men with beards.  (BFBS Radio)

Re: Disable generation of "Sender: " by default.

[Kai Großjohann]

On 15 Dec 2000, Stainless Steel Rat wrote:

> News is different.  For news, the Sender header is generated when
> your From address does not match your local posting address.  RFC
> 1036 assumes that if you are posting from a particular host then you
> can receive mail at that host.  The assumption is invalid, but it is
> in the standard.

Hm.  The only difference I can find between the description in RFC 822
and the one in 1036:

* 1036 says that the Sender header should be verified by software at
  the submitting host.  822 says nothing about verification by
  software.

* 1036 talks about the `entity' submitting the message into the
  network, 822 talks about the `agent (person, system or process)'.

Gack.  I still feel that the intent of the Sender header is the same
in RFC 822 and 1036, but it's only a feeling.

I don't know what Son-of-1036 says, and there is another document for
this, I think, which I also don't know.  I wish that Usenet was better
standardized.  As it is, we have to play by ear which is not good.

kai
-- 
A large number of young women don't trust men with beards.  (BFBS Radio)

Re: Disable generation of "Sender: " by default.

[Stainless Steel Rat]

* Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann)  on Sun, 17 Dec 2000
| Gack.  I still feel that the intent of the Sender header is the same
| in RFC 822 and 1036, but it's only a feeling.

RFC 822: Sender is the agent responsibible for submitting a message to the
network when he is someone other than the originator of the message.  Ie,
if you write a message and I send it, you are the originator (From) and I
am the submitter (Sender).  Sender is generated by hand -- never software
-- when more than one agent is involved.  It exists only to denote this fact.

RFC 1036: Sender exists as a crude authentication mechanism.  Sender is the
authenticated (by software) local mailbox of the originator of the message
when that mailbox differs from the one in the originator header.  Ie, if I
am on, say, skuld.gweep.net w/ login 'ratinox' and I post a news article
with 'grumble@bogus.com' as my From header, then some program somewhere
must generate 'Sender: ratinox@skuld.gweep.net'.

They are not even vaguely close to each other.  My opinion is that RFC
1036's Sender is badly broken, because it utterly fails to accomplish its
goal.  RFC 822's Sender does exactly what it is intended to do, no more, no
less.
-- 
Rat <ratinox@peorth.gweep.net>    \ If Happy Fun Ball begins to smoke, get
Minion of Nathan - Nathan says Hi! \ away immediately. Seek shelter and cover
PGP Key: at a key server near you!  \ head.

Re: Disable generation of "Sender: " by default.

[Kai Großjohann]

On 17 Dec 2000, Stainless Steel Rat wrote:

> They are not even vaguely close to each other.  My opinion is that
> RFC 1036's Sender is badly broken, because it utterly fails to
> accomplish its goal.  RFC 822's Sender does exactly what it is
> intended to do, no more, no less.

But one of the examples given in RFC 1036 is exactly the same as the
example given in RFC 822!

/----
|     For example, if John Smith is visiting CCA and wishes to post a
|     message to the network, using friend Sarah Jones' account, the
|     message might read:
| 
|               From: smith@ucbvax.Berkeley.EDU (John Smith)
|               Sender: jones@cca.COM (Sarah Jones)
\----

This is just like the secretary example in RFC 822.  Hence, there is
at least some overlap between the intentions of the Sender header.
(The RFC 1036 Sender header can be used in more situations, of
course.)

Hm.

Oh, RFC 822 kinda assumes that the secretary was typing the message,
too, whereas here, it's clear that Sarah is NOT typing the message.
Hm.  But that's not the big deal, is it?

kai
-- 
A large number of young women don't trust men with beards.  (BFBS Radio)

Re: Disable generation of "Sender: " by default.

[Russ Allbery]

Stainless Steel Rat <ratinox@peorth.gweep.net> writes:

> RFC 1036: Sender exists as a crude authentication mechanism.  Sender is
> the authenticated (by software) local mailbox of the originator of the
> message when that mailbox differs from the one in the originator header.
> Ie, if I am on, say, skuld.gweep.net w/ login 'ratinox' and I post a
> news article with 'grumble@bogus.com' as my From header, then some
> program somewhere must generate 'Sender: ratinox@skuld.gweep.net'.

> They are not even vaguely close to each other.  My opinion is that RFC
> 1036's Sender is badly broken, because it utterly fails to accomplish
> its goal.  RFC 822's Sender does exactly what it is intended to do, no
> more, no less.

I'm fairly sure USEFOR drops the entire broken idea of a Sender header,
thankfully.

LISTSERV still responds to the Sender header rather than the From header,
checks the Sender header to see if you can post to a list, and if you try
to subscribe to a mailing list, subscribes the content of your Sender
header instead of your From address.  That's why I turned off Sender in my
Gnus settings; it's easier to do that than it is to try to argue with Eric
Thomas that his software is doing something really stupid.

-- 
Russ Allbery (rra@stanford.edu)             <http://www.eyrie.org/~eagle/>

Re: Disable generation of "Sender: " by default.

[Per Abrahamsen]

Russ Allbery <rra@stanford.edu> writes:

> That's why I turned off Sender in my Gnus settings;

But why did Gnus generate it for you in the first place?  It doesn't
do that for me with mail.

Re: Disable generation of "Sender: " by default.

[Russ Allbery]

Per Abrahamsen <abraham@dina.kvl.dk> writes:
> Russ Allbery <rra@stanford.edu> writes:

>> That's why I turned off Sender in my Gnus settings;

> But why did Gnus generate it for you in the first place?  It doesn't do
> that for me with mail.

Gnus used to always generate Sender headers if your From header didn't
match its concept of your e-mail address (which it digs out of the
hostname plus the local username or some such thing).  It's possible that
at some point this was fixed to be less aggressive about generating a
Sender header, but since I never want a Sender header on any of my mail
under any circumstances, I left the setting.

-- 
Russ Allbery (rra@stanford.edu)             <http://www.eyrie.org/~eagle/>

Re: Disable generation of "Sender: " by default.

[Toby Speight]

0> In article <m37l4zau28.fsf@peorth.gweep.net>,
0> Stainless Steel Rat <URL:mailto:ratinox@peorth.gweep.net> ("Rat") wrote:

Rat> RFC 822: Sender is the agent responsibible for submitting a
Rat> message to the network when he is someone other than the
Rat> originator of the message.  Ie, if you write a message and I
Rat> send it, you are the originator (From) and I am the submitter
Rat> (Sender).  Sender is generated by hand -- never software --
Rat> when more than one agent is involved.  It exists only to
Rat> denote this fact.
Rat>
Rat> RFC 1036: Sender exists as a crude authentication mechanism.
Rat> Sender is the authenticated (by software) local mailbox of the
Rat> originator of the message when that mailbox differs from the one
Rat> in the originator header.  Ie, if I am on, say, skuld.gweep.net
Rat> w/ login 'ratinox' and I post a news article with
Rat> 'grumble@bogus.com' as my From header, then some program
Rat> somewhere must generate 'Sender: ratinox@skuld.gweep.net'.
Rat>
Rat> They are not even vaguely close to each other.  My opinion is
Rat> that RFC 1036's Sender is badly broken, because it utterly fails
Rat> to accomplish its goal.  RFC 822's Sender does exactly what it is
Rat> intended to do, no more, no less.

This is pretty much how I see it, too - except that it's impossible
in general for the software to know how to find the user's "local"
mailbox.  So it's not reliable as a deliverable address.  This was
really frustrating when trying to cancel a message sent from a
100-machine workstation cluster, and only being able to do so from
the exact same machine, despite the fact that they all shared the
same filesystem image and MX host. :-(

Re: Disable generation of "Sender: " by default.

[Stainless Steel Rat]

* Toby Speight <streapadair@gmx.net>  on Wed, 20 Dec 2000
| This is pretty much how I see it, too - except that it's impossible
| in general for the software to know how to find the user's "local"
| mailbox.

In the case of RFC 822 this is correct.

In the case of RFC 1036... it is not an RFC 822 mailbox but user@local.fqdn.
Which is not reliable.  Which is why I say that RFC 1036 and son of are
broken when it comes to Sender.
-- 
Rat <ratinox@peorth.gweep.net>    \ Do not taunt Happy Fun Ball.
Minion of Nathan - Nathan says Hi! \ 
PGP Key: at a key server near you!  \