Mailing List Archive

On Thu, 2006-08-10 at 10:57 +0200, torsten@gcrud.org wrote:
> just a quick questions:
>
> Are there any plans to support this:
>
> http://tools.ietf.org/wg/lemonade/draft-maes-lemonade-http-binding-04.txt
>
> Would anyone object to supporting the SMTP HTTP binding described in there?

in my opinion it's premature. notice that this is an individual
submission, and not part of LEMONADE's time table or charter, so it may
never become a proposed standard at all. and it's not as if there's not
enough _real_ LEMONADE RFC's which need implementing already! :-)

but if you want to experiment with it, don't let me stop you.

http://www.ietf.org/html.charters/lemonade-charter.html
--
Kjetil T.



--
## List details at http://www.exim.org/mailman/listinfo/exim-dev Exim details at http://www.exim.org/ ##

On 8/10/06 1:57 AM, "torsten@gcrud.org" <torsten@gcrud.org> wrote:

> Are there any plans to support this:
>
> http://tools.ietf.org/wg/lemonade/draft-maes-lemonade-http-binding-04.txt
>
> Would anyone object to supporting the SMTP HTTP binding described in there?

Interesting stuff. There might be interest (in this chair) when the
proposal is further along. (But by then, the occupant of this chair could
well be retired.)

As Kjetil says, it seems premature now.

I assume that fact that the document has expired is merely a sign that it
hasn't quite yet been resubmitted rather than that interest had flagged.

It seems also as if the IMAP "side" is more important (and more developed)
than the SMTP side. And since suitable IMAP servers can be used as
submission tools, it's not clear that the SMTP side is as important.

--John



--
## List details at http://www.exim.org/mailman/listinfo/exim-dev Exim details at http://www.exim.org/ ##

On Thu, 10 Aug 2006, torsten@gcrud.org wrote:
>
> Would anyone object to supporting the SMTP HTTP binding described in there?

I would have thought this would be a job for a separate program.

More interesting would be support for BURL, which entails support for
CHUNKING.

Tony.
--
<fanf@exim.org> <dot@dotat.at> http://dotat.at/ ${sg{\N${sg{\
N\}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}\
\N}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}

--
## List details at http://www.exim.org/mailman/listinfo/exim-dev Exim details at http://www.exim.org/ ##

> I would have thought this would be a job for a separate program.

Are you referring to something like a mod_lemonade for Apache httpd which
that would just talk back to a plain SMTP server? Well, sounds logical as
it would allow you to leverage all the HTTP server stuff which is already
there in an Apache httpd or any other well established http server.

Regards,
Torsten

> On Thu, 10 Aug 2006, torsten@gcrud.org wrote:
>>
>> Would anyone object to supporting the SMTP HTTP binding described in
>> there?
>
> I would have thought this would be a job for a separate program.
>
> More interesting would be support for BURL, which entails support for
> CHUNKING.
>
> Tony.
> --
> <fanf@exim.org> <dot@dotat.at> http://dotat.at/ ${sg{\N${sg{\
> N\}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}\
> \N}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}
>


--
## List details at http://www.exim.org/mailman/listinfo/exim-dev Exim details at http://www.exim.org/ ##

> I assume that fact that the document has expired is merely a sign that it
> hasn't quite yet been resubmitted rather than that interest had flagged.

IMO there should be tons of people in need for something like that.
Basically everyone in a corporate firewall / http proxy environment who
would still like to use a reasonable MUA instead of webmail. But I guess
these people have never been asked and that very practical issue is not
known in the academic world.

Of course it will still require some MUAs to pick this up and support it.
But I cannot see why it shouldn't be possible to build support for this
into Mozilla Thunderbird for example.

Of course if's kind of a chicken and egg problem. The server service
doesn't make sense if you don't have clients that support the protocol and
you will not see too much client development unless there are servers
available and at least some popular email providers jump on the train.

But the proposed standard we discuss here would at least be a basis for an
open implementation versus any proprietary approaches that would tie the
email provider and the MUA together.

> And since suitable IMAP servers can be used as
> submission tools, it's not clear that the SMTP side is as important.

Could you help me out here, please?

Are you saying that IMAP has extensions in place / under development that
would allow me to *send* an email through IMAP rather than an SMTP server?
Wouldn't that obsolete an SMTP server as as Exim sooner or later?

Regards,
Torsten


> On 8/10/06 1:57 AM, "torsten@gcrud.org" <torsten@gcrud.org> wrote:
>
>> Are there any plans to support this:
>>
>> http://tools.ietf.org/wg/lemonade/draft-maes-lemonade-http-binding-04.txt
>>
>> Would anyone object to supporting the SMTP HTTP binding described in
>> there?
>
> Interesting stuff. There might be interest (in this chair) when the
> proposal is further along. (But by then, the occupant of this chair could
> well be retired.)
>
> As Kjetil says, it seems premature now.
>
> I assume that fact that the document has expired is merely a sign that it
> hasn't quite yet been resubmitted rather than that interest had flagged.
>
> It seems also as if the IMAP "side" is more important (and more developed)
> than the SMTP side. And since suitable IMAP servers can be used as
> submission tools, it's not clear that the SMTP side is as important.
>
> --John
>
>
>
> --
> ## List details at http://www.exim.org/mailman/listinfo/exim-dev Exim
> details at http://www.exim.org/ ##
>


--
## List details at http://www.exim.org/mailman/listinfo/exim-dev Exim details at http://www.exim.org/ ##

Quoting torsten@gcrud.org:

> IMO there should be tons of people in need for something like that.
> Basically everyone in a corporate firewall / http proxy environment who
> would still like to use a reasonable MUA instead of webmail. But I guess

Sure, but most sysadmins wouldn't want their users to use some email
client get mail from the outside without central malware scanning. A
person with some experience could create a tunnel trough the proxy, but
such a person usually knows what he's doing, in contrast to a user which
simply has to enter the proxy address (if not given by the system config).

> Of course if's kind of a chicken and egg problem. The server service
> doesn't make sense if you don't have clients that support the protocol and

Wouldn't it make more sense to do this in a proxy gateway, which talks
to the real smtp/imap servers, instead of implementing lots of code into
every mail server software?

> Are you saying that IMAP has extensions in place / under development that
> would allow me to *send* an email through IMAP rather than an SMTP server?

http://www.courier-mta.org/imap/smap.html
As with many extensions, it is not widely used (I don't know of any
software besides Courier itself and the corresponding client Cone that
supports it).

> Wouldn't that obsolete an SMTP server as as Exim sooner or later?

Courier itself has already a MTA.
A simple IMAP server implementation would hand over the submitted
message to the local MTA. So, no, MTAs are still needed and they do much
more than only being a smarthost. This would be more or less a
replacement for the submission/msa protocol. Inter-MTA traffic will most
probably never use something like SMAP.


--
## List details at http://www.exim.org/mailman/listinfo/exim-dev Exim details at http://www.exim.org/ ##

On Sat, 2006-08-12 at 17:06 +0200, torsten@gcrud.org wrote:
> > I assume that fact that the document has expired is merely a sign that it
> > hasn't quite yet been resubmitted rather than that interest had flagged.
>
> IMO there should be tons of people in need for something like that.
> Basically everyone in a corporate firewall / http proxy environment who
> would still like to use a reasonable MUA instead of webmail. But I guess
> these people have never been asked and that very practical issue is not
> known in the academic world.

why would the firewall administrator accept mail traffic tunneled over
HTTP any more than mail traffic sent over port 587? I would hope that
network usage policy is not tied to physical port numbers, but rather on
an analysis of information leakage threat and similar.

this trend to use port 80 for everything is not making anything more
secure. it just means firewall hardware needs to be more and more beefy
in order to do advanced content parsing to distinguish plain browsing
from bad traffic.
--
Kjetil T.



--
## List details at http://www.exim.org/mailman/listinfo/exim-dev Exim details at http://www.exim.org/ ##