Date: Wed, 12 Feb 2003 02:13:01 +0100 From: Brad Knowles <brad.knowles@skynet.be> To: Rahul Siddharthan <rsidd@online.fr> Cc: Brad Knowles <brad.knowles@skynet.be>, freebsd-chat@freebsd.org Subject: Re: Email push and pull (was Re: matthew dillon) Message-ID: <a05200f36ba6f4def1e78@[10.0.1.2]> In-Reply-To: <20030211144057.GA2382@papagena.rockefeller.edu> References: <20030211032932.GA1253@papagena.rockefeller.edu> <a05200f2bba6e8fc03a0f@[10.0.1.2]> <20030211144057.GA2382@papagena.rockefeller.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
At 9:40 AM -0500 2003/02/11, Rahul Siddharthan wrote:
> But I can now block known spammers from even trying to connect,
> because they can no longer relay their mail and thus can't hide their
> tracks.
Not true. They could create messages to be picked up anywhere in
the world, and then bombard you with notices every second.
There has to be an additional level of authentication built into
that system which is not typically present in mail systems today.
Moreover, not only do you need to have virtually unbreakable proof
between the client and the system, you also need to have virtually
unbreakable proof between the system and the recipient. Both are
easily subverted.
Moreover, if we had this level of authentication built into the
existing mail system, we could improve a lot more things a lot faster
than by trying to completely change how e-mail works across the
entire Internet.
> Equally important, the law can catch up with the spammers because they
> can't hide their tracks.
Again, not true. See above. This proposal *may* create a
situation where this sort of thing might exist, but there's a lot
more that would need to be added before you could be virtually
certain.
> One way to transition to a new system would be for mailservers to
> support both systems for a while, and indicate their support by their
> HELO greeting. Perhaps some indication can also be put in the MX
> records.
Synchronous meta-data updates are the #1 kill for most mail
systems today. You don't improve this situation by making the
messages/notices smaller, more frequent, and then tacking on a
secondary transmission channel.
I would have thought people would have learned their lesson with ftp.
> Once a new system is in place, and supported by the big guys (sendmail
> and Microsoft would be enough), I suspect transition would be pretty
> fast.
Not true. First off, Microsoft would never support the same
standard as everyone else, unless everyone else adopted the Microsoft
standard.
Thinking about this some more, you're basically talking about
single-instance message store, a topic that Nick and I discussed in
depth for my talk at LISA 2000. This is fundamentally unscalable,
and places many orders of magnitude more requirements for reliability
on the system than are in place today.
> Look how quickly the world got rid of open relays: back in 1996
> nearly every mail server was an open relay, now the spammers have a
> hard time finding one.
Not at all. The number of open relays may be going down, but
spammers can still easily find enough to do the damage, and that's
for the people that actually subscribe to the appropriate open-relay
blacklists.
--
Brad Knowles, <brad.knowles@skynet.be>
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
-Benjamin Franklin, Historical Review of Pennsylvania.
GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?a05200f36ba6f4def1e78>