On Mon, 19 Apr 1999 07:13:57 +0100 (BST), Jeffrey Goldberg wrote:
> * External dial-up users who insist on doing their email transport
> directly instead of via their ISP. There is no reason why those
> using dynamic IP addresses via should be doing that.
There is no reason they shouldn't, either. Give me a good technical
reason for it and I'll concede.
> their ISPs mail hubs. The better ISPs force such behaviour.
Worse, not better.
>If I am wrong about that, I would very much like to here your reasons
>(off this list) for going direct from a dynamically allocated address.
How about someone using one of the several free Dyndns projects to run a
mailing list? Or a technical person who simply wishes control over their
mail or wants to do the right thing and have their machine deliver the mail
instead of bogging down the ISP's machine?
>far, I've just said that it is wrong to block these and it is wrong for
>ISPs to prevent them, you have not given any reason why you might want to
>go direct to arbitrary mail exchangers on the net.
Because it is a shift in blame. Now it is guilty until proven
innocent... or rather, until one can purchase the privledge, instead of the
other way around. By removing such a large segment on no other grounds than
the fact that they are on a dynamic IP also calls into question any
possibility of common carrier status on the side of the ISPs. Blocking known
spam sites is different as spam is the electronic equivolent of mail fraud.
>(I can imagine why you might want to go via a specific relay for all your >mail which is not your ISPs). If your only reason is that you don't trust >your ISP... well switch.
The fact that there is no technical reason and possible negative legal
ramifications is enough. I'm not even talking about my personal bias here.
>Admittedly, there is an argument against DUL and DSSL (and in favor of
>MAPS) which is that the former is doing the ISPs job for them, while only
>the latter is actually putting pressure on ISPs which host spammers.
No, the former is declaring a large portion of the internet guilty until
proven otherwise on no good grounds at all.
>At least the problem that you have with Cranfield and sites
>like it is one of policy questions and not of genuine misconfiguration.
I would consider it a misconfiguration in the same vein as leaving a mail
server wide open to relay. That, too, could be considered a "policy"
decision where the individual chooses not to prevent relaying. Yet that is
considered a misconfiguration.
- --
Steve C. Lamb | I'm your priest, I'm your shrink, I'm your
ICQ: 5107343 | main connection to the switchboard of souls.
- -------------------------------+---------------------------------------------
--
*** Exim information can be found at
http://www.exim.org/ ***
