[IP] Spam Controls Imperil E-Mail Reliability
------ Forwarded Message
From: Suresh Ramasubramanian <suresh@xxxxxxxxxx>
Organization: -ENOENT
Date: Sat, 26 Feb 2005 07:06:44 +0530
To: <dave@xxxxxxxxxx>
Cc: <GLIGOR1@xxxxxxx>, <netwriter@xxxxxx>
Subject: Re: [IP] Spam Controls Imperil E-Mail Reliability
On Fri, 2005-02-25 at 17:53 -0500, David Farber wrote:
>
> From: <GLIGOR1@xxxxxxx>
> Spam Controls Imperil E-Mail Reliability
>
That article kind of overstates things. Bad spam filtering? Sure. But
saying that spam filtering imperils email reliablity is wrong, and does
no credit at all to several people working at large ISPs, who walk a
continuous tightrope between rejecting spam inbound to their users
mailboxes and blocking legitimate email.
Or, if you choose, ISPs could shut off all spam filtering, and as some
people advocate, dump all the mail in users' mailboxes and allow them to
sort it out. In which case
1. The users would be buried in a sea of spam
2. Technically less savvy users would not be able to filter it out
3. Once it is delivered and stored at the ISP, costs for bandwidth,
storage etc have been incurred - a fraction of a cent per spam, millions
of spams a day. Guess where these costs will eventually be passed on?
It would have been far better if this article was a call for responsible
spam filtering, that kept in mind the ISP's main job of delivering email
that their users want, to their mailbox.
In fact I'll be speaking on a couple of panels that discuss exactly this
(responsible spam filtering, of both inbound and outbound spam) at MAAWG
(www.maawg.org) from march 1-3 in San Diego. MAAWG is an grouping of
abuse desk managers from several ISPs around the world, and so far as I
can see, is about the only conference of its kind that attracts a bunch
of operationally relevant people - abuse desk and mail system
administrators, my peers at other ISPs, as opposed to the usual mix of
product vendors and marketing folk that you can find at most other ISP
oriented antispam conferences that I've seen in the United States.
There are other conferences too, more academic in nature and slightly
less concerned with the implementation of proposed solutions so that
they'll scale to a large mail system millions of users in size, but
that's a different story altogether :)
Speaking of antispam conferences, I'm just back from organizing an
APCAUCE conference at Kyoto, during APRICOT 2005. The highlight of this
was a panel featuring Dave Crocker (author of BATV and CSV), Jim Fenton
of Cisco (author of the identified internet mail proposal) and Meng Wong
(author of SPF), the focus of which was to discuss these proposals from
an operator perspective as opposed to the purely technological view
you'd get when discussing these at an IETF. More about this when I get
the presentations and conference minutes uploaded.
regards
-suresh
------ End of Forwarded Message
-------------------------------------
You are subscribed as roessler@xxxxxxxxxxxxxxxxxx
To manage your subscription, go to
http://v2.listbox.com/member/?listname=ip
Archives at: http://www.interesting-people.org/archives/interesting-people/