[svlug] Email length

Derek J. Balling dredd at megacity.org
Mon May 15 16:36:54 PDT 2000

At 04:27 PM 5/15/00 -0700, J C Lawrence wrote:
>If you consider the MTA as discrete from the feature of allowing the
>postmaster to implement mail handling policies (which are external
>to the MTA, but the which the MTA supports via
>callouts/filters/whatever), then this whole debate pretty well
>vanishes.  The MTA doesn't dick with message bodies.  The MTA
>supports the functionality of filters and other devices by which,
>among other things, an Admin can contrive to dick with message
>bodies should he so wish.

Agreed. I would (at that point) debate the VALUE of such things. I believe 
an e-mail message is a communication between the sender and the recipient. 
I'd also wonder if, legally, an ISP could legally implement such a filter. 
(Given the ECPA).

> > BUT -- There is a difference between ALTERING the content, and
> > SCANNING the content.
>We were discussing the active removal of active content from
>messages during transit...

Agreed. My point is that I would agree with "The MTA scanning the content 
and deciding to accept or reject the message", but would disagree with "The 
MTA calling some cleansing agent to scrub out portions of the DATA 
block."   While the latter is certainly within the "legal bounds" of what 
an MTA can do, I would contend that - by being called for each and every 
message by the MTA - the filter is acting as a portion of the MTA and not 
as an external apparatus. (To me, if the recipient can't override it, I 
would consider it to be acting on behalf of the MTA).

Interesting that I'm having this very same debate at work as we speak. :)

> > However, there are a few caveats: 1.) There are some (poorly made)
> > MTA's which do not properly handle receiving a 500-series failure
> > AFTER they have delivered the DATA segment.  Depending on the MTA
> > in question, some will not see it as an error, but see it as
> > success and not generate a DSN for the sender. They may also
> > continuously retry to send the message. (Not seeing it as
> > "Success" but missing that it is a "failure")
>Suffice to say, I hate MTAs like that, having more than once been at
>the receiving end of their idiocy when bandwidth was a precious
>commodity.  Some of the MS "mail servers" used to fit this

I think we've ALL been subject to silly mail servers like that at some 
point in time. :)

> > 2.) Most schemes designed to find attachments in messages are not
> > too well-thought-out and will false-positive more than is proper.
>Sadly true.

Also a point I'm arguing at work right now. :)


More information about the svlug mailing list