--On 17 June 2009 11:18:08 +0200 Alessandro Vesely <vesely(_at_)tana(_dot_)it>
wrote:
der Mouse wrote:
Mail does not need perfect - or even very good - reliability in order
to be useful. When I first started using email, it could take a week
to get mail from Montreal to California, with a chance that sometimes
approached even that it would get lost on the way. This didn't deter
lots of people, including me, from using it anyway.
You were pioneers, and obviously expected that whatever steps were needed
to amend reliability would have been taken.
Good question. "Mail does not need...very good...reliability" does strike
me as quite complacent. Sure, email is still useful for lots of things, but
the spam problem doesn't just hit reliability. It also hits costs,
usability, trustworthyness, and speed of delivery.
In exchange for what are we giving up reliability now?
Nothing much. We need to have a clear path to a world where you can trust
that sender addresses aren't spoofed. Then we can begin to build reputation
services around tokens that end users understand (email addresses), instead
of tokens that they don't understand (IP addresses).
The path is long, but it needs to be walked.
--
Ian Eiloart
IT Services, University of Sussex
01273-873148 x3148
For new support requests, see http://www.sussex.ac.uk/its/help/
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
http://www.irtf.org/mailman/listinfo/asrg