[mailto:owner-spf-discuss(_at_)v2(_dot_)listbox(_dot_)com] On Behalf Of Seth
Goodman
From: Hallam-Baker, Phillip
Sent: Monday, November 29, 2004 2:43 PM
I don't need to read sendmail's data. We have our own
implementation and we do not see a two-fold slow down
based on the
size of the message.
The conclusion that Sendmail came to was that the cost was
insignificant. Very very few outgoing mail servers are cpu bound on
sending mail. If you are cpu bound on incomming email then you can
always delay verification and do it offline as CPU cycles permit.
Look at their data, not their conclusion, which does not
agree with their own data.
On the contrary, they demonstrated an UPPER bound on the CPU requirements
for DK. You cannot use the Sendmail data to infer anything other than either
DK is viable or that the case is unproven.
If you believe that the load is unacceptable then the onus is on you to
explain why.
I just bought a new PC for my son to play Dora the Explora on. For $350 I
got a Dell PC with a 2.4 GHz pentium chip. I refuse to believe that any
party sending out huge quantities of legitimate email would find such an
impossible burden.
The parties that are likely to be early adopters of DK are not going to find
that level of cost an issue.
the load is incoming, not outgoing, and you can't do
verification off-line and still reject during the SMTP
session. That is exactly the direction we don't want anyone to go.
Rejecting in the SMTP session is an unnecessary crock, particularly for DK
since there is no way to complete validation until you have read the last
byte of the message.
Pardon my intrusion on this fantasy, but not everything Yahoo
or Microsoft does is sensible or intelligent. They also have
very large budgets and there are political factors that can
outweigh the technical ones.
What you call political factors are known as 'customers' arround here.
Go troll somewhere else.