I found
http://utility.nokia.net/~lars/meter/dkim.html
Which is interesting, but does not give an idea based on traffic. If yahoo,
hotmail and gmail implements dkim, well it could be more traffic using dkim
than may be a few ISPs...
So I think it is important to base these stats on traffic rahter than on how
many sites have it configured.
For the purpose of this workgroup I think we are looking at technologies that
reduce spam. So I guess a graph in time showing how much mail traffic is
dkim/dk/spf(?)/senderid(?) would be interesting. I think making the plot
available would be interesting. I saw with the stop of the recent botnet,
security companies run graph on the number of spams within the mail system., it
would be nice to see the above graph per non-spam traffic and spam traffic.
For your case you would have to monitor only inbound mail traffic, as with
outbound, it would only tell us what you have implemented on your MTA.
Not sure I'm fully clear here...
If several big mailing site do that, then it can be aggregated to get an idea
of what's happening.
----- Original Message -----
From: "der Mouse" <mouse(_at_)Rodents-Montreal(_dot_)ORG>
To: "Anti-Spam Research Group - IRTF" <asrg(_at_)irtf(_dot_)org>
Sent: Monday, 17 November, 2008 3:20:23 PM (GMT+1200) Auto-Detected
Subject: Re: [Asrg] The state of the email system
Looking at the discussions, I'm wondering if someone could point me
to graphs showing (with a timeline):
SMTP vs ESMTP traffic
Mime formatted emails vs non Mime
Text email vs text+HTML vs HTML only emails
Character encoding in emails (which ones are used)
PGP/SMime/Total
SPF/SenderID/DK/DKIM/Total
If they don't exist, how can we encourage, groups, email providers to
give them?
I don't know, but I can keep at least some of those statistics for the
mailers I run (my own) or co-run (at work).
My own mailer stats would border on useless, both because I'm such a
tiny leaf site and because the mail is heavily self-selected. Work is
larger, but nowhere near big enough to be more than one data point.
It's slightly less certain because I'd have to get approval to release
the data, but I have no real doubt the approval would be forthcoming
for aggregated statistics such as you list.
Assuming I (a) can implement it and (b) get approval to release the
data, what statistics would you like (just the above? more?) and what
useful things could I do with the data? Send the data to an
aggregator? Make plots available?
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse(_at_)rodents-montreal(_dot_)org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
https://www.irtf.org/mailman/listinfo/asrg
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
https://www.irtf.org/mailman/listinfo/asrg