Dallas Engelken wrote:
No, you're right, thats not fair. If I compare only recent reactive
listings, minus the subdomain hosters that we list, you hit about 60%
whereas before it was more like 27%.
imvURI stats from last 5000 URIBL black listings
-> 2981 hits
-> 2019 misses
Dallas, I've made some recent *substantial* improvements to ivmURI. (1)
I've added *several* new spam sources... it was always a weakness of
ivmURI that the raw data that fed ivmURI wasn't "wide" enough. That
incoming data is much wider now! ...and... (2) I improved ivmSIP's
response time (previously, it was getting bogged down in some auditing
tasks that was delaying writes to the rsync files... that has been fixed).
RESULTS...
stats from 5/23/2008 (a few minutes ago).
---------------------
322/500 (ivmURI hits from the latest 500 URIBL listings)
(whereas a couple of tests in April showed 186/500 and 225/500)
301/500 (URIBL hits from the latest 500 ivmURI listings)
NOTE: to compare apples-to-apples, subdomain listings in URIBL were removed
Let me know if you'd like a snapshot of ivmURI for your own analysis of
these latest improvements.
ALSO...
In spite of your off-list explanation, I'm STILL confused about what you
mean when you refer to URIBL's *pro-active listings*???
You must be either referring to:
(A) Listings *currently* in URIBL-GOLD, but *not* *yet* in URIBL-BLACK
--or--
(B) Listings *currently* in URIBL-BLACK which were *previously* listed
in URIBL-GOLD
Which is it? "A" or "B"? (or something else?)
OF COURSE: The silly part about all these stats is that the *superior*
comparison between DNSBLs is "hit rates" on spams sent to mail servers
combined with low FP rates. It is possible for a DNSBL to have far fewer
listings, but, in "real world" testing, hit on higher numbers of spams
with less FPs.
Rob McEwen
