Everybody's sysetm is different of course.
I only offer those statistics as additional data.
You might consider that since condtions change over time, and in particular
spam rates for any given system tend to rise over time, you should be
prepared for higher rates in the future.

Hope this helps,
_M

]-----Original Message-----
]From: [EMAIL PROTECTED]
][mailto:[EMAIL PROTECTED]]On Behalf Of Markus Gufler
]Sent: Sunday, January 05, 2003 8:32 AM
]To: [EMAIL PROTECTED]
]Subject: RE: [Declude.JunkMail] External test question
]
]
]Hi Madscientist,
]
]As I can understand we have a different situation here.
]I have no statistics about this but after weeks of research in smtp-
]declude- and spamchk-logfiles I'm 100% sure that we will never reach
]such a value.
]
]Assuming that with our current settings (declude + blacklists + spamchk)
]we're catching only 50% of all spams, (I'm sure the real value is
]appreciably higher) you can simply double the values in the
]ratio-diagramm of our spam-report. Then we have a max value of 60%
]during weekends and 20% during workdays.
]
]Markus
]
]
]
]> -----Original Message-----
]> From: [EMAIL PROTECTED]
]> [mailto:[EMAIL PROTECTED]] On Behalf Of Madscientist
]> Sent: Sunday, January 05, 2003 1:38 AM
]> To: [EMAIL PROTECTED]
]> Subject: RE: [Declude.JunkMail] External test question
]>
]>
]> According to recently collected Message Sniffer logs, on
]> average more than 70% of incoming email is spam. We have an
]> extremely low reported false positive rate.
]>
]> _M
]>
]> ]-----Original Message-----
]> ]From: [EMAIL PROTECTED]
]> ][mailto:[EMAIL PROTECTED]]On Behalf Of
]> Smart Business ]Lists
]> ]Sent: Saturday, January 04, 2003 4:40 PM
]> ]To: Markus Gufler
]> ]Subject: Re: [Declude.JunkMail] External test question
]> ]
]> ]
]> ]Saturday, January 4, 2003 you wrote:
]> ]MG> A.) With "identified as spam" you mean they reached the
]> hold value? ]MG> B.) With 2430 processed msgs you mean
]> inbound + outbound? ]MG> In this case 39% is a very high
]> value if you've not some spammers ]MG> as client that create
]> outgoing spam. ] ]39% is about what we saw a year ago.  Some
]> days we're as high as 80% ]held.  Our False Positives are
]> usually under 4% of held.  We manually ]inspect. ] ]Our stats
]> for last 7 days (incoming only) ](both messages and spam have
]> been down since 12/24 - ]I guess everyone has taken a bit of
]> a vacation) ]
]> ]Date        Fp    Fp%  Held  Total  Held%
]> ]==========  ==  =====  ====  =====  ======
]> ]12/28/2002   5  0.94%   534    841  63.50%
]> ]12/29/2002   6  1.39%   432    755  57.22%
]> ]12/30/2002  13  2.23%   583  1,474  39.55%
]> ]12/31/2002  10  1.74%   575  1,393  41.28%
]> ]01/01/2003   7  1.59%   441    796  55.40%
]> ]01/02/2003  25  4.10%   610  1,546  39.46%
]> ]01/03/2003  16  3.02%   530  1,492  35.52%
]> ]
]> ]
]> ]Terry Fritts
]> ]
]> ]---
]> ][This E-mail was scanned for viruses by Declude Virus
](http://www.declude.com)]
]
]---
]This E-mail came from the Declude.JunkMail mailing list.  To
]unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type
]"unsubscribe Declude.JunkMail".  The archives can be found at
]http://www.mail-archive.com.
]
]---
][This E-mail was scanned for viruses by Declude Virus
](http://www.declude.com)]
]
]---
]This E-mail came from the Declude.JunkMail mailing list.  To
]unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type
]"unsubscribe Declude.JunkMail".  The archives can be found at
]http://www.mail-archive.com.
]
]
]---
][This E-mail was scanned for viruses by Declude Virus
(http://www.declude.com)]

---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type "unsubscribe Declude.JunkMail".  The archives can be found
at http://www.mail-archive.com.

---
[This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)]

---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type "unsubscribe Declude.JunkMail".  The archives can be found
at http://www.mail-archive.com.

Reply via email to