In <20040112200356(_dot_)GB18595(_at_)dumbo(_dot_)pobox(_dot_)com> Meng Weng
Wong
<mengwong(_at_)dumbo(_dot_)pobox(_dot_)com> writes:
But if a provider also gives the total mail volume seen,
you can do a ratio.
Getting accurate mail volumes is A Very Hard Task.
You don't need to be very accurate.
Spam is 95% fraud and there are pretty sharp bends on the characteristic
behaviors. Meng is completely right to talk about greylisting, These issues
are not black/white and there are huge differences between the legitimate
parties generating mail.
Even so, orders of magnitude will serve fine for almost all purposes.
Going back to the prvious discussion on 'ratios', if one provider gives the
order of magnitude of the mail volue from a source and another gives the
order of magnitude of the number of mails caught at spam traps it is still
possible to arrive at a perfectly usable estimate of the ratio of spam.
Even better you can calculate the value you want through simple addition and
subtraction.
Assuming we have to encode the data value in an 8 bit field, the way to go
is probably to use a log scale with one decimal point
That gives a range as follows
index 10^ value
0 0.0 1 message (or less)
1 0.1 1.2 messages
10 1.0 10 messages
255 25.5 31 trillion trillion messages
Precision is limited of course, but is hardly likely to be problematic.
-------
Sender Permitted From: http://spf.pobox.com/
Archives at http://archives.listbox.com/spf-discuss/current/
Latest draft at http://spf.pobox.com/draft-mengwong-spf-02.9.4.txt
To unsubscribe, change your address, or temporarily deactivate your
subscription,
please go to
http://v2.listbox.com/member/?listname(_at_)©#«Mo\¯HÝÜîU;±¤Ö¤Íµø?¡