Abigail -

No. Sorry. You are incorrect. It's understandable that you are, but
let me explain

Base64 is used quite legitimately by Outlook, Notes, and GW, in
international settings. Clearly you don't do much business with
Eastern Europe or Asia - their mail is riddled with it (not just the
spam you get from there).

In fact a big problem the SA-devs have is because so many of us and
them are Westerners, we tend to use BASE64_ENC and SUBJ_FULL_OF_8BITS
as sledgehammers. We don't have enough of a comparative corpus of
spam and non spam here. Because most WE get is spam on average from
these countries, the scores get overbiased. This leads to users in
those countries getting substantial false positives.

Ultimately the only solutions are to a) impress the corpus samples
that initial scores build on so it's less western-biased (and
understands what REALLY makes spam in these cases. b) let westerners
who don't have many clients/business in Europe (particularly eastern)
and Asia to jam up those scores manually (which they can do now)

__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com


-------------------------------------------------------
This SF.Net email is sponsored by: INetU
Attention Web Developers & Consultants: Become An INetU Hosting Partner.
Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission!
INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk

Reply via email to