bogofilter resistant email

[David Relson]

On Thu, 19 Feb 2004 00:00:53 +0100 (MET)
Pavel Kankovsky wrote:

> On Sat, 14 Feb 2004, David Relson wrote:
> 
> > It could be a language issue, since the message was English and (I'm
> > guessing here) that you're in Czechoslovakia.  Is there any relation
> > between ham and spam and English in your messages?
> 
> Hmm...yes, English messages are slighly more frequent in my spam
> corpus(approx. 85 % English, a small number of messages in Czech and
> the rest is a mix of various other languages) than in my ham corpus
> (70 % English, the rest mostly in Czech).
> 
> BTW1: Czechoslovakia was "divorced" in 1993. We have two independent
> states now: Czechia and Slovakia.

My apologies for not keeping current on world geography.  My last world
history class was in the 1960's and much has changed since then.  As
best I can tell, 90% of Africa has new names, as well as much of eastern
Europe and southern Asia.  I think North America, South America, and
Australias are about the same.


> BTW2: I have found something bizzare when I was browsing my spam
> corpus: a few English language spams with random Czech words.

If they're close in spelling, might they be misspelled english?  I don't
expect a high level of literacy from spammers.