[Spambayes] SpamBayes

Amedee Van Gasse amedee at amedee.be
Mon Apr 13 13:38:25 CEST 2009

Hash: SHA1

Thomas Hruska schreef:
> Amedee Van Gasse (amedee.be) wrote:
>> On Sat, April 11, 2009 16:02, John VanSickle wrote:
>>> Hi,
>>> I was just wondering if SpamBayes is still in development?  I noticed
>>> you
>>>  still have the same Alpha version as of June of 2007.  It appears dead,
>>> which is too bad since I liked the product and still use the alpha.
>> I don't think it is dead. It gives no errors for me.
>> And when a product is finished, does it need further development?
>> I don't think so: when it works, don't break it.
> If an anti-spam product were truly finished it would delete all spam,
> keep all good mail, and never need updating/tweaking/training. Spambayes
> is therefore not finished.
> IMO, the moment a single spam message enters my inbox, that is an error.
>  The moment a good e-mail that should have entered my inbox is, instead,
> deleted, that is an error as well.  For the most part, Spambayes works,
> but it falls flat on its face quite often.

Funny, I have a different opinion about anti-spam. I don't think that
anti-spam has to eliminate spam, only that it has to reduce it to
tolerable levels.

With the current state of technology and the constant changing nature of
spam, it is impossible to construct an anti-spam product that is perfect
according to your definition. You would need an artificial intelligence
like in science fiction, and even then they wouldn't know the difference
between a tortoise and a turtle.

If anyone is going to make a sentient SpamBayes, please include this
line in the code:

- --
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org


More information about the SpamBayes mailing list