religion

[Matt Armstrong]

David Relson <relson at osagesoftware.com> writes:

> At 11:09 AM 1/22/03, Matt Armstrong wrote:
>
>>Matthias Andree <matthias.andree at gmx.de> writes:
>>
>>I'm the guy who suggested -u to ESR (I may have sent him a patch
>>too...) back when bogofilter had only the Graham method.

[...]

> So, I guess we share the guilt - you had the idea and I did the
> dirty work.

I'm glad you did the work -- previously I had procmail rules running
the mail through bogofilter twice -- once to classify and once to mark
as SPAM/non-SPAM.


>> I went on using -u with Robinson-Fischer without realizing that it
>> wasn't training on unsures.  I think -u is much less useful with an
>> 'unsure' state.
>
> RespectfuLy, I disagree with you on this.  I find that "unsure" adds
> value.

I'm curious about how you deal with the extra complexity.  Unless I am
missing something, for the way I work I think it adds complexity with
no value.  Maybe you can suggest an improved way of dealing with
things:

    (1) I throw all SPAM into a SPAM folder, and all other mail is
        automatically sorted into a plethora of other folders.

    (2) I periodically read the SPAM folder for false positives and
        "bogofilter -N" them (a rare occurrence).

    (3) When I run across a SPAM in one of my many non-SPAM
        folders, I "bogofilter -S" it.

With this system, all messages get trained on with minimal fuss.

Now if bogofilter doesn't train on 'unsure' messages, step 3 becomes
more complex since I have to decide whether -S or -s is appropriate.
I also have to somehow come up with a way of training on the 'unsure'
non-SPAM.


> Bogofilter currently, i.e. the beta version, has two relevant
> parameters which are named "spam_cutoff" and "ham_cutoff".

Great!