Is bogotune helpful?

[Greg Louis]

On 20031202 (Tue) at 0257:16 -0800, Bill Wohler wrote:
> 
> Once upon a time, I used the information in Greg's tuning paper to tweak
> my values slightly. This time I thought bogotune would be able to do
> this more easily.

David and I have been developing/debugging bogotune rather intensely
over the past week or so.  The next release should be a whole lot
less quirky, and the one after that better still.  Bogotune's weakest
point is the way the false-positive target is chosen; we've greatly
improved it, but to do the best we can needs more experimentation.

I've been using the original bogotune (written in R) for my own and my
employer's bogofilter settings for a long time now, so I would say it's
helpful; but it needs tweaking for each individual dataset.  We want to
make it smart enough not to require in-depth understanding on the part
of the user, but as of 0.15.8 we're not there yet.  Perhaps we should
have said as much in a README or FAQ document.

-- 
| G r e g  L o u i s         | gpg public key: 0x400B1AA86D9E3E64 |
|  http://www.bgl.nu/~glouis |   (on my website or any keyserver) |
|  http://wecanstopspam.org in signatures helps fight junk email. |