New header token tagging

[Greg Louis]

On 20030925 (Thu) at 1402:28 -0400, Greg Louis wrote:

> > Greg uses train-on-error and has seen his false positive rate skyrocket.
> >  It's so bad for him that he has requested a way to turn off the new
> > tagging.
> 
> ... degeneration renders [retraining] unnecessary,
> although of course the benefit of head: tagging (I prefer head: to h:)
> isn't realized as quickly.
> 
No, that's not true.  Degeneration won't help avoid retraining; if one
turns on header tagging and degeneration, the first head: tag
registered for a given token will supplant the existing registrations
of the untagged token and muck up the classification just as badly as
would be the case if degeneration weren't in use.  Bummer.

-- 
| 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. |