trivial nomenclature suggestion

[Greg Louis]

For historical reasons we've got algorithms called graham, robinson and
robinson-fisher (or fisher).  This is slightly bogus, since Gary
Robinson suggested both of the latter two calculation methods.  They
differ only in that the "fisher" variant uses Fisher's method of
combining probabilities to arrive at the "spamicity" figure, whereas
the original calculates spamicity in terms of the geometric mean (GM)
spam and nonspam probabilities.

It would be appropriate to name them Robinson-GM and Robinson-Fisher --
too much typing, though.  How about just GM and Fisher?

-- 
| G r e g  L o u i s          | gpg public key:      |
|   http://www.bgl.nu/~glouis |   finger greg at bgl.nu |
| Help free our mailboxes. Include                   |
|        http://wecanstopspam.org in your signature. |