Re: autolearn vs. small messages

[Axb]

On 10/22/2012 12:24 AM, John Hardin wrote:
All:
Would it make sense to implement a lower limit on the body size of a
message for autolearning, to avoid autolearning as ham any FN spams that
(e.g.) just have a URI in the body and nothing else?

imo, no.

Would it make sense to implement such a limit just for autolearning as ham?

ditto

if anything it should be a done using switchable eval/range plugin & 
rules with tflags noautolearn but should not be done within the core.