bogus bogotuning

[Clint Adams]

> In fact, in the early days, I developed a way to do manual tuning. 
> It's written up in bogofilter-tuning.HOWTO, which is still in the
> 0.16.4 distribution's doc directory.  Even there I make the point that
> you need a statistically valid number of messages in your corpus for
> really worthwhile results, but at least the process explained there
> might do better than bogotune with small numbers.

Is it possible for bogotune to output a number indicating the statistical
reliability of the results?