cleanfeed and wikipedia

[Ian Batten]

On 10 Dec 08, at 1401, Nicholas Bohm wrote:

>
> But it gives no detail of the training, and I would really like to  
> know
> whether the analysts are shown collections of images which "passed"  
> and
> "failed" in cases decided by juries, and are later tested by seeing  
> how
> good they are at distinguishing the one set from the other.  One would
> at least know how good they are at tracking the jury standard.

Does anyone know if juries are, in reality, asked to do this?  Is the  
criminality of the images themselves, as opposed to the circumstances  
of possession and any legitimate defence for an otherwise criminal  
image, actually tried by a jury?


ian