Re: Backend and non-backend (was Re: Stencil bounding box)

[Johannes Schindelin]

Hi,

On Thu, 4 May 2006, David Feuer wrote:

> On 5/4/06, Johannes Schindelin <address@hidden> wrote:
> > Hi,
> > 
> > On Wed, 3 May 2006, David Feuer wrote:
> 
> > > If we're looking to measure small changes, area of union minus area of
> > > intersection, divided by the area of the union, would probably be
> > > good.
> > 
> > If a really nasty bug creeps in, which makes the new bounding box very
> > tiny, but keeps the location, this distance will be about 100%.
> 
> Hm hm...  Dividing by the area of the union is bad.  Maybe divide by
> the area of the old ones?

This is asymmetric, but given that we want to see what has changed (i.e. 
we are more interested in deviations from the old state than from the new 
state), it is reasonable.

Any way you turn it, you will always have false positives. But they do not 
harm, whereas false negatives do.

Ciao,
Dscho