[Openicc] Linux CM ideology, was: meta data in test chart

[Graeme Gill]

edmund ronald wrote:
> Maybe if we could agree exactly what a color appearance transform is, we could move from there.
> Because a simple way to deal with things would be to require apps to use a big space eg.  Lab,
> XYZ, or even (gasp!) ProphotoRGB and then transform from there into display space. But Graeme
> tells us this cannot be done ...

While ideal, it is a long way from what most applications currently expect. Many
are coded in RGB, and have lots of hard coded RGB values in them. As soon
as you move to a large gamut space, you need better precision (16 bits/component)
as well.

I'm not sure where "this cannot be done" comes from. The point is that saying
"do a perceptual transform from colorspace encoding A to B" where A is large gamut
and B is not, is insufficient. 'A' needs to be augmented by a source gamut to
be able to create a useful transform.

Graeme Gill.