[OpenICC] Oyranos APIs update

[Graeme Gill]

Chris Murphy wrote:

> If you're going to have a pop-up menu option called Editing Space  
> (probably RGB Editing Space), then you will have to somehow parse the  
> list of profiles and only display the ones that are in fact well  
> behaved, otherwise the result is going to be an explosion of  confusion 
> (as well a heretical terminology to real world incongruency).
> 
> Unfortunately the ICC didn't get a clue a while ago and come up with  a 
> different profile class for these kinds of profiles;

And a good thing too really. "Editing Space" is a subjective judgement,
so it shouldn't be encoded as a fundamental type such as profile class.

It's your judgement that useful colorspaces to edit in are grey balanced
etc., but other users may have different purposes in mind, and not
need such properties. The correct approach (IMHO) is to categorize
in the operating system or application, and allow the configurers and
users to decide which profiles are in the "Editing Space" category.

Some applications may have various implementation limits on
what they can use as Editing Space (ie. only 3 channels,
RGB like space etc.), so obviously inappropriate profiles can't
be used for such a purpose.

> sorting by profile class. But since the display class was usurped  for 
> this purpose, we now have various problems, including that of  user 
> interfaces showing non-device "editing spaces" in lists for  display 
> profiles to select for a particular display! Bad!

Any profile capable of defining a colorspace in relation to CIE space
should be available. Some people want to edit in CMYK, Lab, HSB etc.

[Now I think in fact that the ICC format has to many distinctions without
  clear purpose. Why class display profiles differently to output profiles ?
  What is the point of Colorspace conversion profiles ? It's one thing to
  mark the nature of the device colorspace in the profile, it's another to
  make it a completely different class. Only 4 or 5 classes are needed.]

Graeme Gill.