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

[Chris Murphy]

On Jan 28, 2011, at 3:06 AM, Richard Hughes wrote:

> It
> also meant the applications themselves didn't need to know what output
> they were on, and things like that.

I think this is important. It makes it much more complicated for smart apps to be smart if they have to track where pixels are being output so they know how to color manage them. Better to allow them to just send tagged pixels and let a window manager deal with where, what, how.

For the explicit off switch to opt-out, you'll need to consider an "anti-tag" or a special kind of tag in lieu of an ICC source profile that basically means "no color management". I think this is easier to implement for both parties than a separate API. 

It could take the form of non-ICC metadata, which informs the code to simply not pass pixels onto the CMM, a "pass through" which in the short term is more reliable but requires coding that pass through. Or create a particular flavor of ICC profile formatted profile that is a "null" profile, perhaps it's just a header with a single "noop" tag, which would require the agreement of CMM developers, so that if they encounter it, the CMM itself does a null transform. That's a longer term approach that the ICC probably should have enabled anyway.

> As large gamut display devices become more and more "mainstream" I
> think this kind of thing is going to be more and more important.

Yes. But what's already the case is mainstream displays are diverging from sRGB. They aren't just becoming wide gamut. They are all becoming different from sRGB, different from each other. Even the wide gamut displays aren't clearly converging on a wide gamut aimpoint. Their marketing materials make it sound like they are targeting Adobe RGB (1998) but they do this in a very circumspect manner with numbers like "reproduces 95% of the Adobe RGB gamut". OK well if I have two displays that reproduce 95% of the Adobe RGB gamut, they could still be rather substantially different from each other.

So yes, I think we already need this. And look at all the effort Mozilla has gone through to get Firefox color managed. I think this should be centralized. Apps should not have to do this individually. It's a service. Their role should be merely to pass through EXIF and/or ICC profiles, and the image stream, and then the system manages color appearance transforms. That's how we get it highly optimized and consistent.

Chris Murphy