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

[Graeme Gill]

edmund ronald wrote:
> Let me ask you guys a serious question - this is not a joke - don't
> you think that if the current CMS paradigm were reformulated and
> simplified (a bit like metric) people who design the OS tools would
> find it worthwhile to factor CMS in directly?

The most basic set of requirements are really, really simple.

Be able to identify a device, and associate a profile with it.

To be widely adopted though, it needs to be made simple for
application writers. That's not all that difficult either
when it comes to displays. All the machinery is there in
lcms, but some stable and universally available API's are
needed. Legacy applications need more drastic action, such
as faking an sRGB device for them.

The problem with display people, is that too many of them still seem
to think that "RGB is RGB is RGB", so that there is no problem worth
solving there. (And I say that as someone who in the late '80's
and early '90's was up to my neck in X11 server code, making
X terminals go really, really fast..)

I wonder how long it will take until someone plonks a really
wide gamut monitor on one of the KMS/DRI/DRM/GEM/TTM/X11/OpenGL developers
desks, and they start noticing that there is a problem worth
solving here. Or maybe it would be faster if Linus had one plonked
on his desk, since people seem to take more notice of his rants...

But I guess that wouldn't work either, since the simple workaround
is to switch the monitor to sRGB emulation mode :-(

Graeme Gill.