[colord] Avoiding double correction in graphical apps

[Alexey Galakhov]

Hi,

I think about supporting simple compositor-based display correction so
that the naive "uncorrected" X output will be magically handled as
sRGB. compicc is another example of this approach. However, many
existing apps will still try to do software correction and this would
result in ICC applied twice. Of course one may give sRGB output profile
to that software but this is unsuitable in case of wide gamut monitors.
So, some mechanism is needed to determine if the software is doing the
correction on its own or not. As an example, compicc
uses net-color_spec
http://www.oyranos.org/wiki/index.php?title=Oyranos/net-color_spec_0.2
to do that. Let's support it?

A similar problem exists in ghostscript+gutenprint pipeline (with or
without CUPS). I'd like to support calibration with or without
characterization in gutenprint. However ghostscript applies
characterization too, and I assume it's done better at the time of
vector rendering. So gutenprint needs to know if it's input is
already corrected or not.

Any thoughts?

Regards,
Alex