[Openicc] Proposed "ucmm" display profile configuration convention used by ArgyllCMS.

[Hal V. Engel]

On Monday 02 June 2008 05:42:53 am Kai-Uwe Behrmann wrote:
> Am 02.06.08, 20:42 +1000 schrieb Graeme Gill:
> > Kai-Uwe Behrmann wrote:
> > > The proposal for the ~/.color/icc and /usr/[local/]share/color/icc
> > > directories appeared quite some time ago:
> > > http://bugs.freestandards.org/show_bug.cgi?id=77
> > > http://lists.freedesktop.org/archives/openicc/2005q4/000636.html
> > >
> > > The OpenIccDirectoryProposal is as stated on the very top a write down
> > > of previous discussion on this list. It comes not out of the blue.
> >
> > Sure, but the quality of any discussion generated is informed
> > by practice rather than theory. Things that seem fine on paper,
> > can turn out to be quite another story when it comes down to
> > implementation.
>
> CinePaint, Krita, Scribus and many other open source applications support
> /usr/[local/]share/color/icc as well as many propriarity ones.

LProf is also amoung these.

> What would you expect to be more practise?
> For subdirectories below this level you are pretty free to propose
> something.
>
> > > I have a LCD which's EDID dissappeared after a xorg version change.

This is clearly a bug in the version of xorg that was installed.  I think it 
should be treated as a bug and that we should expect upstream to fix it. 

> >
> > Right, but it's permanently disappeared, rather than coming or going ?
>
> Permanently. But it would have been nice to not even have to notice that
> as a user.
>
> > > Oh, I meant the Xinerama geometries. It is one of the few and of course
> > > even not very constant properties to identify monitors.
> >
> > I guess I'm not following you. Xinerama seem to list monitors in
> > the same order they appear in the X server with Xinerama turned off,
> > so I don't understand that it adds any value to the underlying X11 screen
> > number identification.
>
> Well a simple move or resolution switch makes the Xinerama geometry
> property failing.

Also XRandR changes this since display geometry can be changed at run time by 
the user in some rahter dramatic ways.  At best using geometry is a poor 
fallback.  I agree with Graeme that X11 screen number ID is a better fallback 
for failed EDID data since it is more stable.

Hal

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.freedesktop.org/archives/openicc/attachments/20080602/101449f6/attachment.html