[Openicc] Helping with colord

[Kai-Uwe Behrmann]

Am 07.03.2011 14:00, schrieb Graeme Gill:
> Andrzej Nowicki wrote:
>> We initially wanted to build on oyranos also. We spent much time
>> studying oyranos in the last weeks but came to the conclusions:
>>
>> 1.    Oyranos had a poor chance of being picked up by distributions
>
> Are there specific reasons you came to that conclusion ?
>
> [ I have my own reservations about Oyranos, but it might be
>   constructive to know what others perceive to be its current
>   shortcomings, so that there is a chance that they might be
>   addressed. ]
>
>> 2.    Oyranos had no dbus interfaces (we use dbus many times in our
>> research projects)
>
> Perhaps a dbus interface could be added to Oyranos ?

That would make sense on top of Oyranos.

> [ It would probably be a mistake to base a core CMM on dbus though, since
>   that would limit the range of platforms it can run on, while not
> addressing
>   any of the the inter-system color management issues posed by printing
>   and (remote) displays. ]
>
>> 3.    Oyranos had no python interface code
>
> Perhaps someone could write python interface code ?

The future release of the object oriented Oyranos APIs would be a robust
base for such a project. Yiannis did a great job at working on a
solidifying the code base. But more object types need to be converted
before they can be exposed as stable APIs for device configuration. The
actual target is just this.

>> We appreciated developing on a new project such as colord is a large
>> amount of risk, but we felt it had the better architecture and had a
>> chance of being included in distributions in the future.
>
> I would hope that any software that has merit would have a chance of
> being included in distributions, although a software ecosystem that
> depends on distribution "king makers" is a poor one in my view.

kind regards
Kai-Uwe