Sorry for being somewhat offtopic but I'd like to rise some
points we already discussed (at least with Jon on a berlin
mailing list).
Berlin tries to be 'resolution independent'. All coordinates
are specified in units of 1 tenth of a millimeter. For this
to work right, one obviously need some information about the
resolution of the output device. The GGI visual structure
does have such a field, even though it appears that it is
not used at all (found out by try and error, *not* in the
docs !).
I understand that not all hardware lets you inspect these
details. However, we feel it is a necessary information to
provide from a central point, not via some calibrating in
each application which makes use of it.
In a nutshell, what we'd like to see is a central configuration
tool which lets you provide the physical data once and for all
such that GGI could then use it and the visual structure
would provide meaningful numbers in the size field.
User interfaces are currently in a sad state in that you can't
rely on sizes of anything. When passing an image around between
gimp, gs, netscape for example I find it unacceptable that it's
differently scaled/sized each time. This is one of our most important
criticisms towards X.
Berlin tries to keep applications completely unaware of device space
details. (Of course they might want to inspect them but they don't
need to) In each redraw the nodes in the scene graph use a given
cumulative transformation matrix. This means that you can embedd
an application whereever you want, it just takes the allocation and
transformation from the parent node and tries to deal with it.
Different output devices are orthogonal to the scene graph, they are
encapsulated into the 'DrawingKit'. We currently have an GLDrawingKit
(mesaggi), LibArtDrawingKit (libArt on top of GGI) and we plan a
PSDrawingKit (postscript stream). They all should result in the same
resolution independent output.
Best regards, Stefan
_______________________________________________________
Stefan Seefeld
Departement de Physique
Universite de Montreal
email: [EMAIL PROTECTED]
_______________________________________________________
...ich hab' noch einen Koffer in Berlin...
