Meeting
notes from the UPDF meeting in Toronto, July 30th 2001.
Attendees
Ronald Bergman, Hitachi Koki Imaging Systems
Jeff Christensen, Novell
Lee Farell, Canon
Mark Hamzy, IBM
Mark VanderWiele, IBM
Jim Sommer, Granite Systems
Jeff Soord, Software2000
Don Wright, Lexmark
Michael Wu, Heidelberg Digital
Peter Zehler, Xerox
Patrick Pidduck, PrinterOn, with a colleague
Norbert Schade, Oak Technology, Inc.
As you can see, it was quite a crowd compared to our usual standards.
As we had some new faces, we took some time to introduce them into the basics of the UPDF architecture.
The group agreed on a group licence on sample source code.
This shall ease the development of UPDF specific functionality. Samples could be a set of functions and even a small test application to demonstrate the use of the Parameter Converter, Interdependencies or other very UPDF specific features.
Mark VanderWiele took a lead on that subject. He will provide a proposal soon, which may be taken over by other PWG groups.
There were a few notes in the device fonts section.
The current mechanism of declaring a font passive has to be modified to be able to do the declaration, but have the driver decide whether it wants to follow the recommendation.
We save the complete character increment, but not the ABC values in detail. It has to be discussed offline whether we have to change that.
Other than that the font section is stable for about a year now and will settle down even more with the ongoing development of two sample fonts.
Discussions with the major font vendors about automated UPDF exports are requested.
We spent the larger part of the afternoon in a discussion about a possible extension of the Parameter Converter.
One of the ideas was to extend it to a kind of language to be able to describe future features, which cannot be predicted in the current standard.
Another one was to use an extended version of it to handle Interdependencies.
A third item was the attempt to eliminate the long lists of enumerated lists of generic features or generic PDL output elements and to get to a more general description of generic features. But that isn’t possible with the current structures. This will be discussed offline in detail the next days and weeks.
All approaches were based on the difficulties experienced during the implementation of UPDF in the host application like a driver.
It is still an open discussion whether the XML files have to be used in its original format or can be converted to some other, likely binary, format.
To solve these problems will keep us busy the next weeks and will have the highest priority.
We agreed that the support for raster graphic and color features plus a review of the event handlers have the highest priority for the summer.
A conversion from our current DTD format to the XML schema definition, which is widely used these days, seems to be appreciated by larger parts of the group, especially those, who deal with different schema dialects often.
That is another major task for the next weeks.
So the rather technical discussions about certain features kept us away from thoughts about how to prepare the publishing of the UPDF standard, level 1.
We all saw there is some more significant work to be done.
As the next PWG meeting may not be in Texas and we may not find enough time to prepare all the things we’d like to, it seems more realistic to concentrate all our strength on the solution of the outstanding technical problems.
Please watch the reflector on upcoming conferences.
Norbert Schade