Summary of major items from the second IPP Fax (Qualdocs) meeting. This was held
in Boston just following the IOC / MFPA meeting on 10/26.
To be honest the turnout was a little disappointing - we had hoped that many
people would stay over from the IOC and that this would make up for the absense
of the IPP people who had been at the 3rd bake-off the week before - didnt
happen.
I think the most important thing that came up is something that Lee Farrell said
(Thanks Lee). He pointed out that we still do not have a common understanding of
what it is we are trying to do; different people have different ideas. I would
therefore ask that people send me usage scenarios, ideas for overall
requirements (quality, compatability, speed, etc), non-goals, etc.
This somewhat took the wind out of the sails of the technical requirements
documents we discussed. But there were still some important areas of agreement.
1. The proposal to split the data format and transport was broadly accepted -
the name Univeral Image Format was liked for the data format and the negotiaion
mechanism.
2. THe issue of drawing surface size, scaling and squishing was discussed at
length. It was clear that we wanted to break away from the constraints of the
classic FAX/TIFF world that specifies the drawing surface size in advance (and
forces A4 and letter to be the same logical size). We all felt it was more
useful to allow the receiver to indicate its drawing surface). It was also
considered useful to allow the sender to specify how it wanted mismatched image
sizes to be dealt with - i.e. whether or not the receiver should scale or if it
should crop and/or leave whitespace.
3. Everybody seem to think that we may as well stick with CEILAB for color.
4. We discussed conneg. Somebody with actual implementation experience pointed
out that it was very unwieldy - you get a combinatory explosion in most
non-trivial cases. Still nobody has come up with a better idea.
For a more detailed record see Ron's email with the minutes.
Paul Moore
This archive was generated by hypermail 2b29 : Mon Nov 20 2000 - 12:00:00 EST