There seem to be two extremes on this whole issue. On the one side
there are those who would like all their favorite states included in
one easy to use job state object. On the other side are those who want
to keep the job state object ultra simple and have lots of job state
reasons. Both points of view have been backed up by many good
arguments. Last week at the IPP teleconference it seemed that a
reasonable COMPROMISE was reached. Unfortunately, several key JMP
folks were not present for that discussion and now we're waffling on
the whole issue again. Based on todays mail volume there does not
appear to be any end in sight.
For the record, here is my position.
IPP and JMP should use EXACTLY the same set of job states and the same
state reasons (the agreement reached at the last IPP/JMP
teleconference seemed reasonable to me). How are we to expect the PWG
to be taken seriously if we simultaneously create two specifications
with different job models? The rest of the world expects and deserves
a consistent set of standards from this working group! Please don't
respond again how easy it is to map between the two different state
models. I understand the mapping because I have been following the
discussions. But the rest of the world will not find the mapping to be
so obvious. If we learned anything from the Printer MIB interop
testing it is that interpreting ONE model correctly is hard enough.
Now, we are on the verge of generating two different models to
represent the same thing. It's time to find a compromise.