Tom,
I think that the job time attributes for IPP are useful, since they give an
indication as to when the printed output will be shipped, or made available
for collection. For high-end printers, the total cost of the job may depend
on the print time.
MOD 1.1 4.4.26 states that a printer that goes down has the option of
resuming its printer up time when it restarts, by adding the down-time. The
server is obliged to do this if it wants to preserve the times of existing
jobs.
That means that the printer-up-time actually represents the time since
installation. If the printer breaks down, and has to be replaced, the new
printer should resume the printer-up-time of the first printer, in order to
maintain the times of existing jobs. That means that the printer-up-time is
only an indication of the time since the business started accepting IPP
jobs.
We could add new dateTime attributes for jobs, but then clients would have
to support two alternative schemes. Perhaps it is better to introduce new
attributes, and drop the "time-at-xxx" attributes
Another point is that high-end printers often have a separate RIP engine, so
jobs can be entirely RIPped in preparation for being printed. Often the RIP
takes much longer than the print, and customers may be charged for the RIP
time. When a job has been RIPped, it may have to wait for some time for the
printer to become free.
The "time-at-processing" attribute would correspond to the start of the RIP
process, and the "time-at-completed" attribute would correspond to the end
of the print process. For certain jobs that have to wait for the printer to
become free, the time between these two attributes could be large, even
though the job spent a short time in the RIP and a short time in the
printer. For these printers, it would be useful to have optional attributes
for the start and end of the RIP time, leaving the "processing" attribute
for the actual print run. e.g.:
date-time-at-creation
date-time-start-RIP
date-time-end-RIP
date-time-at-printing
date-time-at-completion
Printers that RIP each page before printing (like our office printers)would
ignore the RIP attributes.
Regards,
Anthony Porter
-----Original Message-----
From: owner-ipp at pwg.org [mailto:owner-ipp at pwg.org]On Behalf Of Hastings,
Tom N
Sent: Wednesday, 24 March, 1999 1:31 AM
To: anthony.porter at computer.org; ipp at pwg.org
Subject: RE: IPP> MOD - Use of time [and Bake Off 2 Issue 17]
Anthony Porter's issue about time and suggestions are relevant to ISSUE 17
that was brought up at the IPP BakeOff 2 the week before last and that I
just published yesterday. Please consider and comment on these alternatives
as well as the ones that Anthony Porter has suggested.
Thanks,
Tom
>> text deleted