established, however, that many, many print controllers would be unable=
to
discern a blank impression because the "page eject" command from whatev=
er PDL
is in use is still invoked. We also agreed that sheet based accounting =
is more
pragmatic except in the case of very high cost marker supplies.
At and/or following LA, we even went to great length to come up with a
clarified definition of an impression being a sheet side traversing the=
marker
path - whether or not marks were made. We had to do this to prevent cou=
nting
two impressions for every sheet of a simplex job.
Even though I came out the shoot arguing for optimum accountability, I =
was
convinced in LA and remain convinced that the compromise described abov=
e is a
good one. I recall discussing an additional (optional) attribute for "n=
umber of
blank impressions" to accommodate devices that ARE able to detect and a=
ccount
for blanks, but don't remember if this has been
included.
I recommend we remain with these decisions.
Harry Lewis - IBM Printing Systems
jmp-owner@pwg.org on 12/17/97 11:54:27 AM
Please respond to jmp-owner@pwg.org @ internet
To: jmp@pwg.org @ internet
cc: szilles@Adobe.COM @ internet, ipp@pwg.org @ internet
Subject: JMP> URGENT: Should impressions include blank last page back
At the JMP meeting on 12/5, we agreed that the definitions of
impressions would count the number of times a media side goes past
the marker, even if there are no marks made.
I think we agreed to that, becasue impressions is supposed to count
after the sheet is stacked, so that the sheet counter doesn't know whet=
her
the back side of the last page (documents with an odd number of pages),=
was marked or not, so we said that it SHALL count.
Howver, for an accounting application, the customers may get pretty
unhappy with having to pay for the final side they didn't use, as
Angelo points out, when their document has an odd number of pages.
URGENT: I NEED FEEDBACK FROM THE JMP LIST BY THURSDAY 12/18 EVENING.
HEARING NO OBJECTIONS I'M GOING FORWARD WITH THE FOLLOWING:
So how about RECOMMENDING (but not requiring) that the number of impres=
sions
for two-sided printing not include counting both sides of sheets marked=
on
only one side. It may be that the interpreter has to be involved in
counting impressions, rather than the sheet counter in the stacker or m=
aybe
the implementation only worries about the last sheet and so there is ju=
st
an internal status bit that says whether a document has an odd number o=
r an
even number of sides in order to know whether to count the last sheet a=
s 1
or 2 impressions.
I suggest changing the sentence in the definition of impression:
If a two-sided document has an odd number of pages, the last sheet stil=
l
counts as two impressions, if that sheet makes two passes through the
marker or the marker marks on both sides of a sheet in a single pass.
to:
If a two-sided document has some sheets that only have marks on one sid=
e
(such as on the last sheet of a document with an odd-number of
impressions), those sheets SHOULD count as one impression, instead of t=
wo,
even if that sheet makes two passes through the marker.
BTW, the current definition of "impression" in the IPP Model is:
12.2.15 impressions
An "impression" is the image (possibly many print-stream pages in diffe=
rent
configurations) imposed onto a single media page.
So it seems that the IPP Job Model is in agreement with the following
recommendation for the Job Mon MIB:
The full definition of the term impressions (as sent yesterday) is
for the Job Monitoring MIB:
Impression: For a print job, an impression is the passage of the entir=
e
side of a sheet by the marker, whether or not any marks are made and
independent of the number of passes that the side makes past the marker=