Re: Further findings on Print quotas // Re: Pykota

[Petter Reinholdtsen <pere@hungry.com>]

[Jonas Smedegaard]
> There is also multiple techniques to measure the basis for print quota:
> 
>  a) number of pages, as declared in job header
>  b) number of pages, as analyzed by actually interpreting the job
>  c) approximated amount of toner user
> 
> a) can be cheated.  A combination of b) and c) best reflect the actual 
> expenses.

Based on rumors from the printer admins at the university of Oslo
where I work, I can add more info.

The most accurate of b) is to ask the printer how many pages it has
printed before and after the job was processed.  The definition of a
page differ from printer to printer.  Some count sheets of paper,
while others count two pages for each sheet of paper.  Others again
count either, depending on the duplex setting.

One can also try to get ghostscript to interpret the job to count
pages, but this has proven to be inaccurate and possible to fool.

There is also the question of front pages with the name of the job and
the user printing them.  Should they be counted as part of the quota
or not.  If it is printed on only one page, while the rest of the job
is on both sides of the sheets, should a front page count as one or
two pages? :)

Printer quotas is harder than one should think.

Happy hacking,
-- 
Petter Reinholdtsen