><< Would it be reasonable for the poll
workers to take the one or more
>external printers around to
eachAT in turn to print out the results? >>
>
>I think this would be possible in the precinct. What
I'm wondering about is
>when we are testing hundreds of units and
we want to test a line of 20 or 40
>units at the same time, would
using external printers be effective. Maybe
>it's better in
fact. I'm simply encouraging the development team to think
>about the testing phase as well as the election day
focus.
External Thermal
Printer
Steve Ricke has found a
handheld thermal printer that is battery powered (7.2VDC) and prints 52
characters per line and has an Infra-red (IrDa) Interface or a RS-232C
modular jack serial interface. Its dimensions are 6" x 6" x
2-1/2" with the paper holder and weighs less than a pound. Its
the Seiko DPU-3445 and it is already Windows CE and Windows 95/98
compatible. The retail cost starts at $242 (with battery and is
currently available. It prints at a speed of 2.0 inches per
second on a 4 inch wide paper stock. It can print 8000 lines on a
single charge.
Some of its
benefits/disadvantages:
- requires special
paper
- the life expectency of the report
isn't as long as an ink report
- it has the IrDa interface which
means ease of use (no cables). I don't think printing reports on
an IR link can be construed as a security concern (as opposed to
uploading results)
- its light and
portable
Printing Report Data
Question: Do
we want to print a results report from each machine at the end of the
election day or do we want to group the election results on one machine and
print one report that reflects how that particular precinct voted (like the
AccuVote does). I know the polling sites I have visited during
election nights were always keen on seeing that report. Besides, they
had to post the report as well on the door or window of the polling
place.
Another
Question: How's that going to work with five DRE's each
producing a report. Wouldn't they have to post all five reports on the
window?
Say we want to accumulate the
results from each machine onto one machine (i.e., a poll book) and print the
results report from that poll book, wouldn't it make sense to have
that same printer equipped machine print the zero totals tapes? Would
anybody doubt that the zero totals actually came from each machine?
I'd say that if somebody pointed an IrDa printer at an AccuVote-TS and
printed a zero totals tape or a results tape, I'd wonder about the
integrity. It could be perceived that the tapes were generated inside
the printer box and not from the AccuVote-TS. Maybe we ought to stick
with the cable attached option just for sake of
perception.
AccuVote /AccuVote-TS Common
Design Features
Question: Aren't we
trying to design this AccuVote-TS with a common base that can also be used
for the Accu-Vote? If that is the case, then we want an internal
printer in every AccuVote because every AccuVote needs to print zero
totals. I guess that kills the idea of an external printer unless for
election prep, you want to allow the system to print on an external
printer. But if you do that then you wouldn't be testing the internal
printer that would be used on election day.
Conclusion
I have now convinced myself
(regardless answers you might give to my questions above) that the above is
drivel and rambling because I have also convinced myself that we must design
in the space required to house an internal printer even if we don't install
it. You can have the internal printer installed as an option for
AccuVote-TS and use an external printer instead, but why bother if you've
already got the space available for an internal printer and there really is
no cost justification for the external printer over the internal
printer.
So, in the words of Rosanna Rosanna
Danna, "Never mind".
Ian