Re: Printers Revisited

From: Alan Dechert <alan_at_openvotingconsortium_dot_org>
Date: Sun May 23 2004 - 17:03:45 CDT

David,

> It would be very desirable--and not difficult to achieve with normal
> UPS batteries--to allow a voter who is in the middle of voting to
> complete her ballot, including both recording the EBI and printing the
> ballot.
>
> Presumably, in a power-down situation, no voter should be allowed to
> START voting on battery power. But the minute or two that might be
> involved in completing the process can be covered by batteries. I
> think that is pretty much true for either laser or inkjet printers...
> but certainly inkjets are more power-friendly.
>
> I know in border cases it could get more complicated: e.g. a voter who
> prints a ballot on battery power, but then finds it needs to be
> spoiled. But even then, the procedures are probably clearer, and more
> error resistant, if we allow an attempted vote to go to completion.
> The voter might have to wait for power to be restored before trying
> again after a spoiled ballot, but that's the same as any voter who just
> walked in.
>
I see trials. I see Human Factors testing. I see surveys including these
questions to election administrators and voters. I see reviews of existing
procedures and existing election code relevant to these questions in the
various jurisdictions.

It would be nice to find a study titled something like, "The History of
Election Day Interruptive Events at Polling Places, and How They Were
Handled." Power outages would be in a subset of events that I imagine to
include fires, ice storms, lightning strikes, organized human disrupters,
unorganized human disrupters, accidents, medical emergencies (e.g.,
pollworker heart attacks), and whathaveyou. A comprehensive academic study
of the voting system should include something like this along with a study
of recommended procedures.

> However, one more likely advantage for inkjets is physical form factor,
> and the associated ergonomics. To put the components in a compact
> space like a voting booth is likely to favor something of small
> dimension and weight. Lasers have shrunk more than I would have
> believed a decade ago, but still are many times larger and heavier than
> smallish inkjets. As the other issues, I find this concern reasonable,
> but not compelling in itself.
>
I think lasers are small enough and light enough that this is not a major
concern in the voting booth. The main reason bulk and weight might be
significant is when the printer is not in the voting booth--especially if
the plan is to store these printers as dedicated to voting. Lasers may be
too bulky to be dedicated and warehoused for this purpose. Inkjet printers
might be easier to store but the mechanism is not likely to store very well
(a year later, a lot of them aren't going to work).

> You can perfectly well design color schemes that are perceivable to
> red/green colorblindness. The blue/yellow form is MANY times less
> common. But even for rare types of colorblindness, a design would not
> represent anything SOLELY by color, color would only aid in
> distinguishing features.
>
That might be nice, but at what cost? If we get a fresh b/w toner (at 30
bucks or so) AND a fresh color toner (at 30 bucks or so), it could lead to a
cost per ballot of nearly a dollar just for ink! This cost is way out of
line. Unless the county has a clear use for the printer outside of the
polling place, I think county ownership of the printer is out of the
question--storage will be just too expensive. The county could buy the
inkjet for the election then give them away, but this leads to around a
dollar per ballot just for the printing: This cost is clearly prohibitive.

The best bet is to let the vendor worry about the specifics. The vendor
would guarantee that there will be a printer in the voting booth and that it
will print the ballot on the specified paper in no more than x seconds and
that it meets specifications a, b, c, and so on. The vendor would guarantee
enough spares on hand and would train pollworkers on how to remove one
printer and plug in another (or would have a tech on site to handle any and
all hardware issues). The vendor would guarantee that the printer is COTS
and has been certified for use with the OVC system. etc etc. We should not
try to anticipate what will be the best printer in all situations. A vendor
might have a (non election) customer that wants 2,000 printers of a
particular make and model and doesn't mind if they're removed from the box
and have 100 prints made on them before they are delivered. In that case,
the only cost to the vendor is the cost of setup and delivery (and maybe a
few damaged or stolen). Another workable model might be a situtation where a
county has purchased x number of printers for non-mission critical
applications (where they could be borrowed for Election Day) and keeps track
of them so they can be easily located and moved for Election Day. Again, I
don't think dedicated printers will be sufficiently economical.

Alan D.

==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Mon May 31 23:17:58 2004

This archive was generated by hypermail 2.1.8 : Mon May 31 2004 - 23:18:16 CDT