Re: Printers Revisited

From: Alan Dechert <alan_at_openvotingconsortium_dot_org>
Date: Sun May 23 2004 - 22:15:25 CDT

Karl,

> Right, which is why I think that ink jets should not be so automatically
> dismissed.
>
I don't think I've dismissed inkjets. I've said,

1) Lasers are very unlikely to be used in our mobile system: Inkjets are
    very likely to be used in our mobile system
2) Both inkjets and lasers will be certified for use with our pollsite
system
3) The real answer on who uses inkjets and who uses laserjets will be
decided, most likely, locally.

> As for the start up time issue - I believe that we are going to have an
> issue in which a voter presses "print ballot" and then walks away.
>
That's an interesting conjecture.

> Laser printers take several seconds even without warm-up (your example of
> 8 seconds is a long time - long enough for a voter to leave). I just ran
> a test in which I printed a web page from Windows to Linux/CUPs and then
> to an ink-jet that had been sitting idle for a few hours. The printer
> started working within 3 seconds.
>
But how long did it take to finish? How long will an inkjet take to finish
printing a really long ballot?

> I ran the same test against a Lexmark 312 - because of warm-up time
> it was 30 seconds before the paper even began to feed. ...
>
Maybe we won't try to get the Lexmark 312 certified for use with our system.
Other lasers for $300 and less take much less time.

> On the second test, with the printer warmed, it was 8 seconds
> before the paper feed (and significant noise) began.
>
OTOH, most ballots would be printed with the printer already warm, no?

> That's probably true. However, we're not talking about situations in
> which we run the printers efficiently - our goal is a problem free
> election.
>
I'm not sure that's our goal. We want several things for sure:

1) Transparency -- everyone can see and understand how everything is
supposed to work ("nothing up the sleeve"). It will be a system people can
trust.
2) Auditability -- in case of any questions/problems with tabulation, we can
go back and look at the comprehensive audit trail--electronic and paper--and
figure out where errors occurred so we can correct the problem(s) if
necessary.
3) Affordability -- cheapness is important so we can preserve public money
for other important uses, and also so democracy can flourish elsewhere in
the world--few find $3,000 voting machines affordable.
4) Speed -- for the voters, time is money. If voting takes too much time
and trouble, people will be discouraged from voting altogether. Steps must
be kept to an absolute minimum. Don't make the voter jump through a bunch
of hoops. Have enough machines available so they don't have to wait in
line. On the election administration side, speed is also very important.
Develop standard ways of doing things and standard file formats to minimize
manual labor in counting votes. Get the results done and published before
it becomes known that if a few hundred votes in Rebol County in FL could be
swung one way or the other, it could decide the outcome of the election. If
the count is slow, decisive counties surface and become points of attack for
people that want to manipulate the election result.

There are other things we want but I'm pretty sure "problem free" is not
high on the list. There will always be problems. We need clean, fair,
accurate, and rational ways to recover from these problems.

--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 23:17:59 2004

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