Re: Barcode Redux

From: David Mertz <voting-project_at_gnosis_dot_cx>
Date: Fri Jun 04 2004 - 16:23:18 CDT

On Jun 4, 2004, at 5:05 PM, charlie strauss wrote:
> OCR pro:
> machine readable ballots.
> voter can with his eyes verify the information that will be read by
> the scanner.


> OCR con:
> Hideous design issue concerning paper jams( and fatal to project if
> not resolved)

This one is a barcode "con" also though. The BVA probably won't use a
sheet feeder. But "election central" almost certainly will need to
autofeed sheets to read each barcode. Wasn't it you, Charlie, who
stated a requirement that it is possible to re-canvas 100k ballots in a
day. You certainly can't manage that with a hand scanner waiving over

> Garden variety systems are very slow (both at the user level and the
> sheet fed level.)
> Garden variety systems have huge error rate ( probably fairy easy to
> fix with post processing, redundant info)


> potential for stenography is high yet we may not care
> No, but if the only machine that reads ballots is a bar code reader,
> the reality of accessing things embedded in the text portion is
> essential nil.

Well... read by whom?

Suppose the main ballot hides something malicious, but the barcode is
copasetic. The BVA and the BRP that utilize the barcode won't detect
the sneaky stuff. But the planted pollworker spy who handles the
ballot will have an opportunity to compile a blacklist of the people
who didn't vote as commanded/purchased (or whichever other attack we

Establishing narrowly that the BVA/BRP don't utilize the covert channel
isn't really any harder for OCR than for barcodes. Just because those
apps don't take advantage of the leak doesn't mean that the bad guys
don't have their own "Leak Reader" application.
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Wed Jun 30 23:17:06 2004

This archive was generated by hypermail 2.1.8 : Wed Jun 30 2004 - 23:17:29 CDT