Re: bar code bit encoding

From: <Adechert_at_aol_dot_com>
Date: Sat Sep 13 2003 - 16:24:49 CDT

In a message dated 9/13/03 2:00:05 PM Pacific Daylight Time,
voting-project@gnosis.cx writes:

> While I'm not sure about the EXACT answer to this question, I am
> confident that the answer is "significantly more than 11 bytes."
>
I am basing it on what I've heard talking with a number of people
knowledgeable about 1-dimensional barcodes... about 30 symbols is about as much as you
want to try to scan. Jan's scheme needs 24 symbols. We know this will work.

> Maybe Jan can give a more precise answer about common scanner
> technology, but I would guess we have room for three or four times that
> much without real difficulty. The decision about encoding should be
> made out of a desire to save $20 by finding the cheapest -possible-
> scanner,
>
I get no sense that the 30 symbol warning depends much on the quality of the
scanner. If you need much more than that, you go to a 2-dimensional scheme.
Then we're looking at a $500 scanner. This complicates matters a lot.

> btw (especially since that $20 might cost us 50 hours of
> programmers' time).
>
But there are 200,000 polling places in the USA -- we need probably at least
two (one plus a back up) for each polling place. So people looking at the
system will be multiplying this figure.

> I think it's pretty much agreed that actual write-in votes won't be on
> the barcode (for the demo), just the fact the vote *is* a write-in. To
> actually count it, tabulators will have to look at the ballot face.
>
Not to put too fine a point on it, but the write-ins can be ocr-scanned.
Furthermore, the write-ins will be included in the data taken from the PCs
themselves. It should only be necessary for tabulators to look at the ballot face
in the event of manual re-count and while doing random checks for accuracy.

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 Tue Sep 30 23:17:05 2003

This archive was generated by hypermail 2.1.8 : Tue Sep 30 2003 - 23:17:09 CDT