[EM] Electronic Voting Bill of Rights?

[Dave Ketchum]

On Mon, 17 Nov 2003 22:37:09 +0000 Gervase Lam wrote:

>>Date: Sun, 16 Nov 2003 17:59:28 -0500
>>From: Dave Ketchum <davek at clarityconnect.com>
>>Subject: Re: [EM] Electronic Voting Bill of Rights?
>>
> 
>>> It depends on whether you do Disk-At-Once or Track-At-Once recording.
>>> If you do Disk-At-Once (i.e. write all the ballots in one go), then
>>>what I said above would work.
>>>
> 
> Whoops.  I think it is better to use Session-At-Once here.  This is 
> because all computer data files are written into just one track per 
> session.
> 
> A disk can contain many sessions, which in turn can contain the required 
> number of tracks.  Normally CD-ROMs and Audio CDs only have one session 
> per disk.
> 

We are getting in pretty deep here.  I will repeat what I have said a few 
times:
      Vendor (whoever has this responsibility) writes program (whatever is 
standard and in common for all precincts using this program) on a clean 
disc of the variety of CD (assuming that CDs have enough capacity for 
following purpose).  These shall be write-once CDs in the sense that 
nothing gets erased or overwritten, and voting machines shall be incapable 
   of overwriting even if the CDs would tolerate this.
      If I understand the technology well enough, a record can be 
logically overwritten by writing the new version in new space and changing 
the pointer that points to this record.  Hopefully this can be detected by 
analyzing record lengths and positions (what is supposedly record 2 
positioned other than the position calculated for it by considering 
location and length of record 1).
      Detailer or detailers add definitions of ballots for each precinct. 
  This can be multiple records such as a record for state elections, a 
record for county elections, and a record for local elections within a 
county.  Point is that this data can be multiple records written at 
different times.

     Voting machine does a diary in several records (I read "sessions" above):

           What is recordable as part of polls opening, including program, 
all the counts (which better be zeroes), etc.
           One or more sessions of recording copies of ballots.  Technique 
and quantity recorded in any session shall make it impossible to identify 
a voter's ballot by its position in the record (other than the obvious 
case of a single voter voting in that election using that voting machine). 
  Multiple copies of ballots are appropriate to aid in recovery from 
defects in writing.
           Poll closing record, including at least the counts.

> 
>>Remember that the disk has other content - I want the single disk to
>>start with program, etc., to read when polls open, and for everything to
>>get recorded that might be of interest later.
>>
>>As covered above, I expect multiple ballots must go in each ballot
>>record written.
>>
> 
> I assume the multiple ballots here are duplicate ballots, which act as 
> backups just in case of data corruption.  Even with the start program and 
> duplicate ballots, you should be able to put in about 1/4 million ballots 
> on the disc.  However, this does very heavily depend on the computer file 
> system used on the disc.
> 

No, this was assuming that quantity of voters required writing a record 
whenever work space in memory filled up and required this.  I like better 
a work space large enough for the whole election, but am not sure whether 
this is affordable (remember that someone, someday, is going to beat 
California's record of 135 candidates and do ranked voting and/or 
write-ins that chew up lots of space).  Duplicates use up more space on 
the CD, but not a bigger work area.

> 
>>Does not really answer our problem, but it could simply be whether
>>anyone has written CD burner code to run on a Z80.
>>
> 
> Using the Domesday Project as an example, all I was trying to say was that 
> it is possible to get an 8 bit CPU to run an optical disk.  The Z80 is an 
> 8 bit CPU.  However, I doubt anybody has written CD burner code in Z80.
> 

Hopefully there would be enough voting machines to justify writing some 
burner code if needed.


> Anyway, the Z80 would be virtually dependent on how the CD burner was 
> wired to it.  This would depend on how the "wiring up" has been designed. 
> Only after knowing this would you be able to program the Z80 
> appropriately.
> 

Again I assume enough potential customers to justify suitable wiring.


> Thanks,
> Gervase.

-- 
  davek at clarityconnect.com    people.clarityconnect.com/webpages3/davek
  Dave Ketchum   108 Halstead Ave, Owego, NY  13827-1708   607-687-5026
            Do to no one what you would not want done to you.
                  If you want peace, work for justice.