What A Fellow Activist Has Learned about L.A.'s Voting System (not good)

Written by: F. Gluck on Jan 29, 2007 4:15 AM EST

Linked to groups: DFA Santa Monica

I received this from a fellow election integrity activist in L.A. County:

January 2007

What I've learned about L.A.'s voting system

1. L.A. county scanners are not very accurate, but I'm not sure to what degree. As I recall, optical scanners typically have an error rate of 1 to 2%. Analysis of a couple 1% mandatory manual tallies showed that most of the time the machine and hand counts do not agree.

2. The 1% mandatory manual tally is still being done in a suspect manner. First I found out the selection of precincts for the recount was NOT being done in a random manner, because a truly random selection might result in many more precincts to count. This is due to CA election code which uses the word "random" when describing how precincts are selected. I prefer not to go into a lengthy explanation of why a truly random selection could result in many more precincts to be hand counted. Eventually we seem to have gotten the attention of the registrar by raising public awareness of this issue. After that she said it was a sort of random selection, but did not describe the process. Then at a recent election there was a meeting open to the public where a supposedly random list was presented to be used for the selection of precincts. A list printed up before the meeting may or may not be entirely random. I'd much prefer the use of dice at the meeting to select the precincts over any random list produced by a computer, especially one programmed by a staff that formerly did not want to do a random selection. An alternative would be a book of random numbers.

3. The necessity of installing a new tabulating computer system is suspect. The old one, developed by county personnel, still works and can still be programmed. Diebold said they could build a new tabulator (GEMS2), but despite a rich contract still
haven't finished it yet. The staff says Diebold was the only vendor who had the ability and willingness to build a new tabulator, and thus was the only choice. With all the computer
science departments at major universities here in L.A. the registrar could have asked for expert help and probably received it. This is not rocket science. We could have built our own
new tabulator, but the staff says that would have cost more. I seriously doubt that.

4. Finding out how the election system works is very difficult. You have to ask a lot of questions and be persistent. I've never seen a flow chart of how ballots are handled and tabulated. Yes, there is a thick book of procedures, but in all my talks with the staff nobody has told me where I can get a copy of this book.

5. Citizens cannot learn much about the computer system that actually tabulates the votes. In a request for information on the computer file system the registrar said giving out that information would endanger the voting system, and then denied the request. The only possible threat would be her own staff members. We've been told that the tabulator computers are not connected to the Internet and are only accessed by the staff.
The tabulating computer is programmed by 5 staff members. We haven't met them yet. If revealing how the computer system works is a threat, then our voting system must be really vulnerable to insiders. The implication of the registrar's response to this request means that we have faith based voting in L.A. We have to have faith in the registrar and her staff, since it's all so secret what they do with the computers. We can't really know what is going on, even though the software is not proprietary to any corporation. Citizens paid to develop the software, but citizens cannot examine or look at the software in any manner. If we persist, I think observers with special status may someday have access to the software.

6. The registrar has decided to keep things even more secret from the public. For 2 years I did videotaping of logic and accuracy tests and a 1% hand count without any problem. But now there is a new policy. Now nobody can do video or audio taping of these events or any other public election events inside the Norwalk office. Yes, I can stand out in the hallway of
the MTS room on election night and shoot through a glass window with my camera. They didn't stop that, even though it's technically against their new rules. I can't hear what's
going on inside and most of what goes on is distant and blocked by workers. There is no point in videotaping that.

During past logic and accuracy test observers were allowed inside the scanner room. That's how we discovered new Dell PCs in each scanner cabinet. The Dell PCs are part of the new GEMS2 tabulation system that is not yet certified. We asked to have them removed on election night and I believe they were. Now there is a new policy. Observers at logic and accuracy tests are kept outside in the hallway and have to observe by looking
through a glass window. You can't hear much and you can't see very well because the computer screens are a distance away from the window. If this policy was in place in the past we never would have discovered Dell PCs in the scanner cabinets.

7. L.A. County apparently uses software that was never certified. I asked to see certification documents a few years ago and got the brush off from the staff member who supposedly maintains these records. More recently, other people have asked to
see the documents, but they didn't get to see them either. There may not be any documents. The certification system is so muddled that verbal and letter written "OKs" from the
secretary of state may be the way software is "certified."

The registrar' staff has told me that sometimes last minute software changes are made and there is no time to get approval from the secretary of state. This happens when a judge makes a last minute ruling demanding changes be made on the ballot.

8. We discovered that a computer in the MTS area is connected to the Internet, and is used on election night because election workers want to see election results while they are working on the election. We asked to have that computer removed on election night but were denied. We were told this particular computer is not part of the election system.

I've been a poll worker in L.A. for many years. I do give the registrar and her staff credit for improving the precinct level polling operations during the years I've been involved. The training is pretty good.

When I started asking questions a few years ago it was easier to talk to staff members. Now, with the number of activists increasing, I get the impression from some the staff that they feel harrased and that this is a big headache for them. I get the impression that they don't see the need for so much citizen involvement and questions. They do have a difficult job to do with constantly changing regulations, and they put trust and faith in Diebold and their computer election systems to help them do the work.

Tags: