Citizens Oversight Status Update 2016
Citizens Oversight (2016-12-30) Ray Lutz
This Page: http://www.copswiki.org/Common/CitizensOversightStatusUpdate2016
This has been a successful year for Citizens Oversight and for everyone who has helped in our projects. And very recently, we have had a few breakthroughs putting us on a track for more success in the coming year. Mainly, our focus on Election Integrity
was the most productive but have also made progress on several fronts in the San Onofre shutdown project, and a few others we can note below. Be sure to follow the links for more information.
We started watching San Diego County Registrar of Voters (SDROV) carefully after the 2004 elections working with some other groups. This was partially the impetus for the formation of Citizens Oversight
as an organization. From 2008 to 2010, we did an in-depth investigation of their procedures and created a comprehensive report
. The most important finding and the reason it took so long was the fact that their procedures were largely undocumented, but we started to really understand that the 1% manual tally was our most important opportunity to exert oversight to catch election manipulation. We realized that we needed a special data file to allow us to be able to make sense of the 1% manual tally results (to insure that they did not just swap out the report) and this was the basis of the Snapshot Protocol
, which we continued to roll out this year (and covered more fully below).
One of the key activities of the Snapshot protocol is the careful oversight of the random selection meeting, which we video recorded. At that point, we realized that the number of batches being randomly selected for audit was insufficient. Michael Vu, the SDROV, simply omitted about 285,000 ballots from the 1% manual tally audit selection process. After asking him to include the ballots and getting a "respectful" refusal, I wrote up a lawsuit and filed it pro-per. Since I am not an attorney, can only represent myself, and that only in clumsy fashion. Once the lawsuit was filed, I worked with Attorney Alan Geraci to represent both myself and Citizens Oversight in our lawsuit against Vu.
Our case against Vu was very strong, as the Election Code section 15360 on the 1% manual tally explicitly says it will include Vote By Mail ballots. In the early drafts of the law, it also said provisional ballots were to be included too, and letters to the Governor of the final version included references to the fact that both VBM and provisional ballots would be covered by the audit. We had a four-day trial, with our star witness being Professor Phillip Stark of Berkeley, the inventor of the Risk-Limiting Audit. The County had witnesses from several other counties who largely said they also violated the law (LA, Sacramento, San Luis Obispo) and Deborah Seiler, the former SDROV who had a hand in the restructuring of the election code and a big player in CACEO (see below). Unfortunately, with all the testimony from the county, we were unable to record the trial due to a request by the County that it not be recorded. It is amazing that our elected officials are not proud of what they do and want it publicised. There was a lot of good information that will only be in the written transcript (which should be available once the appeal is filed).
The county argued that provisionals were omitted in the legislative process, and it was true that the words were stricken. But this was more about eliminating redundancy in the words than actually intending to omit provisionals from the audit, or at least that is our position. The lobbying group of election officials, the California Association of Clerks and Elections Officials (CACEO), actually sent a letter to Gov. Schwarzenegger when the bill sat on his desk saying that despite the efforts of CACEO to make sure VBM and provisionals were not included in the 1% manual tally, that they failed, and they were still required by the bill. So even they admitted that the new law did require that the audit cover both all VBM ballots and validated provisionals. (Fortunately, we learned that the legislative history of all bills processed by the legislature is archived by the Secretary of State. This was all submitted as evidence in the official record and will be available in the eventual appeal.)
Later, the county used a very disingenuous argument that we wanted the county to audit even provisionals that were never validated as legitimate ballots in the audit. We have never said that, but the term "provisional ballots" is used carelessly sometimes. There are 1. the provisionally cast ballots which are placed in envelopes on election day, and 2. the subset of those ballots which are not validated and left in envelopes and 3. the others which are validated and are then treated as any other ballot cast at the polling place. Validated provisional ballots are not to be treated any differently than a regular ballot, and we argue, that means they are also to be covered by the 1% manual tally audit. But the judge ruled against our request for provisionals based on this ridiculous argument, which we believe we can win on appeal.
Our lawsuit was about the primary. The processing of the VBM ballots by the SD ROV was extremely suspicious. I don't think this is just about incompetence. The hoops they jumped through in the primary to make sure the audit was not trackable was astounding. First, they selected only about half of the VBM ballots for audit by batch. This I already mentioned was the basis for the lawsuit. But they say it is impracticle to produce a report of the entire election, batch by batch, prior to the selection of the random sample. This allows us to total up all the batches and verify that it matches the grand totals for each race, and then we can compare the votes in each random batch once they are selected to what they actually manually tally. So we had to live with this shortcoming for the last few years because the GEMS software they use from Premier Election Systems (Formerly Diebold Systems) is "very difficult to make that report." So they always make the report AFTER the batches are selected. This breaks our protocol (the Snapshot Protocol
) because it would be easy for an insider to correct the reports so they match the batches to be audited, thereby concealing any manipulation of the VBM ballots. It is key that we get the report BEFORE they do the selection, and that means it will be of all the batches, because we don't know which ones they are selecting.
What they did next was very strange. Instead of just adding a few batches to the audit so as to include another 8 batches or so (as I requested in my email), they went overboard and spent an entire week with 40 people working, rifling through nearly 1300 batches to find the ballots corresponding to the originally selected precincts. I carefully checked the actual totals from these precincts and they did not exactly match the snapshot file. That is, the raw number of ballots in each batch was incorrect, meaning they had created brand new reports. This was something I tried to bring to light in our trial but it was just too complex and detailed to get across. But more worrisome is the fact that by preselecting the ballots from all the batches meant they could make sure they matched exactly. We noticed in their 1% manual tally report that the first two VBM precincts had minor "variances" -- just enough to make them look legitimate. All the rest of the VBM batches matched exactly. Suspicious to say the least. This, combined with the fact that the election results were substantially different between the polling place ballots (where Bernie Sanders one by a substantial margin over HRC, 58% to 42%, while early VBM ballots had HRC win by the same margins -- 58% to 42%. These are the ones they essentially pre-counted prior to the audit, and made new reports for the batches. Given that they scrutinized the selected ballots so much prior to the audit and assembled the precincts for review by pulling ballots from hundreds of batches, and the fact that new reports were generated, there was any variances at all. That's why we think the published variances in the first two batches were just for show.
We asked that the court rule that Vu must go back and redo the audit in the primary, but the judge said he would not go back to that election. It was over, "moot" in his mind. Our attempt to stop certification was unsuccessful. At the most recent hearing in early December, the County argued that according to election code section 13314, we could not force them to deal with the current election because any writ of mandate could not be effective because it "will substantially interfere with the conduct of the election." This is a ridiculous claim, as section 13314 applies to ballot statements and anything regarding printing of the ballots, not the processing of the audit. But the county wants to be immune from the courts. They say we can't stop an election in its tracks because they then say we are interfering. And we can't go back later, because then the election is moot. If they get their way on this point, then the courts will be all but worthless to get anything done by our out-of-control elections officials.
We think the primary was flipped from Sanders to Clinton using the suspicious VBMs that were pre-stacked by Vu and his staff.
They are certainly dragging their feet, delay delay delay, stretching out actually getting to the judgment as much as possible, apparently so the new president will be sworn in on Jan 20 and our judgment won't be done until after that. How convenient. We will push to get the judgment earlier so we can deal with the current election, and the inevitable appeal.
We want to appeal to get provisionals included, they want the appeal to get VBMs to be made optional, while we want provisionals to be included in the audit. Stay tuned.
Sign-in Roster Review
One of the projects we completed as a team was the gathering of data about the election from the sign-in rosters maintained at the precincts. This work took about four weeks of steady work scanning documents and counting signatures, followed by detailed analysis. In this election, we wanted to test the theory that the SDROV processed Crossover ballots as provisionals, thereby processing them last, subjecting them to additional scrutiny, and omitting them from any audit process. The data does show, however, that the SDROV largely did process crossover ballots as traditional ballots and not as provisionals. We also did some calling of NPP voters to see what happened to them. We noted some precincts had about half the NPP voters requested crossover ballots and voted for the president in the Democratic party, while in other precincts in the same area, none did. From our phone calls to voters, it appears these differences were due to precinct worker errors rather than reality. This is not to say there is not continuing concern that NPP voters were not treated correctly in the primary, as the county made it very difficult to get the crossover ballot, requiring votes to say "Democratic Crossover Ballot" exactly in some cases. But many voters we called also said they had no problem at all and the process was easy.
It is important to mention the many people who helped with this process, including Patricia Gracian, Christine Mose, Joie Kincaid-Peringer, Daniel Osman (who supplied some computers and helped with logistics), Josephine Piarulli, Ramona Irwin, and Pat Herron. (Let me know if I missed listing someone). These are true patriots who are dedicated to clean elections and this data will be useful for years to come, and was an important addition to the evidence in the trial.
Other Election Lawsuits
Citizens Oversight worked on the primary to file a number of lawsuits with Attorney Bill Simpich. One, against SOS Alex Padilla and several cases to "contest" the primary. But sadly, Sanders "gave up" prior to the convention and the conventions were held very early this season (several months earlier than normal) making any legal recourse moot. Also, it was unclear how to proceed on these cases so they sit idle at this time unless we have any bright ideas.
Snapshot Protocol was developed based on our experience in San Diego, once we realized that we needed a file to be able to make sense of the manual audit and make it more than just theater. We extended the request to all counties in California in 2014, with a few counties very cooperative with our requests but most blowing us off. This election season, we strove to extend the Snapshot Protocol
to the top 175 counties in the nation, comprising 50% of the voters in the country. But many of these counties either do not have audits of any sort or are still using only all-electronic voting machines with no trail to audit. in others, we only scratched the surface on getting volunteers lined up to help. But we did sign up more than 200 volunteers to help in districts around the country.
Well, we did not complete the goal, but we did make substantial progress in at least extending the notion of carefully watching election audits to a few other states, and in more counties in California. With the initial action from the court under our belt, we got a bit more respect from some of the counties and some did change their methods based on our recommendations.
I was able to tour the districts in Southern California during initial VBM processing.
One state we made inroads in was Florida, because they had their primary on August 30. One specific race was of particular interest: Debbie Wasserman Schultz vs. Tim Canova in the district largely including Broward County (Ft. Lauderdale area).
We learned that audits in Florida are particularly weak, as they choose only one random race, and then 1% of the precincts comprising that race in the manual tally audit they perform. And, if there is any recount (which is automatic if there is any race within 0.5%) then there is no audit at all, and the recount is a machine recount and scrutiny only of under and over votes (with no actual review of the paper ballots by hand). The most important thing here then is the selection of the random race. If you want to fix one race, then all you need to do is make sure the race selected for the audit is not that race and you are home free.
We must thank especially Margaret Carlock in Citrus County as she helped get the state organized and gathered up the times and dates of the audits, and filmed both Citrus and Marion counties. This was very helpful. I took a trip down to Florida after the General Election and toured several counties, video recording and documenting a recount in Pinellas County, and also in Miami-Dade Counties, met with Tim Canova, and captured apparent election audit fraud in Palm Beach county.
To support this work, we created:
This year, we had an unprecedented opportunity to witness recounts in several states, thanks to the Jill Stein campaign and the Green Party, while HRC stood by idly, telling her volunteers not to object to anything, and the Trump campaign actively worked to derail the recount. Election integrity activist John Brakey of Tucson, AZ (Audit AZ
fame) went to WI so I decided to go to MI to see what I could find out. I'm not sure if anyone else would have noticed the specter of the uncountable precincts. I talked to the reporter from the Detroit Free Press in Macomb County and tipped him off on the significance of the unrecountable precinct issue, and it was then taken up by reporters in downtown Detroit where about half of the precincts were unrecountable. Interesting was the fact that in the precincts actually recounted (and incorporating computer results for "unrecounted" precincts, HRC was up 75,000 votes over DJT. The results published by the SOS in Michigan included only the counties where recounted was at least started. Only a few counties finished the process. At that point, the court pulled the plug because the Trump attorneys pushed that Jill Stein was not an "aggrieved party" because there was no way she could win the election no matter what, even though they were charging her $125 per precinct for the recount. Yet another example of why our election system is a mess but those in charge (like the Trump campaign) like it that way, so it is very hard to clear up.
There is still a lot to do to finish gathering up the information from this trip, but here is my report thus far: Michigan Recount 2016
John Brakey and his team in WI discovered that most recounting there was only done by rescanning the ballots through the same machines again. But worse, he discovered that the most recent ballot scanning equipment includes cell-phone connectivity which means those scanners could potentially be contacted by hackers who could alter their programming, or the data from the scanner could be intercepted by a "man-in-the-middle" vulnerability known to exist in cell phone networks. So anytime you hear that these machines have no internet connectivity, we need to also ask if they are outfitted with SIM cards to allow them to connect to mobile networks.
In 2008, I devised an approach to processing elections that takes the best of hand-counted paper ballots while leveraging the internet for crowd-based oversight and constant checking of the election. The Open Ballot Initiative
is based on the creation of images of all the ballots with cryptographic security measures so any change of ballots will be detected, making any form of "stuffing the ballot box" impossible. These images can be processed by any software (or by "hand", reviewing them on a web site) so as to determine the vote expressed on each ballot. With this in place, we can easily perform 100% audits of any election without any appreciable cost (or at least far less than the $3.5 million price tag on the bogus WI recount).
We want this to be our goal for the future in election integrity. The industry is tending in this direction anyway. There is some interest by the Git Hub
software community in the idea. Also, there is some work on a standard Cast Vote Record that can help standardize the data across all election districts. Sounds good, but everything is a double-edge sword. Standardization also means any vulnerabilities can be more easily be exploited. But with the Open Ballot Initiative
, there are fewer opportunities for such hacking because the system includes constant 100% audits by third parties.
To support this work:
Another angle we are working on is to push for legislative changes, particularly in California to:
- clean up audits
- frozen (snapshot) results before random selection in a machine-processable format.
- open random selection using random number generators (like dice) and a list (rather than the easier to hack raffle method).
- No re-scanning ballots (to generate a new report) when variance are reported (as they do in LA and many other counties which does more to cover up fraud rather than expose it).
- Improved audit reports, including disclosure of what ballots are covered by the audit.
- Provide for better citizen involvement
- Random selection meeting must be able to be video recorded by anyone.
- Loosen rules on video recording/photographing ballots if the identity of the voter is impossible to determine.
- Support Open Ballot Initiative
- Future election management equipment must produce ballot images, and create standard Cast Vote Record.
- Use cryptographic mechanisms to secure the ballots such as SHA digest of each image and Block-Chain digest of ballot set.
- Ballot images data and CVR data shall be made available to the public prior to certification.
- Since use of this approach supports 100% independent audits of the election, 1% manual tally is not necessary, but can be replaced by statistical checks that image data accurately represents the physical ballots.
Since the emergency shutdown on January 31, 2012, Citizens Oversight (sometimes under fictitious name Coalition to Decommission San Onofre) has been working hard to make sure the plant did not restart, that ratepayers are treated fairly in the settlement, and that 3.6 million pounds of high-level nuclear waste is not stored only 100 ft from the seawall. On one of these, the permanent shutdown of the plant, we can report that the result comports with our intentions. The plant is indeed shutdown and defueled, and no new nuclear waste is being produced. (YAY)
On the other issues, we have not been so lucky YET. However, it does appear that things are starting to go our way. This is good news.
During 2016, the CPUC "reopened the record" on the settlement and asked for parties to comment (again) on the settlement deal. After ALJ Bushey worked on it for a while and then retired, they finally say we should all talk it over again. The last time we met in this fashion, the utilities had a settlement deal all concocted -- fait accompli
-- and we were asked to join the settlement which was not open to any changes at all. Of course, that settlement was the deal struck by CPUC President Peevey in his secret meeting in Warsaw Poland with SCE executives. They had a bogus settlement hearing where in 3 hours we considered the 3.3 billion deal, and Peevey lost control and told Mike Aguirre to "Shut up" and said "I don't have to answer any godammed questions." The deal was approved about two years ago right before Peevey retired. Citizens Oversight participated in these proceedings as the Coalition to Decommission San Onofre (CDSO) and submitted a request for reconsideration of the decision jointly with Ruth Henricks (represented by Aguirre & Severson LLP) which was apparently permanently lost in the history, with "petitions for modification" of the decision being the main modus operandi for pushing for changes to the decision. Earlier this year, the record was reopened and parties submitted more comments. Not until ALJ Bushey left did anything happen, and it is more discussion of the settlement among the parties, still ignoring our motion for reconsideration. Here is the most recent ruling: http://www.songscommunity.com/docs/171205944.pdf
This coming year, it looks like we will get a fresh bite at the apple to get ratepayers a fair settlement. Still talking to the players to figure out what our best course of action will be, but whatever it is, this will be big news (albeit probably suppressed by energy-stock invested news outlets).
Nuclear Waste Dump on the Beach
The main approval for the new semi-permanent nuclear waste storage facility, called an iss-fissy (ISFSI which means "independent spent fuel storage installation") occurred at the California Coastal Commission Meeting in October, 2015. This meeting was hardly even publicized by the utility at their "Community Engagement Panel" which is more about keeping people distracted rather than providing real engagement. We objected at the CCC meeting and filed a lawsuit with the help of the Aguirre Severson LLP law firm to stop this ridiculous idea. Our lawsuit is based on the fact that SCE (Southern California Edison) did not explore other alternatives, such as moving the fuel to the Palo Verde plant near Phoenix (which already stores nuclear waste without the danger of the coastline). We also like several other options, such as moving it three miles inland within Camp Pendleton, and locating it in the California Desert near a railroad line so the fuel can be moved without as much danger of an accident (our "Fishel CA" Proposal).
The judge initially ruled against our lawsuit based on administrative details but now is allowing it to proceed. This is good news as we now have a chance to block this project. An important agenda to help on this project is to get City Councils, County Boards, and other elected officials to create letters of endorsement that they do not support locating the nuclear waste on the beach at San Onofre without first carefully reviewing all options, even if those options seem to be slightly more expensive or require additional paperwork with the NRC.
Citizens Oversight has been involved in a number of additional projects, such as:
- Video recorded and documented the Aliso Canyon methane blowout.
- Assisted with the March Against Monsanto, San Diego -- We had a great march in 2015 but the 2016 march was largely overshadowed by the Bernie Sanders event also in town on the same day. There are a lot of leftover T-shirts and other supplies for one more event of this kind but I think it has run its course and we should concentrate on other projects. Decision making and logistics for this event (which has been a lot of work) will be left up to that committee.
- Save Ecpac - We worked to provide a plan to reopen the East County Performing Arts Center. The City of El Cajon countered with a proposal to turn it into a church venue, with the universally hated Rock Church (just ask anyone who lives near any of their locations) selected as the proposed "anchor" tenant. Were it not for the 35 year low-cost lease, additional building for Rock Church offices and other violations of church-state separation, there is no legal reason to block is (although there are common sense reasons why it is a bad idea. We sent them a letter giving them guidance on how the Rock Church could use the theater and allow others to use it too. Since then, they have made no progress although the City Manager says the roof has been repaired. Now going on 7 years closed for no reason.
- Grossmont Healthcare District -- COPs member Allan Goetz has been doing a great job providing citizens oversight of this local governmental organization. They have sort of come around a lot from the early days. Many of my proposals are now being embraced and we have a good reputation with them whereas in the past, I was a pariah. Allan has also become a representative attending the hospital board of directors. Allan has been successful in bringing up key issues such as quality of service at the emergency room and avoid side-issues. If you want to pursue oversight of any body, Allan has first-hand experience and tips of what has been effective.
- Police Oversight -- Participated in documenting the fatal shooting by police of Alfred Olango in El Cajon. We are pushing for the use of body-cams (which reduce complaints by 93% in other studies) and an independent citizens oversight board of police practices in El Cajon with Subpoena power.
- EMI emissions from Sunrise Powerlink in Alpine -- Since we worked against the Sunrise Powerlink years ago, we revisited the issue now in Alpine where the high-voltage transmission lines are buried under the main street in town.
- Submitted a complaint about the use of federal HUD money to support the religion-based East County Transitional Living Center (ECTLC).
Our work on Election Integrity
, our Election Audit Lawsuit
and the Snapshot Protocol
have been supported by our fundraising efforts which have done reasonably well this year. As we have been keeping our costs to a minimum, our fundraising thus far has been used only for out-of-pocket costs for the lawsuits, travel, and events. We will be sending out confirmations to our donors in the early part of next year. We have been fortunate to have some very generous donors this year which should enable us to continue the pursuit of our agenda.
Organizational Plans for the New Year.
- Membership platform -- we have installed a comprehensive platform oriented to nonprofit organizations (called Civi CRM) on our web server. We hoped have this up and running this summer but the installation took longer than expected, and the election forced us to put this project aside for the new year. The nice thing about this platform is each member will have dashboard to control subscriptions, donations, and other involvement. It also supports event management and much, much more.
- Regular Meetings -- We've been using the Zoom web conferencing solution which allows up to 50 people to be on the same call, with screen sharing and anyone can speak. This has been a great help in doing the limited organizing we were able to do for the Election Team. We need to set these on a more regular basis and also a local meeting place would be a help, particularly as we are interested in helping anyone react to current events with open dialog in the public space.