Thread. "Kris Kobach, Kansas’s top election official, recently declared victory in the highly publicized Republican gubernatorial primary in Kansas, surpassing his opponent (sitting Gov. Jeff Colyer) by about 300 votes." via @tytinvestigatestytnetwork.com/2018/08/17/kob… 1/
2/ "At 9 p.m. on August 7, election day, with 10 of Johnson County’s precincts reporting (and just one other, smaller county left to report), Colyer led Kobach by 44 votes in the state and by 13 percent in the county."
3/ "But at 9:07 p.m., a reporter for the Kansas City Star posted that Johnson County would not display further results for at least two hours due to a 'computer glitch.'"
4/ "At 5:50 a.m., the county finally published its remaining unofficial results, putting Kobach na"rrowly in the lead."
5/ "Johnson County later explained that the problem involved its new ExpressVote touchscreen barcode ballot-marking system, made by Election Systems & Software, LLC (“ES&S”). More than a week later, ES&S has yet 2 explain the precise cause of the problem." kansascity.com/news/politics-…
6/ The recommendation to buy these machines came from Johnson County Election Commissioner Ronnie Metsker, who Kobach appointed in January 2016. shawneemissionpost.com/2018/08/10/met…
8/ Before the board of county commissioners accepted Metsker’s recommendation, election-integrity advocates and independent computer science experts expressed serious concerns about the security of these machines. shawneemissionpost.com/2018/05/16/ele…
9/ "As reported by the Shawnee Mission Post in May, the ExpressVote produces a paper “summary card” containing a computer-generated barcode. The barcode, which humans can’t read, & thus can’t verify, is the only part of the summary card that is then scanned & counted as your vote
10/ @marilynrmarks1, who spearheaded the still-pending paper-ballot suit in Georgia, told TYT that “the official vote that is counted is a barcode—not a human-readable vote. The Johnson County voter has no idea what vote they actually cast."
11/ As Marks further explained, "Voters can’t read barcodes, nor can they know how the machine is programmed to print or interpret barcodes.”
12/ Here is my own article about the dangers of touchscreen barcode ballot markers (like the ES&S ExpressVote purchased by Johnson County), including quotes from IT experts expressing significant concern about them as well. medium.com/@jennycohn1/st…
13/ Although the barcoded “summary cards” also include computer-generated text, Kansas (like many states) does not require manual audits and will fund a candidate’s request for a recount only if the margin of the victory is less than 0.5 percent. ballotpedia.org/Recount_laws_i…
14/ I'm told that the state will not pay for a a HAND recount (as required to detect electronic vote tally manipulation) even if the requesting candidate falls within .5 % margin. Rather, it wld be a useless machine recount. I am still investigating and will report back.
15/ Source showing that Kansas does not require post-election audits of ANY sort, much less post-election HAND audits. verifiedvoting.org/state-audit-la…
16/ Given the absence of mandatory MANUAL audits and meaningful state-funded HAND recount laws, Kansas is unlikely to use the text on the ExpressVote "summary cards" in a manual audit or manual recount.
17/ This is a problem "bc a manual audit or manual recount is the only means 2 verify the legitimacy of electronic vote totals, since ES&S & other private vendors block forensic analysis of their equipment based on the 'proprietary' nature of the software" tytnetwork.com/2018/08/17/kob…
18/ "A further issue is that the results of any such manual audit or recount would be unreliable due to the inability to know whether voters reviewed the text to verify that the machine didn’t print the wrong selections or omit some races entirely."
19/ In Johnson County, poll workers apparently gave conflicting instructions as to whether voters should undertake such a review. fox4kc.com/2018/08/08/joh…
20/ The ExpressVote training manual is equally unclear, describing the “post print verification” of the “summary cards” as “OPTIONAL” and stating that the machines can drop the cards directly into a bin without voter verification. verifiedvoting.org/wp-content/upl…
21/ Per Marks, “Even if a ... voter attempts 2 verify that the [ExpressVote] card is a reflection of how she voted, there are very few voters w/ 100% recall of everything that was presented to them on the ballot," so that "most... do not & cannot verify that the print out...
22/ ... is actually what was on their ballot. Simply put, the [ExpressVote] text summary of what voters purportedly voted is almost worthless from an auditing perspective. Voters must simply trust that the barcodes haven’t been erroneously or maliciously programmed.”
23/ "Another problematic aspect of the ExpressVote process that occurred on Aug. 7 is that the candidate names couldn’t fit onto a single screen—voters had 2 push the 'More' button 2 see them all. This problem also ... confuse[d] voters in Johnson County." kshb.com/news/local-new…
24/ In addition, all touchscreens, including the ExpressVote, "typically cause longer lines than hand-marked paper ballots because they limit the number of voters who can vote at once to the number of touchscreens at the polling place." verifiedvoting.org/testimony-of-v…
25/ Indeed, Metsker attributed last week’s long lines to the county having an insufficient number of machines. kansascity.com/news/politics-…
26/ Touchscreens can also disenfranchise voters if, due to a glitch or denial-of-service attack, they fail 2 work at all. In Johnson County, “some of the new machines simply didn’t work..., leading some voters to give up & go to work. Poll workers seemed unsure of a backup plan.”
28/ ES&S, which accounts for about 44 percent of U.S. election equipment, is no stranger to controversy. trustthevote.org/wp-content/upl…
29/ ES&S (previously called AIS) 1st raised eyebrows in 1996, when its chairman, Chuck Hagel, resigned a few weeks b4 announcing his intent to run 4 the U.S. Senate & then won the race by 15 points, despite pre-election polls calling it a dead heat. tytnetwork.com/2018/08/17/kob…
30/ With his unexpected landslide victory, Hagel (ES&S chairman until a few weeks b4 announcing his candidacy) became Nebraska’s 1st Republican U.S. senator since 1972. Election officials said ES&S machines had counted about 85 % of the votes in the race. tytnetwork.com/2018/08/17/kob…
31/ In 1997, ES&S’s founder, Bob Urosevich, became a VP of voting machine vendor Global Election Systems. His brother Todd is a vice president at ES&S. Bob and Todd’s sister, Sueann Devereaux, née Urosevich, works in ES&S’s programming department. tytnetwork.com/2018/08/17/kob…
32/ In July 2000, Global promoted Urosevich 2 president. Jeffrey Dean, a convicted cyberfelon, became a Sr. VP in Sept 2000. Dean had previously “pleaded guilty to 23 counts of embezzlement involving sophisticated manipulation of computer …records..." tytnetwork.com/2018/08/17/kob…
33/ Several months later, a Global voting machine lost 16k Gore votes in Volusia County, Florida, during the 2000 presidential election. The Volusia error was discovered only bc an alert poll worker noticed Gore’s vote total going down. tytnetwork.com/2018/08/17/kob…
34/ In 2001, Johnson County bought Global's AccuVote-TS touchscreen voting machines. The county’s election commissioner at that time, Connie Schmidt, appeared in marketing materials for the vendor, along with officials from Georgia, Maryland, & California. tytnetwork.com/2018/08/17/kob…
35/ In the marketing materials Schmidt says that Johnson County was the "first county in the Midwest to deploy Diebold's innovative touchscreen systems." (Note: This presumably was published after Global changed its name to Diebold in 2002.)
36/ Diebold acquired Global in January, 2002. In April 2002, the machines miscounted hundreds of votes in six different races in the Johnson County. Diebold reportedly “later attributed the glitch to a software error.” tytnetwork.com/2018/08/17/kob…
37/ Diebold told the AP that Jeffrey Dean left the company in 2002. But election-integrity advocate and author Bev Harris obtained Dean’s court file, which included internal Diebold memos showing that Dean had remained as a consultant. tytnetwork.com/2018/08/17/kob…
38/ In 2003, a review in Ohio “reported security flaws” with the type of machines used in Johnson County. But Schmidt—who has served as the co-project manager for the U.S. EAC's Election Management Guidelines”—said she was not personally concerned, shawneedispatch.com/news/2003/dec/…
39/ Schmidt said she was unconcerned bc the voting machines didn’t connect to the internet. She did not apparently mention that all voting machines receive programming before each election from centralized computers that can & often do connect to the net. tytnetwork.com/2018/08/17/kob…
40/ In Aug. 04, the U.S. CERT issued a Cyber Security Bulletin re: Diebold’s GEMS central tabulator, stating that “a vulnerability exists due to an undocumented backdoor account, which could [allow] a local or remote authenticated user [to] modify votes.” tytnetwork.com/2018/08/17/kob…
41/ The Guardian later reported that Dean, the convicted cyber-felon, had programmed the GEMS central tabulator system and that it had “counted a third of the votes in the Bush-Kerry [2004] election in 37 states.” tytnetwork.com/2018/08/17/kob…
42/ In 06, IT experts reported that Diebold's AccuVote TS is “vulnerable to extremely serious attacks.... [A]n attacker could...create malicious code that spreads … from machine to machine during normal election activities—a voting-machine virus.” tytnetwork.com/2018/08/17/kob…
43/ In 06, Voters Unite compiled a FIFTY-ONE PAGE “partial list of documented failures” involving election equipment from ES&S, Johnson County’s current vendor, which was founded by the same man who hired a convicted cyber-felon while at Diebold. tytnetwork.com/2018/08/17/kob…
44/ In 07, IT experts reported that “a voter or poll worker w/ a Palm Pilot & no more than a minute’s access" to ES&S's iVotronic touchscreen voting machines could surreptitiously "recalibrate" them to "prevent voters from voting for specific candidates... tytnetwork.com/2018/08/17/kob…
45/ … or cause the machine to secretly record a voter’s vote for a different candidate than the one the voter chose.” This would require “no password, & the attacker’s actions ... would be indistinguishable from the normal behavior of a voter in front of a machine...
46/ ... or of a poll worker starting up a machine in the morning.”
47/ In 09, ES&S acquired Diebold & assumed the Johnson County contract. It was thus on ES&S’s watch that the County’s election system caused a long delay uploading the results of the 2016 election, a disturbing prelude to the August 7 debacle. tytnetwork.com/2018/08/17/kob…
48/ Because the AccuVote-TS (which Johnson County, KS used in 2016) has no software-independent paper record of voter intent, there was no way to confirm the legitimacy of Johnson County's results in the 2016 election. tytnetwork.com/2018/08/17/kob…
49/ Commissioner Metsker (Johnson County's Commissioner) could have "righted the ship" by implementing hand marked paper ballots with post-election manual audits and a secure chain of custody between Election Night and the audit.
50/ Instead, as in Sedgwick County, Johnson County Commissioner Ronnie Metsker chose the ExpressVote, ES&S’s latest touchscreen. Due to this decision, the precinct totals were provided courtesy of the unverifiable barcodes on the ExpressVote "summary codes."
51/ As discussed, the text portion of the summary cards is a wholly UNRELIABLE record of voter intent & is also unlikely to be used in a manual recount or manual audit bc Kansas does not require manual audits and makes it incredibly difficult to get a state-funded manual recount.
52/ But the ExpressVotes likely also produced paper vote-tally reports when the polls closed. If so, those reports could at least show whether the reported totals correctly reflected the precinct totals (regardless of whether the precinct totals were themselves legitimate).
53/ Here is what the vote tally reports look like from an Accuvote TSX in Georgia. If the ExpressVote touchscreens used in Johnson County don't produce something similar listing the vote tallies I would be surprised.
54/ TYT has made a public records request for any such vote total reports & has not yet received a substantive response. But Kobach rejected a prior similar request by Dr. Beth Clarkson, a Kansas statistician, 4 Sedgwick County’s 2014 vote tally records tytnetwork.com/2018/08/17/kob….
55/ Clarkson sued Kobach for the vote tally reports, but lost in the district court (where she represented herself) & filed an appeal (using an attorney this time). Although the appeal was briefed and argued LAST YEAR, the court has yet to issue a ruling. tytnetwork.com/2018/08/17/kob…
56/ Even Georgia, which is not exactly known for its election transparency, publicly posts its vote tally reports at the precincts on Election Night. What is going on in Kansas?!
57/ Meanwhile, it seems that Johnson County has not yet paid for the ExpressVote machines. It remains to be seen whether the county will return or keep them. tytnetwork.com/2018/08/17/kob…
58/ For now, like his predecessors, Metsker is standing by his vendor:
“We need to give them space to make the corrections & I am assured that they will find that and that we will all be happy with this equipment for years & years to come,” Metsker said. tytnetwork.com/2018/08/17/kob…
Study shows that people of all political persuasions are willing to modify their beliefs based on corrective info from reliable sources, but “subjects ‘re-believed’ the false info when retested a week later.” 1/ news.northeastern.edu/2018/06/18/tir…
2/ The author of the article says It may help to warn people in advance that they are likely to forget the correction bc “this helps them mentally tag the bogus information as false.”
3/ It’s also “important that the corrective information be repeated as frequently, and with even greater clarity, than the myth.”
I hate to be the bearer of bad tidings but elections have been electronically suspect starting long before the Trump/Russia scandal. This article is lulling folks into a false sense of security, which is dangerous. Domestic hackers & insiders were always an equal threat. 1/
I agree, tho not enuf time (and 0 political will) to do this in Nov. Wish it were different. For now I hope to stop states from doubling up on electronics w/ touchscreen ballot markers. Using electronics to count votes is bad enuf. Having them mark our ballots too is nuts. 1/
Nuts except for those who are unable to hand mark their ballots. Once you have hand marked paper ballots they can be either scanned or hand counted (my preference) or both. 2/
Any time u put a machine between the voter and the paper record of voter intent there is an opportunity for programming mischief. Here is just the latest example.: 3/
I’m hoping some of the cyber experts who signed the letter about the risks of using cellular modems to transfer election results can answer this question. Thx! @philipbstark@SEGreenhalgh@rad_atl@jhalderm
Seeing as no one has answered yet, I will say that even if the cellular modems CAN be configured to bypass the internet, we should not have to blindly trust that vendors or whoever else is hired to set them up will do that.
Kathy Rogers, the face & voice of @ESSVote, which has installed CELLULAR MODEMS in tabulators in WI & FL, is cozying up to @DHSgov which refuses to advise states to remove the modems despite a letter from 30 cyber experts & EI groups stating it should do so. #CorruptElections 1/
The notion that cellular modems affect only “unofficial” results is bogus bc, among other reasons, in certain jurisdictions, unofficial results become the official results once added to absentees & provisionals—sometimes w/o ever comparing them to the precinct results tapes! 1/
And Wisconsin doesn’t even require that counties publicly post the results tapes so that the public itself can make this comparison! (I don’t know about Florida, Michigan, & Illinois.) 2/
Thus, we must simply trust that someone trustworthy is conducting this due diligence. In Johnson County, Kansas, the County acknowledged that it does NOT conduct this basic due diligence. 3/