The Green Party’s candidate for California Secretary of State this year is a strong proponent of Internet Voting. David Curtis, who was also the Green’s 2010 candidate for Governor in Nevada, is now running to replace CA’s term-limited Democratic Sec. of State Debra Bowen and he’s staking out a position that contradicts computer scientists and security experts who warn that online voting cannot be done securely.
Recently, I had a “conversation” with Curtis on Twitter about his advocacy for Internet Voting, after Steven Dorst, an election integrity advocate who follows The BRAD BLOG on Twitter, responded to a tweet of Curtis’ declaring his support for an optional “online method of voting”.
My enlightening conversation with Curtis follows in full below.
The BRAD BLOG has long been on record documenting the many dangers of Internet elections. While the ease of hacking such elections is certainly a major concern (among many others), the over-arching problem with Internet Voting is that, after voting is complete, it is 100% impossible for citizens to oversee their own election results in order to determine that any vote has been tallied as per any voter’s intent.
As with electronic touch-screen voting, which is 100% unverifiable in any form, even when done securely — if there is even a way to measure such things (and virtually all computer security experts have told us for years that there isn’t) — Internet Voting can never been done in such a way that the citizenry can know that its been done securely. Thus, no matter how “secure”, Internet Voting is ultimately a threat to confidence in elections and, along with it, representative democracy in the U.S.
Nonetheless, Curtis, like at least one other candidate in this year’s SoS race in California, disagrees. He strongly advocates in favor of Internet Voting…
Curtis suggested, during our conversation on Twitter, that voting online can be both “secure and verifiable” with votes printed out onto PDFs or “other paper-like substance” and by allowing individual voters to check a database with a Personal Identification Number (PIN) after an election to see if their own vote was recorded accurately. That option presents a number of problems, such as the possibility of buying and selling votes, as I mentioned during the conversation. More importantly, it misses the point that an electronic database available to a voter can show the voter one thing, while the actual recorded and reported result can be entirely different. Unless every single Internet Voter in the state checks their results and then reports it publicly somehow (thus, violating their right to a secret ballot), such a method of “verifying” ones own personal vote offers no way for all voters to know that all electronic votes were accurately recorded and counted as cast.
The Green Party candidate is not alone among California Secretary of State candidates in advocating for Internet Voting. Democratic candidate and state Sen. Leland Yee, voiced strong support for Internet Voting upon declaring his candidacy, though he has largely (or, at least, publicly) backed off his advocacy after being informed of the many concerns about IV from computer science, security and e-voting experts.
Democratic candidate and state Senator Alex Padilla, believed to be the front-running among the pack of SoS candidates to date, has not explicitly endorsed Internet Voting schemes to my knowledge. Though, as we reported last year, Padilla has been less than forthright concerning his advocacy for unverifiable electronic voting systems. He misleadingly shepherded a radical new election reform bill (SB 360) into law last year which ended the long-standing requirement for federal testing of new e-voting systems in California and granted unprecedented power to the Secretary of State to approve new e-voting systems for use in “a legally binding election” even without testing by state authorities either.
Among the leading Democratic candidates, only former Common Cause official Derek Cressman has avoided vocal support for Internet Voting or other electronic voting tabulation schemes, focusing his candidacy instead largely on campaign finance reform and related government transparency issues in the wake of Citizens United.
On the Republican side, neither Pete Peterson nor Dan Schnur, a former GOP strategist and spokesperson who is running this year as an independent, are on record in regard to Internet Voting at all, to my knowledge.
Thanks to California’s new “Top Two” primary system, the two leading vote-getters from any party — including from the same party — during this year’s statewide primary on June 3rd, will go on to face each other in the General Election this November.
My Twitter conversation about Internet Voting with the Green Party’s Curtis — and with Dorst, who helped kick off the discussion — follows below.
PLEASE NOTE that each remark is not always directly responsive to the one before it, since the conversation went off in several different directions at a time, as various tweets were replied to even as comments were made in regard to a different response, etc. It’s impossible to perfectly translate that “three-dimensional” conversation into an absolutely linear narrative here, but I have tried to order the conversation below with as much accurate context as possible. Moreover, to make the conversation a bit easier to follow, I’ve taken the liberty of lightly editing some of the tweets to remove necessary Twitter abbreviations and/or to clarify what each person was responding to. The original tweets, in any case, are each linked along with each remark. Any misleading ordering or edits are unintentional and would be solely my fault, where and if they exist.
DAVID CURTIS: I support multiple modes of voting. I would support an (optional) online method of voting that is secure and verifiable.
STEVEN DORST: And there’s the rub. I have grave doubts that online voting can ever be both secure AND verifiable. And Auditable. And Recountable.
CURTIS: I think it can be done, we can set up voter accounts with PINs. Just like a credit union account.
DORST: That doesn’t address the Auditable or Recountable criteria. And if I vote that way, how can I know my vote was counted as I intended?
DORST: Check out @TheBradBlog. He’s convinced me that paper ballots, marked by hand & counted in public, in precinct, is necessary.
CURTIS: Your account would show your votes 24/7. The SOS database would report from the accounts.
BRAD FRIEDMAN: Nice. So I could buy or sell my votes 24/7 under that plan as well?
CURTIS: No, but you could save the 30 minutes in line. SO there would be a “savings”.
FRIEDMAN: If my vote is “verifiable” online, what stops me from selling it? OR showing it to my boss so I don’t get fired?
CURTIS: Selling it to whom?
FRIEDMAN: I can offer $ to folks who can prove they vote how I wanted them to. I can beat my wife if she doesn’t.
DORST: And if your vote is anonymized to allow recounts, how do you verify it was anonymized correctly?
CURTIS: Same way the credit unions do it, partial account number of receipt, or could be random generated code id.
CURTIS: Paper would still be an option, online voting would be an optional mode, like with registration.
FRIEDMAN: So only SOME votes would be 100% unverifiable then?
CURTIS: Do you bank online?
CURTIS: Is the money in your credit union verifiable or unverifiable?
FRIEDMAN: Sounds like you haven’t studied elections enough to understand the difference between banking and voting. Secret ballot makes voting totally different.
FRIEDMAN: Online banking is totally different. All transactions are 100% overseeable forever by all parties. Secret ballots are totally different entities.
[For more details on this point than I was able to make in 140 characters, please see Dr. David Jefferson’s “If I can shop and bank online, why can’t I vote online?” (PDF). Jefferson is a computer scientist at Livermore National Laboratory, as well as the Chairman of the Verified Voting Foundation and a longtime voting systems adviser to both Republican and Democratic California Secretaries of State.]CURTIS: In order to have an online transaction an alternate mode would be required.
CURTIS: The blocks to online voting are based on current statutes, and would have to be modified to allow online voting.
CURTIS: If an online vote was created it would have an opt in that included new statutory language.
FRIEDMAN: There are reasons (good ones!) for the statutes you want to change. Among them: Security, Overseeability, Buying/Selling votes.
CURTIS: The existing modes of voting stay, it would be an additional optional mode.
FRIEDMAN: “Optional mode” for whom? Do I get the option of deciding whether 100% unverifiable votes are cast my election??
CURTIS: If a voter opts in and it is viewable only by them via PIN, it is not equal to being viewable by everyone.
FRIEDMAN: So when I “opt in” and my boss requires I show him how I voted, what’s your plan for that intimidation?
FRIEDMAN: When I, as a citizen, want to oversee the tally, how can I EVER know that any online vote was registered as cast?
CURTIS: Secret code, like when grades are posted publicly.
CURTIS: The SOS [Secretary of State] would have the tally just like now.
FRIEDMAN: “The SOS would have the tally just like now” — The SoS doesn’t “have the tally”. It’s tallied locally and sent to the SoS. Who counts the BALLOTS in your plan for Internet voting?
DORST: Moreover, how do you REcount the (nonexistent) ballots?
CURTIS: It would be a database, with a back up.
DORST: A database only allows the election to be re-tallied, not re-counted.
CURTIS: The transaction would generate a PDF (or other paper-like substance.)
FRIEDMAN: Please speak to ANY of the EXPERTS in Internet Voting so they can explain the MYRIAD reaons we don’t/shouldn’t/can’t do it.
CURTIS: Online voting would also address accessibility issues.
FRIEDMAN: What “accessibility issues” are you referring to?
CURTIS: I did an online vote for the CC [Green Party Central Committee] in Nevada.
FRIEDMAN: What do you man, “did an online vote”?
CURTIS: I facilitated an e-vote for the Green Party of Nevada. They elected their executive council, online, with Ranked Choice Voting.
[Ed Note: I didn’t even respond to Curtis’ reference to Ranked Choice Voting (RCV) — also sometimes known as Instant Runoff Voting (IRV) — which is something long supported by the Green Party, despite the legendary complications it presents to citizen oversight of vote counting. When applied to actual public elections it makes it virtually impossible for precinct-based counting to take place, since it usually requires several rounds of counting to take into consideration the voters’ 1st choice, 2nd choice, etc. for candidates. The tallying process is also very difficult for both voters and candidates to even understand, much less oversee, which is why many RCV/IRV tallying schemes have been abandoned in localities after they have tried to use them. The Green Party and other “third-parties” often support RCV/IRV under the generally well-intentioned (if arguably misguided) belief that it allows for such parties to have a better shot at breaking the two-party duopoly that exists in most U.S. localities. It was for that reason that, years ago, we supported the general notion of IRV, until investigating further and discovering all of the complications and dangers, only some of which I mentioned above, that go with it.]CURTIS: The e-transactions were printed to PDF, so we could do a recount if necessary.
FRIEDMAN: PDFs are not ballots nor verifiable by voters. Please talk to some e-voting experts and/or read CA Sec. of State Debra Bowen’s “Top-to-Bottom” review of state e-voting systems. [Ed note: A 2007 state-sponsored study by world-class computer science and security experts that uncovered innumerable and gaping security issues and flaws in every single e-voting system and computer vote tabulation system used in the state, leading to the decertification of many of those systems in California and elsewhere.]
CURTIS: Will do. Perhaps what I am not making clear is this mode would require statute language and tech that is not in place.
FRIEDMAN: No. That was clear. But a VERIFIABLE Internet election also involves TECH that doesn’t exist, unless you’ve got a secret. 🙂
CURTIS: People like to say “no”.
FRIEDMAN: By “people”, do you mean world class computer science, security and Internet Voting experts??
FRIEDMAN: Or by “people” do you mean voters who would like to be able to oversee their own elections?
CURTIS: All the people of earth, they like free speech.
FRIEDMAN: Again, the problem is that ALL citizens need to be able to verify and oversee ALL votes, ballots and results in an election. They can’t do that with Internet Voting.
DORST: Sounds like you should invite @dc_us [David Curtis] to be a guest on The BradCast — WELL before the [June 3, 2014] CA SoS primary!
FRIEDMAN: Would love to discuss having David on the show.
CURTIS: Great, thank you. I would appear on your show.
FRIEDMAN: Great. We’ll try to make that happen in the coming weeks, if possible.
Great work Brad and Steven!!
Sounds like David Curtis may possess enough flexibility and integrity to look a little deeper into this and come to a more reality-based understanding. Hope so!!!
What kind of hack would support Internet voting? 😉
Thank you for the Green Party story.
Applause to the Green Party candidate David Curtis for continuing to lift the Green Party banner as a Green Party candidate.
The Green Party’s Green New Deal is a winner. Green sustainable jobs. Green Party Solar Jobs. Green Party Wind Jobs. Green Party Conservation Jobs. Green Party weatherization jobs! Green Party rail jobs. The Green Party is right. We need More Trains, Less Traffic.
http://www.youtube.com/watch?v=zegUmkpLw_w
Thanks for Green Party story!
I guess ANY press is good press [#3] for the Green Party (who are no doubt used to being ignored by the media). I agree with the Greens on most things, but not internet voting — an AWFUL idea.
The main reasons why Internet voting is not ready for use in government elections are:
1) it is not safe;
2) it will not be safe any time soon;
3) even if it were “safe”, it is not transparent; we cannot know what is really going on inside the computers, and should not trust the results; and
4) being paperless, we cannot independently check or recount the results, so there is no way to prove to the losers that they lost, nor to recover from the inevitable cyber-meltdown.
http://countedascast.org/internet-voting-risks/
I am appalled that someone from the Green Party wants to give the Chinese army, Russian mafia, or, need I say it, the NSA, an opportunity to vote (massively) in California elections.
I can’t believe that we are having this discussion.
If Internet voting can be made secure, can Mr Curtis please tell us all HOW?
I’m sure that the rest of the world would love to know too. After all, several countries are building their own internet cables to avoid connection to the spying 5 eyes.
How can anyone be so naive? And are there more like him?
Scary.
Brad, you wrote the following, and I think the issue is even worse than you describe:
"Unless every single Internet Voter in the state checks their results and then reports it publicly somehow (thus, violating their right to a secret ballot), such a method of "verifying" ones own personal vote offers no way for all voters to know that all electronic votes were accurately recorded and counted as cast."Even if every single voter could reliably verify whether or not his or her own vote was recorded properly, that still leaves two fundamental problems.
1) Phony ballots stuffed into the online ballot box would not correspond to any voter, so know one would be able to complain that they are wrong. Ballot box stuffing would not be detectable.
2) If a voter discovered that his own ballot was recorded incorrectly, he or she would need proof of that to present to election officials. Without incontrovertible proof, but just the voter’s claim, the officials could do nothing. They would (rightly) surmise that in most cases the voter was simply wrong. And if they did not do that, then election officials would be at the mercy of thousands of disgruntled voters who could falsely claim that their votes were misrecorded. So exactly what kind of incontrovertible proof would voters be able to present?
Keep up the good work, Brad.
Sorry for the typos in my previous post. Darn autocorrect on iPad. I don’t seem to be able to re-edit the post. If you are able to edit it, Brad, be my guest.
David Jefferson @ 8:
Fixed the typos. And thanks for pointing out those additional important issues. You almost always do.
And thanks as well for the kind words.
Hi readers, I would like to point out that no where in my platform do I call for internet voting. I am willing to discuss the concept, which I have done. That said, CA Greens is already doing internet voting for their SGA, standing general assembly. It has worked ok so far. I’m not sure what security they use for it but will ask them.
Green Party candidate David Curtis writes @10:
With all due respect, David, you need to do much more than discuss this topic. You badly need to educate yourself on the topic, because the colloquy above reveals that you are dangerously uninformed.
I would strongly urge that you take the time to read the articles that Brad Friedman linked to in this piece. I would also highly recommend watching Dr. David Jefferson’s scientific presentation Electronic and Internet Voting: The Threat of Internet Voting in Public Elections.
Frankly, the degree of ignorance about electronic voting (touchscreen, op scans and Internet), displayed by those who have the integrity to stand with ordinary citizens in the midst of our global class war is both disappointing and frightening.
Slightly off topic. If you will be in the vicinity of My Fair City this Thursday evening, come to a free panel discussion entitled “Voter Suppression, Equal Rights and the Promise of Democracy” at 4:00 in the Tsai Auditorium at Harvard (1730 Cambridge Street). For more information, see http://www.scholarsstrategynetw...mise-democracy.
What I get from reading this story is that (1) David Curtis is OPEN to the idea of internet voting if it could be made secure, but not if it could not be made secure; (2) he has enough intellectual honesty to try to have a reasoned conversation about it while not pretending to be right or intolerant of differing opinions; and (3) Brad is more interested in making Curtis and others who disagree with him look like an idiot, even if it means distorting the conclusions of a very open-ended, imperfect (Twitter-based, for god’s sake) conversation.
In other words, this “story,” and Brad’s spin on it, makes Brad look like a meanspirited jerk, and makes David Curtis look reasonable and open-minded. Probably not the result you were trying for, Brad, but that’s the outcome here.