The state's election department website talks about how it is holding a series of demonstration sessions for counties where voting machinery vendors can show up and present their latest wares. There is a spectrum of scanning and tabulation systems, some more transparent than others. The worst, to me, do not let a voter make his or her own ink mark on a ballot; but print out a ballot that doesn't even look like a ballot. It is a bar-code representation of whatever was just voted. The better systems have clearly readable paper ballots that are scanned by dual independent systems, which not only is a way to audit the initial tabulation but creates a library of digital images of every ballot card and every vote cast. This approach can be used to trace machine errors, quickly examine questionable votes and pull the paper if necessary for adjudication.
Your skepticism about whether counties will be buying a new generation of opaque voting technology is legitimate. In all likelihood, some will and some won't. That same is true with whatever audit protocol is eventually adopted, as there are distinct choices. In general, more opaque systems tend to be welcomed by administrators and vendors who don't want public scrutiny. The opposite is also true. That's where Pennsylvania stands, which is not entirely different from many states and jurisdictions in late 2018.
If anything, this is a moment for concerned citizens to be heard on this.
JB: So, let's look at this pragmatically. The average person who may be deeply concerned about restoring and maintaining some semblance of legitimate elections doesn't really understand all the ins and outs regarding electronic voting. And it has been shown that the various vendors have lied about the machines, their susceptibility to hacking and even whether they are connected or can be connected to the internet. I admit that, after writing on electronic voting since 2005, I'm not altogether clear about it myself, beyond an abiding distrust of this technology, borne out by past elections and faux hacks (like DefCon or Haldeman's infamous hack that programmed a voting machine to spit out the University of Michigan fight song). Vendors are fighting to maintain market share and are notorious for their opaque management of voting equipment. Election officials are not computer scientists. I wonder, how can We the People come out ahead in this scenario?
SR: My answer may not please you. I think there is a balanced place for using the best of paper records and the best electronic analysis to give the public trustable election results. I've looked very carefully in recent months about the emerging technologies and audit protocols; those that can and must be used before official results are certified.
We are in the 21st century. I don't see how anyone rationally thinks it is possible to have have country with multi-millions of voters, and multi-page ballots, have completely manual hand counts--that will be accurate. I have heard the mantra for years that only hand counts can be trusted. That is an old narrative that is not serving the public on many levels. If anything, it is a distraction from spotlighting the pros and cons of the current range of vote count systems. Exhibit A for that conclusion is the recent Florida recounts.
Look at how the hand counts and ballot custody broke down in the biggest counties in Florida's simultaneous three statewide recounts. I suspect many people will stop reading right now. But consider that some of the biggest problems seen in the most problematic counties were a result of technical limitations in the state-certified commercial scanning systems, old Florida laws that never envisioned multiple recounts and yet imposed very short deadlines, and the complexities of literally managing millions of pieces of papers--including some with ink marks in more than one recounted race.
Florida, like Pennsylvania and Michigan in 2016, didn't conduct a complete recount either. It looked at a subset of ballots in a tight window and said, 'Good enough. Election over.' What was seen was another state lacking a system designed to verify that all the votes cast were counted accurately. (That said, some Florida counties have that kind of system, but they were not allowed to use it under the state's dated recount laws).
The best protocols and technology is not hard to understand. They involve
using ink-marked paper ballots (a record that cannot be hacked) and then two independent tabulations to double-check the results against each other. The most impressive approach I've seen uses image-based software in the second scan to ascertain how the votes on the first scan of the ballots was done. That second scan, done by a system with nothing in common with the first except the paper run through it, helps create the most granular library of the ballots (down to the individual machines used in precincts); quickly identifies sloppy or questionable ink marks, and gives anyone observing a ready path to pull the actual paper ballot in recounts and challenges. I've heard people say, 'you can't trust any digital image--it can be hacked.' In the approach I'm describing, the technology is being used to help put the paper in order so there can be a full accounting of every vote. If there's a question or legal fight, you have the quickest access to the questionable ballots--as opposed to looking for needles in a haystack of paper.
Who created this dual system? The answer is a company called Clear Ballot and Ion Sancho, who recently retired as the Leon County, Florida, supervisor of elections--who was the technical advisor to the Florida Supreme Court's 2000 presidential recount that was shut down by the U.S. Supreme Court. I'm not making this up. I'm just describing the best and most granular vote count accounting and verification system I've seen.
Now, Leon County is the gold standard. Most of America isn't anywhere near that. But I can honestly tell you that the mix of paper and technology exists for the most transparent and verifiable elections in our lifetimes. Whether it is used is another story. Who doesn't want transparency like this? A lot of people, including some progressives who distrust anything electronic. We have spent a lot of time in this discussion, but the bottom line is simple. What's the goal here? To account for every vote or be less precise.
JB: There's so much to talk about here. I'd like to bring up two aspects of your last comment. One is the voting machine vendors. Their ownership, proprietary software and machine maintenance continue to be shrouded in secrecy. Why have election officials and the public at large put up with this, all these years?
And, after you answer that, let's talk about those election officials for a moment. I have no doubt that there are plenty of honest ones who also seek accurate vote counts. On the other hand, there have been egregious examples of ballot destruction, often despite state statutes and court orders, that have made any kind of election re-examination all but impossible. I'm thinking specifically now about Brenda Snipes of Broward County, Florida, in the 2016 Democratic primary. Democracy was definitely not served yet there were no ramifications on any front - despite a Republican governor, state's attorney, etc. If election officials can cavalierly destroy ballots or otherwise deflect or defeat voters, where does that leave the voters? Can you talk about this as well?
SR: On the proprietary nature and privatization of voting systems: of course, it's offensive to have a public process held hostage to private businesses where the public cannot see what features could be abused by political insiders. But consider this paradigm shift. If two independent analyses of the paper ballot count was done, that would do a lot to break open the so-called black box.
(Note: You can view every article as one long page if you sign up as an Advocate Member, or higher).
1
1
1
