Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
The Voting Machines Need to Go
The voting machines America is currently using need to go. They breed mistrust, and there’s a reason for that: They are not trustworthy. They are also inherently undemocratic.
No, I’m not talking about 2020–well, not as such. I’m talking about all the past elections since we started using these machines, as well as all the future ones until we come to our senses.
These Machines Are Antithetical to Democracy
Let’s do the second problem first. Here’s a question that pretty much answers itself:
Can a system be democratic if the people are not permitted to know how their votes are counted?
It’s not a democratic system unless the people have freedom to vote and have some understanding of how their votes are counted. So why are we using machines that have secret coding to count the votes? That’s one reason why the voting machines need to go: Their coding is kept secret. Get rid of those machines, and replace them with paper and pen, or with machines that use open-source software.
A related question also pretty much answers itself:
Can a system be democratic if the people are not even capable of knowing how their votes are counted?
Again, it’s not a democratic system unless the people have freedom to vote and have some understanding of how their votes are counted. Under the current system, only people with years of training in just the right areas of computer science are even capable of understanding how votes are counted. So why are we using machines that use fractions to count votes, instead of just simple arithmetic?
And why are we using machines that condemn us to ignorance about how a vote-count is going until the software is updated? Why are we using machines that guarantee an election cannot be run smoothly unless election workers are properly trained in the new technology? These aren’t online Trumpy conspiracy theories, mind you–this is the New York Times!
And why are we using machines where you can’t know how your vote is being counted–or understand much of what anyone even says about how they’re counted–without understanding what a router is, whether it’s connected, what a LAN is, what a firewall is, what a batch is, what packets are, what packet captures are, what an IP address is, what logic and accuracy tests are, what a client is, what a server is, what domain controls are, what computer logs are, and how to read those logs?
That’s another reason why the voting machines need to go: Their workings are an inherent mystery to the voters.
Get rid of those machines, and replace them with paper and pen, or maybe with machinery no more complex than the Scantron machine you remember from high school. Use open-source technology that doesn’t need a lot of software updates, doesn’t require any fancy training, and isn’t a mystery to anyone–or, at a bare minimum, that isn’t a mystery to your neighbor, uncle, or friend who at least knows computers and engineering enough to program a machine to count. I don’t know how to do that, but it makes a big difference that people I know can understand such technology. But no one I’ve ever met knows what’s going on inside these voting machines.
They Are Not Trustworthy
To some extent, you need only read the above.
To a greater extent, the massive untrustworthiness of these machines is best understood by carefully reading what Dominion Voting Systems actually says about their machines, reading an important report from a committee in the Michigan State Senate, and letting G. K. Chesterton remind us what sanity looks like.
Hath it not been said by Dominion Voting Systems?
Voting systems are, by design, meant to be used as closed systems that are not networked (meaning not connected to the Internet). It is technologically impossible to “see” votes being counted in real-time and/or to “flip” them.
Were the machines incapable of online connectivity, would Dominion not simply say that?
Well, many vote-counting machines really do have online capabilities, including some but not all Dominion machines and not only Dominion machines; see this NBC News story, for example. But, more importantly, see page 22 of this report from a committee of the Senate of Michigan:
A report from the Michigan Senate Oversight Committee:
Upon completion of the election, tabulators print the final results on paper. Clerks then connect a modem and transmit by secure, cellular connection or transfer by flash drive the unofficial results to the county clerk.
The Report then says the second-most important thing I’ve ever read about voting machines, consigning it to a footnote in miserably small font size: Voting machines (not including Dominion products in Michigan, apparently) have internal modems, but don’t worry:
…they are not turned on until the polls are closed and tabulation has concluded.
So first the machine counts the votes, then it prints the results on paper, and only after that is it allowed to access the internet for the fast, electronic reporting of provisional results. Or, for some of the machines, to be accessed by a USB drive for the same purpose.
In other words, the technology is secure if we keep all the modems off till the right time. Our elections are secure if we use the technology correctly.
Hence some wisdom from Father Brown, G. K. Chesterton’s mystery-solving priest in a book named after his wisdom, is the most important thing I’ve read about voting machines. Father Brown is explaining “The Mistake of the Machine” in a story bearing that name, and he says this:
“You always forget,” observed his companion [Father Brown], “that the reliable machine always has to be worked by an unreliable machine.”
“Why, what do you mean?” asked the detective.
“I mean Man,” said Father Brown, “the most unreliable machine I know of.”
The voting machines are not safe if they are not used properly. The only way for them to be used properly is for there to be safeguards in place–starting with something on the order of rules in 50 states saying no such machine may be used to count votes until poll observers from both parties plus government officials swear in writing that the modems are switched off.
If those rules were in place, we’d have had their existence and important drummed into our skulls by every fact-checking punk from CNN to the state governments. But that hasn’t happened. Therefore, the voting machines are not trustworthy.
We should listen to the feds–if only just this once. Specifically, we should listen to the Election Assistance Commission, a US government agency, laying down VVSG 2.0 standards–“Voluntary Voting System Guidelines 2.0”–right here. Those rules require “systems to be air-gapped from other networks and disallows the use of wireless technologies.”
That’s right–not only do voting machines with online connectivity exist, but even the feds recognize that this is a problem, and they tell us we should get rid of them.
A broken clock is once twice a day, and once if it’s a federal government clock on military time that broke down in the afternoon. But now is that time.
Published in Elections
Part of a long, long series of trying to figure things out. Here’s some more of it:
Not to mention floating point variable declarations.
Paper ballots are the way to go. I believe France does this, and gets election results in a reasonable time frame. If the media has to wait, too bad.
I don’t think open source is sufficient. There is still the question of which source version was actually compiled into the machine which is running, and operational procedures (such as connecting or disconnecting a modem) that may be properly followed or not.
And yet, ironically, under the current system the media has to wait for weeks.
Open-source would help. But it’s not sufficient. Many things would help but are not sufficient.
I pretty much agree – but you and I both know that it just ain’t gonna happen. Connectivity is here to stay: For that matter, even if I were wrong about connectivity, will you really trust an electronic voting apparatus?
In 2004, all of my friends on the left were convinced that Bush stole Ohio by rigging the voting machines there.
Paper ballots and purple fingers.
Of course, France is also a lot smaller, and many ballots people deal with in the US can have a large number of candidates, measures, etc.
That’s something I’ve had trouble getting people to understand. Especially the software part. I’ve written programs even on business systems 40 years ago that modified themselves while running. I don’t mean just processed configuration data etc, I mean that changed the actual code of the programs. It made future operation more efficient than processing the same data each time.
My Like on this Comment is the first. It is reserved for actually Liking the Comment once I read it if I feel like it.
Say it? They should be ready, even eager, to demonstrate it.
How many states have purchased these machines? Were security audits conducted? (And if they weren’t, why weren’t they?)
The noises those bozos make about their software being proprietary is just too funny. It’s tabulating software; it counts things!
I don’t even trust pen and paper. But I’ll trust people if the processes are right. Maybe even a connectable machine!
But I’d be pleased to see any improvement. A simple bill saying something like “The State of Texas shall henceforth adopt VVSG 2.0 standards for our elections” might be possible in a couple dozen states. At the county level, citizens can take action modeled after the Parents Revolt and go Office Space on the bad machines.
Figuratively, of course.
Until it’s done. Then go get the machines at the city dump and make it literal. And put it on the internet.
Bigger than Alabama? Alaska? Arkansas? Utah?
Elections are a state thing.
Which implies no absentee ballots, no 2 weeks of early voting, etc etc.
Oh, I should include this again:
California alone has almost half the population of France.
I meant by state. You can find state ballots with over 100 things to vote on. Judges, state reps, commissioners, various new laws…
Ok, cool. I was just thinking of population size while focusing on your first clause.
Excellent post @saintaugustine! I suspect that elimination of the arduous hand-counting was the impetus behind adopting them- at least at the local level (that’s giving giant benefit of the doubt!). But how do you convince the average voter it’s actually progress to do something by hand? And the human element is far from fool or crook-proof. Many large metro areas seem to produce more votes than voters (when needed). I’m with you on all your major points though. I just wonder whether the Secretaries of State are convinceable?
Beats me. All I got is the oldest tools–logic, rhetoric, and prayer.
But prayer doesn’t overrule free will, I’m not good at rhetoric, and logic does not easily persuade cultivated stupidity.
So I’m left trying to figure out and explain the truth where I can. Maybe the future will listen even if hardly anyone today does; worst-case scenario, if that’s all I’ve done then maybe I’ve done my duty, which is enough.
Still–surely a (peaceful) revolt of citizens at the county level can work to reform things, and surely some states at least can take some simple and obvious steps like VVSG 2.0. I suppose I could mail my own state and local officials a copy of this post.
And surely some voters can understand that less room for corruption is progress even if it’s lower-tech.
So if there are, say, 30 million ballots in California – assuming less than half the population actually votes – with 100 issues/candidates/etc, that’s 3 BILLION total votes. And most of them are probably not just Yes/No. They’re going to be a list of names…
I don’t know if 3 BILLION votes can be counted by hand in time for the 11 o’clock news.
Actually I wouldn’t be surprised if the total number of real voters drops by quite a bit, perhaps half or even more, if people have to vote in person on Election Day.
And I would consider that a Good Thing. People who can’t be bothered to vote on Election Day can be safely ignored, as far as I’m concerned.
First, that’s better than the system we have now. The computers had it all ready by 11:00 AM three or four weeks later.
Second, yes–they can be counted by hand quickly if they have enough counters and a good process.
Third, it’s worth a delay if it secures more trust and more trustworthiness.
Fourth, even in Cali it’s not a 30-million situation. It’s the counties that count, isn’t it?
Sure, and partly because the counties will be voting for different Commissioners etc. But all of the counties have more people in them now than in the 1940s or 50s, and they’re voting for more offices and measures too. These days, counties seem to have trouble getting enough people involved even with the machines doing most of the “work.” Getting people once again used to showing up for these functions could be difficult in many areas. It didn’t fall away in one swell foop, and it probably can’t be restored instantly either. And what happens if you get mostly leftists volunteering to do the counting, because they know it’s not the people who cast the ballots who really matter, it’s the people who count them?
Well, yeah. If we want to have votes counted honestly–with trust and trustworthiness–then we need good processes, and people helping, and some time to change things.
If we can have, over some time, even that much local civic engagement by decent people with basic math skills, then the country’s lost anyway. We can’t screw anything up worse by trying to do better.
Then we have the same corruption we have now.
Before we started using all this computerized technology to count the votes, we’d get results the same night.
Aren’t computers supposed to make things more efficient?
Nah, they don’t even need to wait for the returns to know it was the fairest election, ever.
As well those too incompetent to obtain a photo ID, or to register to vote a couple months before an election.
No two weeks of voting. No absentee ballots except for overseas military, state department folks stationed overseas, etc.
Sure they can. When you break it down to precinct level it’s totally doable.
I don’t see how voting machines are inherently more or less secure than human counting. I don’t actually observe my vote being added to the tally in either a system. The security of either one is predicated on the design of the system, the implementation, and the honesty of the humans involved.