A while back, I wondered: “How long can the equilibrium of technically incompetent rulers lording it over technologically advanced societies be maintained?”
Possibly not much longer at all. Via InstaPundit, we find Robert X. Cringely’s diagnosis of the touchscreen-voting … uh, situation (emphasis added):
In the case of this voting fiasco, there was a wonderful confluence of events. There was a vague product requirement coming from an agency that doesn’t really understand technology (the U.S. Congress), foisting a system on other government agencies that may not have asked for it. There was a relatively small time frame for development and a lot of money. Finally, the government did not allow for even the notion of failure. By 2004, darn it, we’d all have touch screen voting.
Oh, and there are only three vendors, all of whom have precisely the same motivation (to make as much money as possible) and understanding (that Congress would buy its way out of technical trouble if it had to). This gave the vendors every reason to put their third string people on the project because doing so would mean more profit, not less.
One definition of insanity is doing the same thing over and over again, somehow expecting a different outcome. In this instance, the issue isn’t whether Diebold and the other vendors were insane (they aren’t), but whether the government is.
Cringely’s analysis is worth reading in its entirety for its insight into how project management in IT usually works, which is to say, not well.
Returning to the problem at hand, here’s your geeky analogy of the day: if this doesn’t get fixed in the next few months, the effect on American democracy will be analogous to acid rain from a cometary impact turning Earth’s oceans into vitriol — the annihilation of the base of the marine food chain. The reliable exercise of the franchise is the base of the electoral food chain. Cast enough doubt on enough results and the legitimacy of every elected officeholder dissolves like the calcareous shells of so much phytoplankton doused in nitric and sulfuric acid. Publicly-funded bureaucracies suddenly and explicitly become the tools of an arbitrarily-chosen oligarchy, and one that doesn’t even know enough to demand elementary accountability, at that.
Others have suggested that the election of 2004 may be pivotal, as important as, say, 1860 or 1932. Since most such 72-year cycle arguments rely on astrology or numerology, I remain unconvinced. But unless the homeostasis characteristic of American society operates on the voting security issue, universal suffrage could become, at least temporarily, a bad joke in the next eleven months. Not knowing who’d been elected President until December 13th last time around was one thing. What if we didn’t know who’d been elected to anything 5 weeks after the election, and had no good way to find out?
Jay — Good catastrophe scenario. Better founded than most. The geeky analogy works for me.
The 72 year thing can be projected back, pretty much, one more iteration to the initiating of the Constitution. I actually kinda/sorta believe in the cycle, though not in any rigid way. It just seems that our institutional arrangements need a major overhaul every three generations or so.
On the main point, we should be moving to even LOWER technology. We should have a law mandating paper ballots where you put an X in the box with a pen. How the Hell else do you know whether anyone actually voted the way they meant to? I’m serious. So what if it takes an extra day to count. It will just make the whole process more tense and interesting, and easier to check and recount.
Lex, we use that system right now in NE FL and it works well. The ballots are scanned into a computer, but there’s a paper trail for verification. –s
I still think electronic balloting is a good idea. However, it looks like it will take some time to make it work. As usual, the human systems, rather than the technology per se, are the problem. As Jay points out, the problem is exacerbated by the general incompetence of the political people in charge of making reforms.
In the early ’80s I met a lady who had voted in the first post-dictatorship election in El Salvador. She showed us her little finger, which she had been made to dip in indelible ink as a condition of voting, to prevent multiple voting. It was then a couple of weeks after the election and her finger still showed the ink. At the time I wondered why they couldn’t use a similar control method in Chicago. I assume the reason was that too many people benefited from vote fraud, and that this is one of the reasons why it is still difficult to implement effective controls.
Even though fraud now benefits, I think, mainly Democrats, it appears that Republicans are reluctant to do anything about it. Perhaps this is because they want to avoid similar scrutiny if they come to dominate local politics, or perhaps it’s because they want to avoid scrutiny on other issues WRT which they are relatively vulnerable. Either way, I don’t think we can expect the parties to be effective in this area.
I’m with Scott on this one and, unusually, against Jonathan. There is no reason whatsoever to have electronic voting.
There is no need to go high or even medium tech on this. This is a classic case of technology being adopted for its own sake.
I was a poll watcher in two Chicago elections. I saw no vote fraud. The ballots used the metal poker-type thing. Those ballots were quick and easy to count by machine. But all the problems with fraud, etc. would have arisen from the human element, and that won’t change, and the more “modern” technology actually facilitates fraud, a major problem, in exchange for minor even trivial efficiency gains. For example, it took about ten minutes to count the ballots at the precincts I visited. But it would only have taken a few hours to count paper ballots by hand even with no technology at all.
And we are at the point where an optical scanner could read sheet-fed ballots with a nice big X written in pen by the voter, making hand recounts easy and accurate, no hanging chads. Then, feed the ballots again to make a pdf record after the election is finally over and there is no need to do a recount, to keep a record, and pitch the paper ballots.
This is a place where retro- low-tech is the way to go.
There is much fraud. Your personal experience doesn’t refute this fact. There are two broad categories of fraud: ID fraud (multiple voting, inappropriate people voting) and ballot fraud (stolen, altered or destroyed ballots, polls held open beyond closing time).
Electronic voting isn’t likely to be useful against ID fraud, except in instances of people attempting to vote multiple times under the same name, but then neither are paper ballots. OTOH electronic voting can be made highly robust and relatively tamper resistant. The problem is that current electronic voting systems are badly designed. (Anecdote: I voted electronically Nov. 2, and the system didn’t give me a receipt to prove that I had cast a vote.) There is no technical reason why there can’t be an electronic system that 1) records my vote, encrypting it using a one-time key (so that it’s indecipherable except to the counting computer), 2) gives me a machine-readable receipt, with a printed private key that would allow me to login to the counting computer and confirm that my vote was registered correctly, 3) creates an electronic audit-trail (so that, for example, votes cast after polls close could be detected). These are just hypothetical examples that I pulled out of the air; I’m sure there are other ways to do it, and other benefits.
The point is that there could be a lot of benefits from electronic voting if the systems were thoughtfully designed. So far the systems have not been thoughtfully designed. The flaws of these systems have been widely discussed on the Internet for some time, but the pols have failed to follow up and get the problems fixed.
We all use electronic banking machines without thinking twice. The stakes are high but the technology has proven itself. There are benefits to getting rid of the old, paper based systems. The technology for electronic voting is not necessarily more complex than that of the ATM, but it has been designed by incompetents and developed by people who aren’t going to lose a lot of money if they screw up. I think that the remedy is to change the development incentives and keep working on the technology, not to keep doing things the old way.
It’s a big country. Electronic voting technology is in its infancy. Let it continue to be developed and let jurisdictions experiment with it if they wish. There are costs to continuing to do it the old-fashioned way. Well designed electronic voting systems may have benefits that make them worthwhile. It’s too early to write them off.
The project-management lesson here is that missed requirements are costly, and major missed requirements are devastating. The likelihood of such incidents approaches unity when the requirements-gathering session consists of a bunch of technical illiterates elected from gerrymandered districts, waving vague concepts of “computerization” like so many magic wands. I don’t, of course, think that the US electoral system is going to collapse next year. But continued practice of borderline rule-by-decree by ignoramuses in Congress virtually guarantees the occasional interesting disaster, and I can’t help but wonder if they’re all going to be survivable.
I concur that continued development of electronic voting in a non-prescriptive environment is likely to result in a much more robust system in only one or two election cycles; just one change, the addition of a printed voting record of some kind, would go a long way toward fixing what’s wrong right now.
“There is much fraud. Your personal experience doesn’t refute this fact.”
What? This wasn’t my point at all. My point was that the technological “improvements” would make it EASIER rather than harder to perpetrate fraud if there was fraud.
“The point is that there could be a lot of benefits from electronic voting if the systems were thoughtfully designed.” Without being flippant — what benefits? I just don’t see it.
There are various instances where we absolutely requires a paper record. For example, a signed original of a will. If the stakes are high enough, we demand something tangible. Choosing who gets to exercise government power belongs in this category. So I see it.
I think you are assuming that the current electronic voting systems, which are first approximations, are the way electronic voting has to be. My point is that this isn’t true. The current electronic voting systems are flawed in part because they don’t give the voter a paper record of his vote. There is no reason why such flaws can’t be remedied.
Well designed electronic voting systems can be more robust than simple paper systems. They can provide secondary paper records and have robust audit-trails and can do it in a way which is more resistant to fraud than are paper-based systems. (How do you steal or modify ballots when the ballots are encrypted and you can’t tell what choices voters made? How do you steal ballots or create bogus ones when there’s an independent, encrypted electronic audit trail authenticating each ballot?) Electronic documents, properly protected via well understood electronic encryption and signature techniques, can easily be made more resistant to tampering than are paper documents. Citing the practices of the legal profession, which is notoriously behind the times technologically, is not convincing.
The problems with electronic voting are technical and ought not to be difficult to solve once we have more experience. The benefits of well designed electronic voting systems should be: 1) less fraud, 2) greater precision and 3) more rapid availability of results.
Note that the problem of authenticating voters remains, no matter how we record and count votes. But that’s a separate issue.
BTW, here’s a photo (open it in a new window if you want to read the instructions easily) that I took in my rather New Labor-ish electronic voting booth Nov. 2. Notice the painful adherence to multi-lingualism, accompanied by a disturbing lack of attention to such minor details as feedback and controls (e.g., you don’t get a printed receipt). I don’t see why it shouldn’t be possible to do better.