Hacker News Reader

Comment by llambda | Hacker News Reader

What I've read[1][2] in regard to this incident is that it is most likely a calibration issue: i.e. the touch screen is improperly calibrated and as a result is not selecting the proper region of the screen. Now this is concerning because it likely means other machines could be or are miscalibrated. However the important takeaway here is that this is not some malicious attempt to rig the vote. If that were the case the likely method would be completely invisible from the UI; why would an attacker bother to actually show a user they were being manipulated? Of course, they wouldn't.

[1] Joseph Hall comments here, also provides a link to further commentary by him: http://gawker.com/5958114/an-expert-weighs-in-on-that-viral-...

[2] http://www.theawl.com/2012/11/the-truth-about-voting-machine...

View on HN

Unfortunately no, if you read what he wrote in the comments of that submission:

Being a software developer, I immediately went into troubleshoot mode. I first thought the calibration was off and tried selecting Jill Stein to actually highlight Obama. Nope. Jill Stein was selected just fine. Next I deselected her and started at the top of Romney's name and started tapping very closely together to find the 'active areas'. From the top of Romney's button down to the bottom of the black checkbox beside Obama's name was all active for Romney. From the bottom of that same checkbox to the bottom of the Obama button (basically a small white sliver) is what let me choose Obama. Stein's button was fine. All other buttons worked fine.

The obvious question to ask: Why is that part not shown in the video?

Edit: Just watched the video again. The video suspiciously freezes from 11s to 18s. Probably, something inconvenient being edited out amateurishly?

It's too bad that the video doesn't provide any evidence supporting that story.

The video would be truly compelling if he had repeatedly tapped from Romney to Stein. As it is, the video appears to show miscalibration, and only his testimony suggests fraud. If the video showed what he claimed, it would be lawsuit material.

That still doesn't establish malice.

That doesn't actually mean anything. Resistive and infrared touchscreens (especially cheaper ones) can have all sorts of hilarious calibration errors where part of the screen area works perfectly fine, but other parts of the screen are off completely.

I used to work on kiosk software for restaurants. At the start of every day we had to go through all of the touchscreens in the restaurant and calibrate them. Sometimes they would lose their calibration halfway through the day and need to be recalibrated multiple times.

The consumer electronics industry has shipped hundreds of millions of touch screens in the last few years and i've not seen a single article complaining that they opened up their new tablet/phone to find it mis-calibrated and that their touches were off my such a huge margin. Why are voting machines apparently subject to calibration problems that other touch screen devices are not?

Miscalibrated devices shipped to consumers get returned, resulting in extra costs to the retailer or manufacturer. Miscalibrated voting machines don't.

1. Because they probably have to standardize on an older model and can't push out newer models at will.

2. Because these machines are open to the general public who are hitting the screens with their meaty fingers all day.

Most likely it's the difference between capacitive versus resistive screens. If you purchased your phone based on the lowest bidding vendor, I'd bet there be a lot more problems with it too.

Not all touch screens are created equal.

I remember the old Nintendo DS touch screen, which was resistive, required user calibration where you would press on the corners of the screen. If you missed the corners (which, with the relatively thick stylus and the relatively thick bezel on the original DS, wasn't uncommon), the accuracy of your touch positions would be very notably off, to the point that using touch-screen keyboards in some games would be difficult.

Obviously I don't know for sure if this was the exact case with this voting machine, but the point being, some touch screens do require calibration and there is a chance for error. Or, of course, it could simply have been a factory mistake.

Either way, I doubt there's any malice behind this error.

I have owned a ton of touch screen devices. None of them were miscalibrated, even after years of use.

I have used a ton of touch screen devices in public. I've encountered quite a few that were miscalibrated.

Public devices put up with abuse that your private devices do not.

Honestly curious - what kind of abuse could cause a calibration error?

My kids abuse my iPad pretty heavily but generally smacking on the screen with their hands and fingers a lot but I still have never seen it mis-calculate a touch.

iPads used capacitive touchscreens which aren't prone to the same sorts of calibration issues that resistive touchscreens are.

I don't know much about electronic voting machines but I wouldn't be surprised if they mostly used resistive touchscreens (which are historically much cheaper to produce).

iPad uses a capacitive touch screen. Older technologies are very prone to mis-calibration, even if they are not in heavy use.

I recently had to help my mother recalibrate her sewing machine's touch screen.

If you ever used a Palm Pilot, you've seen this sort of calibration issue. Given the state of government procurement, they're likely using that cheaper, crappier tech in most areas.

This is an example where technological ignorance is dangerous because it leads to the wrong conclusions.

Multi-touch screens as used in smart phones use capacitive touch sensing technology, they do not require calibration.

Cheaper, older touch screen systems (such as those used with ATMs or voting machines) use resistive touch technology, they absolutely require calibration.

Don't fly much I'm guessing. It's the rare checkin kiosk that is accurately calibrated.

The touchscreens you see in modern consumer electronic devices are capactive touchscreens. I think the iPhone was the first major device to ship with a capactive touchscreen, prior to that, resistive and infrared touchscreens were a lot more common (and cheaper).

If you've ever interacted with a touchscreen at an ATM, mall kiosk, or airport kiosk, it's likely that they were resistive or infrared based, and more susceptible to calibration errors.

Yes, very likely a calibration issue. There was a similar instance occurring here in Colorado [1], but in the opposite direction (a vote for Romney went in as Obama). The machines definitely need to get better, but it is probably not fraud or a hack of any sort.

[1] http://www.chieftain.com/politics/pueblo-county-voting-machi...

> why would an attacker bother to actually show a user they were being manipulated?

Because the person who wishes to tamper with the outcome of the election only has access to (or, very likely, skillz to modify) the calibration template (which has to be reconfigured each election, hence is modified by more people), not to the entire source code and tool chain of the voting machine. Most attacks are not gigantic conspiracies with unlimited resources.

Reply on news.ycombinator.com