When the Internet of Things fails

For the past three or four weeks, I've had a blizzard of technological failures. There was a period when my home computer was kernel-panicking regularly; there was a period when my router would intermittently stop working entirely, but only for certain applications on one computer. Then my Time Machine backups got corrupted. All of that stopped happening as mysteriously as it had started, but now my work computer has kernel-panicked twice in the past three days. A week ago, my iPhone stopped syncing properly, and I spent a fair bit of time last weekend poking at it, including a factory reset and restore from an old backed-up backup, but it just kept getting worse; it finally got to the point where it contained 35GB (more than half of its storage space) of “Other”—apparently the remnants of corrupted syncs.

(I finally talked with Apple phone support about that one, and got a completely incompetent guy who told me weird things about having to uninstall five components of iTunes, then asked me what version of OS X I was running, then asked me whether I was on Mac or Windows, then told me to delete iTunes, then got confused by the fact that that's impossible, and then put me on hold and never came back. I kind of think he was an impostor who had snuck into their call center. The next guy who answered the phone said there was no record of my having spoken to anyone, but he connected me to an advanced-support person, and she was very helpful. I ended up backing up to iCloud, which I've been trying to avoid doing for years, and then doing a factory reset and restoring from that backup, and now all seems to be well.)

But that's not what I'm here to talk about today. The other thing that's been up around here lately is that I've been acquiring high-tech home devices, like a thermostat and a doorbell and a door lock. And this past week started me wondering: as my home gets increasingly wired, what happens when these devices inevitably fail to the same extreme degree, and/or in some of the same ways, that my computer and phone have been failing?

I suppose the answer is to keep the lower-tech older devices on hand so I can swap them in if need be. But the science fiction scenarios in which high-tech citywide systems fail catastrophically are seeming less implausible than they used to.

I was already thinking of writing this entry this morning, but hadn't yet gotten around to it, when I happened across a Charlie Stross blog entry from December: Trust Me (I'm a kettle). Regardless of the veracity of the malware-kettle scenario that he opens with, his other scenarios are all-too-plausible. Sf writers have been talking about ubiquitous surveillance (and/or sousveillance) for a long time now, with “smart dust” cameras and the like, but Stross is talking about an intermediate stage, feasible in the very near future if not already feasible, in which it becomes very easy to hide embedded computers in the physical objects whose presence we take for granted. In one sense, this is pretty similar to the kinds of bug-hiding that spy agencies have been doing for decades; but now the bugs can be much much smarter, and can acquire information of different types (like a log of all the keys you type on your computer).

Interesting times.

Join the Conversation