A measurement cannot have such great precision that the inaccuracy in the measurement is shorter than the plank length.

That is not known to be the case. Got a reference for that?
It's also something entirely different from suggesting that space is discretized in Planck-length units, which is certainly not the case. In fact, it's a fundamental postulate of QM that the wave function is smooth and continuous (and hence, so is the location-probability distribution). If it wasn't continuous, then you'd end up with undefined momentum.

I would think that the number of people who have a cell phone with no contract, but have broadband and a computer on 24/7, is vanishingly small.

I've got a prepaid cell phone (I use maybe 20 minutes airtime per month, it's mostly for when I'm out on jobs or actually want someone to be able to reach me when I'm not home, then I forward the landline to it). At least one computer on my LAN is always on (and is connected by DSL through an ISP whose TOS doesn't object to my servers so long as I don't do anything stupid). My total connectivity (cell, landline, DSL, ISP) is about US$95/month.

But in general, you're probably not too far off.

That isn't necessarily true. It's just as possible people are wasting time fixing unimportant issues and ignoring more important ones.

I'm not trying to disparage the OpenBSD team or anything. It's just that no development team is perfect.