The professional privacy crowd is being intentionally obtuse about the possible uses of location data in epidemiological tracking, as demonstrated in this white paper. Thread: https://twitter.com/granick/status/1247983480572534790

As the paper explains, you phone is a surveillance device that works three ways. It talks to cell towers, it has a GPS thingy in it, and it potentially interacts with its environment (by bluetooth, wifi, photographing a QR code, etc) in ways that help determine location

The technical argument against contact tracing right now is that the resolution of these devices is not sufficient to figure out contact history. The phone knows you are at South Station, but not who you passed within six feet of. So it's useless, right?

But the point is we can use this data IN CONJUNCTION with all other methods, and we can use it to reduce the search space so that other, more labor intensive methods (like interviews) can be brought to bear. We also live in a surveillance society with things like camera footage

The correct way to look at this technical capability is not as a magic surveillance solution that replaces the hard work of contact tracing, but an additional tool, *already deployed at scale internationally*, that we can put in the hands of epidemiologists and doctors

In particular, phone based contact tracing can retroactively answer vital questions like "did anyone who attended this party that was a spreading event then fly to another city? Where did they go while infectious?" https://www.nytimes.com/2020/03/23/us/coronavirus-westport-connecticut-party-zero.html

It doesn't matter that your phone doesn't give pinpoint location data. The rough location data may already be enough to help investigators cut the possible number of contacts from tens of thousands to hundreds. And that is the kind of capability we need as we move to containment

Note that *in particular* this implies that measures that just do contact tracing, while trying to keep location ambiguous or undetermined, are not sufficient. We need the maximum resolution data these devices are already broadcasting to advertisers, platform owners, the GRU, etc

The more general problem with privacy advocates' arguments against technical measures is that they are not being made in good faith. They have a desired conclusion ("this doesn't even work!") based on their policy preferences, and they fit their thinking to meet it.

I'd be very curious to hear from people in the epidemiological world whether a data stream that let them track long-distance travel by infected people, and generate a list of potential contacts based on even rough proximity data, would be useful in containing this disease

There are all sorts of privacy arguments you can make for why we shouldn't do this kind of contact tracing. I'll fight you, but they're strong arguments! The technical arguments, however, are weaksauce. Anything that helps prune a list of potential contacts is better than nothing

A final point needs emphasis. The debate isn't how we use these technologies right now. There is no point to doing that until we've reached a state like Taiwan or South Korea, where it's back to tracking individual cases, and want to reopen the economy and resume normalish life

The upshot is this: for the first time in human history, we have a working, automated real-time and retroactive location tracking tool for something like 80% of the US population. Do we make it part of our pandemic response, or strictly limit it to data brokers and advertisers?