I am surprised that when I do a Google search for 'COVID-19 "google timeline" ', I see essentially no results. We all know now what COVID-19 (nCov, Coronavirus) is. From the reports I see, this virus has the unusual characteristic of being very contageous for up to periods of weeks prior to a person's feeling symptoms. That is presumably how a large fraction of a cruise ship became infected. This is quite ominous. If epidemiologists were to ask a patient, "tell us where and when you've been each minute over the last 15 days" the vast majority of these victims wouldn't have a prayer of providing that information. And even worse, finding the other people who were "in that AM/PM at 5:07-5:21 10 days ago" would be essentially impossible. Or potentially dozens of other involved locations, over those 15 days. Before one of you accuses me of "advocating" the use of Google Timeline to track potential cases of COVID-19 by means of Google Timeline, I don't need to "advocate" it. Rather, I simply point out that there are a lot of people out there scared of this virus, and probably be looking for a way to determine if their paths have crossed with a victim, even if that victim wasn't symptomatic for 1-2 weeks after the contact. So at some point, I think there will be discussion of this possibility. The data is already being collected, in all Android phones (is there an Apple equivalent?) In principle, if a new infectee is identified, it would be technically possible to work backwards, figure out where he has been over the relevant period, and find anybody who was close to him during a multi-week period. One reason this could be important is that there may be a drug which might reduce (or, hopefully, eliminate) a person's contageousness if it taken during this pre-symptomatic period. One possibility is an old anti-malarial drug, chloroquine, which I have mentioned before. https://www.ncbi.nlm.nih.gov/pubmed/32074550 But if people only begin taking chloroquine when they begin exhibiting symptoms of this new flu, that means that they will be spreading that virus for as much as two weeks, or even more. In principle, hundreds of people could be infected, directly or indirectly, merely because there is no early warning. Suppose you receive a text or email notification that you were in a small store, 5 days ago, with a person who just developed symptoms of COVID-19. You MIGHT be infected. So, you MIGHT want to take a chloroquine pill. (The half-life of chloroquine is 45-55 days). Or some other pill that could assist if taken long before symptoms were likely to appear. Not only might you not get sick, maybe you'd be able to avoid transmitting the virus to many others. (My speculation...) And maybe you'll live, when you otherwise wouldn't. We Cypherpunks are SUPPOSED to be more concerned, than average, about the privacy and freedom implications of technologies. What I have described, above, might be handled in a completely-voluntary fashion. But, we want to ensure that this doesn't turn into a permanent form of tracking. So we should debate the implications of all this, ideally before everyone else is talking about it. I would be surprised if Google isn't already considering something like this. They have much of the data to do so. They might hesitate to announce such an idea, for fear that people would think this is some sort of generalized people-tracking system. Jim Bell