Re: One would have throught...
Define 'too close'. Kissing distance, then you may swap spit and virus particles. In bluetooth range, and you'll probably be fine. [...]It's all just technofetishism.
You are suffering from 'one-bit mind'. This is where people assume that quantities which need to be represented by real numbers (or collections of real numbers) can be represented by a single bit, which is either true or false. It's a common problem among computing people: I suffer from it too.
In this case the quantity you care about is risk: what is the risk I have been infected with COVID-19. This is a real number, between 0 (no risk) and 1 (definitely infected). And you're trying to do things to lower it. And if you know that you have been close enough to someone who is infected that your phone talked to their phone over bluetooth then, that risk is higher than it would otherwise have been. And that's what an app can tell you, and that's why these apps are at least potentially useful.
You could argue of course that in fact there is a single bit of information here: am I actually infected? Well, for a start that's not really a single bit for various reasons (how many infections die out early, are you infectious, blah blah blah), but even disregarding that the problem is that without extremely aggressive testing regimes this is not something you can know, so you need to use a probabilistic proxy – a real number – for it, and keep that proxy as low as you can.
Of course aggressive test-and-remove regimes are a way of uncovering the underlying single-bit, and they are therefore better than the app approach, but, again, this is not a single bit: testing regimes are better, apps are worse, but this is not the same as saying that apps have no value, especially when very aggressive testing regimes are not available.
Biting the hand that feeds IT
