That's always a bit of a dodgy metric, because of two issues: The simpler issue, that people are more likely to bring up, is that you need infinite memory (or,

That's always a bit of a dodgy metric, because of two issues: The simpler issue, that people are more likely to bring up, is that you need infinite memory (or, if you're a member of the Church of Church, you need to be able to represent infinite symbols). The tricker problem is that, since all computers are analog while all theoretical formulations are discrete and symbolic, Turing-completeness is a good-enough-is situation. You can't really build a finite state machine or a lambda applicator, so what it means for something to be a computer isn't as theoretically robust as you'd hope. It seems that, in practice, a computer is something a computer scientist regards as a computer (i.e., relative to the things they expect to be able to do with it, and the ways they expect to interact with it). I say this because of the weird question of whether CSS is Turing-complete. On the one hand, it seems that existing proofs of concept (e.g., [1]) require the user to play tippy bird with key inputs. On the other hand, real digital computers require a clock to play tippy bird with register commits. To what extent does the clocking of a digital computer have to be "out of sight, out of mind" for people to agree that it's a computer? Also, of course, CSS has to enumerate memory positions (i.e., DOM elements), but then a digital computer has to do the same, so the extent to which one cares about infinite memory is put in the spotlight. [1] http://eli.fox-epste.in/rule110-full.html http://eli.fox-epste.in/rule110-full.html