Vector Clocks: So what time is it?
distributed-computing-musings.com
distributed-computing-musings.com
Special relativity.
Just because newtonian physics is a sub part of GR, doesn't mean "GR" is the best way to describe basic F=ma.
The smallest subset theory that incorporates the need for vector clocks, and the most unique part that introduced and required vector clocks is special relativity - which is what thats what's most appropriate to discuss and not GR.
The key insight of General Relativity is that in addition to, tidal effects of gravity cannot be transformed away even in arbitrarily accelerating reference frames. This is not particularly related to vector clocks.
Vector clocks, roughly speaking, build upon the idea of Lamport Clocks. This isn't to say that one type of clock is better than the other. The choice between them depends on the task at hand and involves a set of trade-offs.
Anyway, Leslie Lamport, who is considered the founding father of distributed systems as we know them today, introduced the concept of Lamport Clocks in his 1978 paper "Time, clocks, and the ordering of events in a distributed system" (for which he later won the Turing Award).
According to Lamport, that paper was directly inspired by the special theory of relativity. In his own words [^1]:
"[...] special relativity, in particular, the four-dimensional space-time view developed by Minkowski in the famous 1908 paper. I realized that the problems in distributed systems are very much analogous to what's going on in physics [...]"
So, I don't know if these algorithms were reinvented or (re)discovered because, as you rightly say, they happen to operate in our universe.
Vector clocks are valuable precisely because you can tell when concurrent events occur. Eg [1,2] and [2,1] are concurrent but [1,2] and [2,2] are ordered. You need this info for things like MVRegister CRDTs.