When did the term "tuple" get its current meaning?

Question

In a recent discussion, someone told me tuples in the modern meaning (in particular, tuples are heterogeneous: that is, different elements of a tuple can belong to different sets/have different types) first appeared in Codd's tuple calculus. I was surprised it would be so late, but searching Google Books before 1970, I can't see any clearly heterogeneous examples, and quite a few clearly homogeneous ones ("tuple of ones and zeros", "tuple of natural numbers", etc.)

Сan anybody confirm that Codd introduced heterogeneous tuples or point out an earlier appearance?

Answer 1

Tuples appear in essentially all formal treatments of set theory, and those go back to way before the 70s!

Answer 2

I'm not sure having really understood your question, but it seems to me that mathematicians make no difference between heterogeneous and homogeneous tuples. Consider $\left(5,b,f,8,3\right)$: one can say it is an heterogeneous tuple because the set $\left\{5,b,f,8,3\right\}$ can be partitioned into $\left\{b,f\right\}$ and $\left\{5,8,3\right\}$, letters and digits respectively. But are letters and digits of different type? What do you mean by type? In mathematics, as far as I know, there is only one type: the set type. Everything is a set. In this example letters and digits should be both defined as (particular) sets. So every tuple is homogeneous by default. Also, consider $\left(2,1,9,7\right)$: there are only digits this time, so one can say it is an homogeneous tuple. But what if I split the set of all digits into those less then $5$ (the "low" digit type) and those equal to or greater than $5$ (the "high" digit type)? It becomes heterogeneous. I would say the context is of great importance here.