Have you ever wonder why typically people use i, j, k variables in “for loops”? Example from wikipedia article:

for (i = 0; i < 100; i++) {
    for (j = i; j < 10; j++) {
        some_function(i, j);
    }
}

It is believed that this tradition comes from Fortran, 1954. There were only two types of variables: fixed point (integer) and floating point (real). All variables starting with i - n were integers. It is not specified in manual, but I think that i - n were selected as mnemonic technique - it is two first letters in the word i-n-teger.

Here we are 70 years later still using this convention, even so many modern developer haven’t even heard about Fortran.