Cool paper - thanks for posting! I love the typesetting of the math in these old papers - I have no idea how it was done, but it's ~good vibes~ for me.
If you're interested in reading a more recent take on notation, check out this post [1] by Terence Tao on MathOverflow. It's pretty cool, and a shorter read than this PDF - if you're in a rush.
While I'm posting overflow questions I like, check out this gem [2] on parsing HTML with a regex.
> I love the typesetting of the math in these old papers
I agree except that the first thing the paper does is define "1" to be a function that "produces a vector of the first N integers". Highly confusing. It turns out that symbol isn't a "1" and rather is a weird squiggle that unfortunately looks like one.
It should be the Greek letter iota (i.e. I from indices).
Besides iota, APL uses frequently the Greek letter rho (i.e. R from rank; rho applied once gives a vector with the array dimensions and applied twice it gives the dimension of the rho vector, i.e. the tensor rank of the array).
If you're interested in reading a more recent take on notation, check out this post [1] by Terence Tao on MathOverflow. It's pretty cool, and a shorter read than this PDF - if you're in a rush.
While I'm posting overflow questions I like, check out this gem [2] on parsing HTML with a regex.
[1] https://mathoverflow.net/questions/366070/what-are-the-benef...
[2] https://stackoverflow.com/questions/1732348/regex-match-open...