Oops, your’re right, Numpy uses tuple-of-slices, not slice-of-tuples. I can’t think of a usage of tuple-of-slices from the top of my head either, but slices are definitely not limited to integers (and None). It is common in Pandas, for example, to use slice of strings (where the strings are titles), e.g.
df["C1":"C3"] # This returns data between labels C1 and C3.
Are you talking here about type annotations? That might be something to discuss on the typing SIG. I don’t know what the current annotations for the slice builtin are (it doesn’t have an __annotations__ attribute).
SIG is short for Special Interest Group, and the Typing SIG is a bunch of people that are interested in stuff related to typing, including type annotations. They discuss things on the Typing SIG mailing list: https://mail.python.org/mailman3/lists/typing-sig.python.org/