are operations on integers in scope for this library?

[3b]

It would be nice to have fast endian swap, signed<->unsigned, unsigned<->float, etc. operations for various sizes of integers for cases where they can't be done while reading from an octet vector (for example when parsing data that isn't byte-aligned).

Is that something that would be reasonable to add to nibbles, or would it be better as a separate library?

[brown]

Nibbles seems to be in maintenance mode at this point and all the current code works on byte-aligned data. I think a new library probably makes the most sense.