While I totally understand your point: namely that the prefix for the type is not the same thing as the variable name - and your point about Satoshi's own documentation is correct, still, saying that "scriptSig" and "scriptPubkey" are wrong names coming from misunderstanding, I would claim itself is a misunderstanding :) The best way to think of these two things are as *generalizations* of the concepts of signature, public key in a public key digital signature scheme. For the obvious reason that Bitcoin Script itself is designed to allow that generalization: you're not limited to validating a signature against a single public key, but are instead evaluating the predicate script-verify(script-pubkey, script-signature, message) instead of just verify(pubkey, signature, message). From what I know that's why scriptPubkey and scriptSig stuck.