I mistakenly inferred from the following clause that a parent extended
public key plus a child private key would be equivalent to knowing the
extended *child* private key---meaning that the *parent* private key was
still secure:
> knowledge of the extended public key + any non-hardened private key
> descending from it is equivalent to knowing the extended private key
This patch's addition of the word "parent" (twice) removes the ambiguity
and may help other readers draw the correct inference that the parent
private key is no longer secure in this case.
I also changed "+" to "plus" to avoid confusion with the actual
mathematical operations used in this BIP.
Slightly re-worded description of the signature field in PaymentRequest
The previous phrasing confused me and I had to check the Bitcoin Core source code to see what it was expecting for a signature.
David A. Harding