add implementation considerations section to address gap between serialization format and APIs in FIPS 203 #79
Conversation
…alization format and APIs in FIPS 203
Co-authored-by: Mike Ounsworth <[email protected]>
Co-authored-by: Mike Ounsworth <[email protected]>
| Similarly, the algorithm that expands seed values is not listed as one of the "main algorithms" and features "internal" in the name even though it is clear that it is allowed to be exposed externally for the purposes of expanding a key from a seed. | ||
| Below are possible ways to extend the APIs defined in {{FIPS203}} to support serialization of seed values as private keys. | ||
|
|
||
| To support serialization of seed values as private keys, Algorithm 19 in {{FIPS203}} should return (ek, dk, d, z) on line 7 and Algorithm 16 should be promoted to be a "main algorithm" for external use in expanding seed values. |
There was a problem hiding this comment.
Algorithm 19 in {{FIPS203}} should return
This sounds like we're redefining FIPS203 instead of extending it. Better to phrase it like let "alg 19b" denote the same procedure as algorithm 19 except that it returns (ek, dk, d, z) on line 7.
There was a problem hiding this comment.
@carl-wallace are you going to suggest some changes I can merge in?
There was a problem hiding this comment.
Applying Bas' suggestion to the PR I sent yields something like this:
To support serialization of seed values as private keys, let Algorithm 19b denote the same procedure as Algorithm 19 in {{FIPS203}} except it returns (ek, dk, d, z) on line 7. Additionally, Algorithm 16 should be promoted to be a "main algorithm" for external use in expanding seed values.
No description provided.