26315 2014-07-12 01:25:19 +0000 ECDSA/ECDH: "namedCurve ASN.1 type" is ambiguous 2014-09-26 23:26:52 +0000 1 1 1 Unclassified Web Cryptography Web Cryptography API Document unspecified PC Windows NT RESOLVED FIXED P2 normal --- 1 sleevi watsonm bal public-webcrypto watsonm oldest_to_newest 108876 0 sleevi 2014-07-12 01:25:19 +0000 Raised by Brian on The current ( https://dvcs.w3.org/hg/webcrypto-api/raw-file/ee10c81e1141/spec/Overview.html ) spec language for ECDSA/ECDH handling of EC keys states "If params is not an instance of the namedCurve ASN.1 type defined in RFC 5480" This was seen as confusing. The intent was to specify that of the choice of ECParameters, only the namedCurve choice was acceptable (e.g. implicitCurve and specifiedCurve) are NOT supported. As Brian interpreted, and as others may reasonably do so, this was seen as constraining the contents of the OIDs to those specified in 2.1.1.1 of RFC 5480 ( http://tools.ietf.org/html/rfc5480#section-2.1.1.1 ). While 2.1.1.1 is clear to indicate other specifications may describe additional types, it's still ambiguous because WebCrypto may choose to allow OIDs for which no specifications updating 5480 exist - e.g. the NUMS curves One possible language modification is "If params is an instance of the ECParameters ASN.1 type that specifies a namedCurve" but that may still be seen as ambiguous. 111958 1 watsonm 2014-09-22 17:58:59 +0000 I suggest we adopt the wording proposed in Comment #1: "If params is an instance of the ECParameters ASN.1 type that specifies a namedCurve" 112327 2 watsonm 2014-09-26 23:26:52 +0000 https://dvcs.w3.org/hg/webcrypto-api/rev/dbdc7abe4e32