Fair enough, I guess.  But all of these libraries already have algorithm
IDs for SHA256/SHA512, so some new convention is going to have to come for
SHA3/512.

I can just see the administrators saying "Damn, I forgot the '2' again!"


On Tue, Dec 10, 2013 at 4:34 PM, Olafur Gudmundsson <[email protected]> wrote:

>
> On Dec 10, 2013, at 4:29 PM, Richard Barnes <[email protected]> wrote:
>
> > (Sorry if this has already been raised, but…)
> >
> > The digest identifiers in draft-ietf-dane-registry-acronyms-02 seem a
> little silly, in that nobody else in the world really seems to care that
> these are variants of SHA2.  The standard practice across many libraries is
> to just use some variant of "SHA-XXX", where XXX=256,384,512.
> >
>
> Richard,
>
> First version had this but an comment was made that we could have both
> SHA2  and SHA3 in 512 bit variant thus the recommendation
> was to future proof us.
>
> > OpenSSL: shaXXX
> > WebCrypto: SHA-XXX
> > BouncyCastle: SHAXXXDigest
> > CNG: BCRYPT_SHAXXX_ALGORITHM
> > PKCS#11: CKM_SHAXXX
> >
> > So I would suggest we just change these to "SHA-256" and "SHA-512".
>
> Unless the chair's tell me to make the change it will not be made,
> feel free to bring this up in the IETF LC if you think this is important.
>
>         Olafur
>
> >
> > --Richard
> >
> > _______________________________________________
> > dane mailing list
> > [email protected]
> > https://www.ietf.org/mailman/listinfo/dane
>
>
_______________________________________________
dane mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dane

Reply via email to