Re: Algorithms and referenced documents

On Tue, Jun 25, 2013 at 7:16 AM, Alessandro Di Federico
<alessandro.difederico@mail.polimi.it> wrote:
> Hi, I'm a Ms.C student from Politecnico di Milano, and I'm using the
> WebCryptoAPI (PolyCrypto, right now) for my thesis.
>
> I've some suggestions/questions about the WebCryptoAPI:
>
>      1. Why don't you include the newly standardized SHA-3 hashing
>         algorithm in the document?

While Keccak has been chosen for SHA-3, NIST has not yet released
recommended parameters that would define what SHA-3 actually is.

>      2. It'd be nice to have an alternative to RSA for asymmetric
>         encryption, why don't you include ElGamal and EC-ElGamal?

Per the charter and original discussions, this was based on exposing
cryptographic functionality that browsers/user agents already had
implemented (and applied for various controls for, such as export
licenses) and exposing those to content scripts.

With this criteria, ElGamal/EC-ElGamal are not widely implemented.

>      3. For our project we need an authenticated Diffie-Hellman key
>         exchange, we plan to use HMQV [1]. We can implement it over the
>         standard DH primitive, but having it native would be the best.
>         Can you consider to include it?

For the same criteria given above, I think it's unlikely.

Likewise, begin the debate on HMQV vs FHMQV.

>
> Thanks in advance,
> Alessandro Di Federico
>
> [1] http://link.springer.com/chapter/10.1007%2F11535218_33
> [2] http://www.w3.org/TR/WebCryptoAPI/#dh-EcdhKeyDeriveParams
>
>
>
>
>

Received on Wednesday, 26 June 2013 17:36:51 UTC