Re: [kitten] Request for review of a SASL mechanism using the OPAQUE aPAKE

[Alexey Melnikov <[e-mail]>]

Hi Nadja,

Thank you for your draft. I read it and overall it looks good, other 
than missing specific parameters in Section 7.

Also some specific comments:

> 6.1.  Default Channel Binding
>
>     'tls-exporter' is the default channel binding type for any
>     application that do not specify one.
>
>     Servers MUST implement the 'tls-exporter' [RFC9266] channel binding
>     type if they implement any channel binding and use TLS.  Clients
>     SHOULD implement the 'tls-exporter' [RFC9266] channel binding type if
>     they implement any and use TLS.
>
>     Servers MUST use the channel binding type indicated by the client, or
>     fail authentication if they do not support it.
In order to maximize interoperability I suggest you also mention 
tls-unique for TLS 1.2, unless your exlusion of TLS 1.2 is deliberate? 
In the latter case your draft should say that.


In Section 8:

>    client-first-message-bare =
>                [reserved-mext ","] username "," auth-request
>                ["," extensions]
"auth-request" is not defined.
>
>    client-first-message = gs2-header client-first-message-bare
>
>    validator = "v=" base64
>
>    server-message-bare =
>                [reserved-mext ","] channel-binding "," ksf-params ","
>                credentials-response ["," extensions]
"ksf-params" and "credentials-response" are not defined.

Best Regards,

Alexey