further details for authenticatorCancel operation

[equalsJeffH]

We may wish to add to the authenticatorCancel operation description some considerations for rolling back state (counters, keys, etc.)

[equalsJeffH]

see also #125

[equalsJeffH]

see also issue #292

[AngeloKai]

I am moving the issue back to WD-07 because I am addressing the cancel issues together at once.

[equalsJeffH]

@AngeloKai said:

I am moving the issue back to WD-07 because I am addressing the cancel issues together at once.

Ok, changing assignee to @AngeloKai

[equalsJeffH]

@jcjones notes in webauthn f2f tpac 2017-11-09: that we ought to have some hand-wavy general impl-cons for authnrs

i.e. for cancellation, go look at the transport specs... (ie ctap subsections)

[nadalin]

@selfissued Mike please add a sentence in the implementation section to say this is job of the transport specs (CTAP)

[jcjones]

(and U2F, which leaves it undefined but still isn't webauthn's problem...)

[equalsJeffH]

on 6-Dec-2017 webauthn call: move to this to PR milestone, we'll have more data wrt the "authenticator model" spec section's level of abstraction.

[AngeloKai]

The CTAP spec described the detailed workings of canceling authenticator operation and the current webauthn spec described the rough model of authenticator cancel works: https://w3c.github.io/webauthn/#authenticatorcancel. I don't see a need to further define authenticator cancel operation. Therefore, closing the issue for now.

@equalsJeffH can you please inform me what further detail you think we should provide? If so, we can re-open the issue.