[webauthn] clarify normality of authenticator model - is it actually authenticator API ?

equalsJeffH has just created a new issue for 
https://github.com/w3c/webauthn:

== clarify normality of authenticator model - is it actually 
authenticator API ? ==
Presently, we characterize the webauthn authenticator (authnr) model 
as a "specific abstract functional model".

However, we are refining this authnr model (e.g. PR #154) such that it
 is becoming more concrete. Perhaps we need to update the language in 
the spec such that instead of an "authenticator model", it is an 
"authenticator API". 

Note that if one looks at the extant example of the FIDO U2F USB-HID, 
Bluetooth, and NFC "authenticator transport protocols"..

https://fidoalliance.org/specs/u2f-specs-1.0-bt-nfc-id-amendment/fido-u2f-hid-protocol.html
https://fidoalliance.org/specs/u2f-specs-1.0-bt-nfc-id-amendment/fido-u2f-bt-protocol.html
https://fidoalliance.org/specs/u2f-specs-1.0-bt-nfc-id-amendment/fido-u2f-nfc-protocol.html

..one can imagine that a similar protocol(s) and subsequent low-level 
"authenticator API(s)" could be defined for WebAuthn, i.e., as is 
tacitly imagined in the webauthn "Authentication" use case where one's
 smartphone is acting as the authenticator for a browser running on a 
laptop (a protocol between the laptop and the phone is needed to 
fulfill this use case). 



Please view or discuss this issue at 
https://github.com/w3c/webauthn/issues/176 using your GitHub account

Received on Wednesday, 17 August 2016 23:21:12 UTC