Difference between revisions of "OpenID in Smartphones"

Revision as of 21:32, 7 December 2020

Full Title or Meme

This use case focuses on the parts of enabling a Self-issued Identifier within a battery-powered mobile device that will impact the protocol that is used when the device acts as a Self-issued OpenID Provider.

Context

• This use case assumes the user with a Smartphone that wants to enable a Self-issued Identifier using only that phone
• Both Apple and Android have learned that energy consuming applications will drain a user's battery and work to limit apps that do that.
• Also no app that resulting in huge drains on the Smartphone battery would survive for long in the marketplace.
• The goal for would be a Self-issued OpenID Provider User Experience that is as good as the front channel OpenID Connect experience.
  • The user experience with creating the new identifier must result in 80 - 90% success rate.
  • The user Experience with navigating and registry with a new RP must be nearly as good as that using Facebook to perform the same function.
  • The user experience of loosing access due to accident of technology upgrade must be 90 - 95% successful on first attempt.

Problem

• Given the current browser reality, SIOP will not duplicate the same seamless, in-browser window experience,
• In other words, a user cannot get a signin experience on a RP site for first time signin with a DID that they can get with, say, Google or FB.
• Facebook and Microsoft have had years to create an account recovery regime that works of a wide range of access denials in the real world. SIOP needs to come close to duplicating that functionality or it will suffer from bad user reports on the technology.

Preconditions

• The user has a smartphone that can handle the secure storage of user secrets.
• A native app can be down-loaded that will act as a SIOP. Some folk call this a use wallet if it can also store other credentials, but that is outside the scope of this use case.

References