<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <p>Thanks David.</p>
    <p>Unfortunately server side session isn't an option for the JS SDK
      use case, where the app might not have a backend (and even if it
      does, enlisting it to acquire and renew tokens to be used by the
      JS frontend would entail adding legs to the protocol).</p>
    <p>About your conversation with Apple: would you be able to keep the
      list updated on what you learn from them? I would be happy to join
      the conversation and articulate the SDK use case, if that helps.</p>
    <p>Use of iFrames for renewing tokens has never been trouble free
      (the zones in IE Brock mentioned in a different branch, disabled
      3rd party cookies etc) but this change would make the problem far
      more ubiquitous, to the point that standard workarounds (don;t
      disable 3rd party cookies; etc) will go from controversial to
      unfeasible.</p>
    <p>Thx</p>
    <p>V.<br>
    </p>
    <br>
    <div class="moz-cite-prefix">On 6/6/18 10:58 AM, David Waite wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:8A71E9E9-8309-42E3-9276-E3AC1F66665F@alkaline-solutions.com">
      <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
      Hi Vittorio,
      <div class=""><br class="">
      </div>
      <div class="">Yes, Apple seems to be further moving from a model
        where all state is isolated not just on the origin of the
        content, but segmented on both the top-level URL bar location
        and the remote origin, e.g. a (local location, remote location)
        pair.</div>
      <div class=""><br class="">
      </div>
      <div class="">They had this blog post about the change: <a
          href="https://webkit.org/blog/8311/intelligent-tracking-prevention-2-0/"
          class="" moz-do-not-send="true">https://webkit.org/blog/8311/intelligent-tracking-prevention-2-0/</a></div>
      <div class=""><br class="">
      </div>
      <div class="">The option to prompt the user for storage access
        that apple has provided should only prompt once per site
        (hopefully), but can only be triggered once the user has
        interacted with that site, e.g. clicked on the iframe. So
        prompting is likely not only a bad UX from prompting, but would
        require the user to interact with a component that isn’t
        providing obvious value.<br class="">
        <div><br class="">
        </div>
        <div>The RFC is for the session access API that they have
          implemented above, prompting the user and requiring user
          interaction to use.</div>
        <div><br class="">
        </div>
        <div>Hopefully it is not too self-serving to note that the DTVA
          proposal uses back-end API to coordinate session management,
          so it should not be affected by this change.</div>
        <div><br class="">
        </div>
        <div><br class="">
        </div>
        <div>As a second point, I reached out to the web evangelist at
          Apple for clarification on how federated login sites can avoid
          being classified as tracking under ITP. In particular, it
          seems a fully transparent SSO (without user interaction with
          the IDP site) may cause the IDP to be classified, at which
          point future redirects for SSO will get a (RP, IDP) segmented
          state, with the user appearing unauthenticated and the browser
          looking like a unique browser. </div>
        <div><br class="">
        </div>
        <div>There are a lot of technical, security, and user
          knowledge/empowerment reasons to always have an IDP
          interaction on SSO, but it is a behavior that a lot of
          deployments strive very hard to avoid.</div>
        <div><br class="">
        </div>
        <div>-DW</div>
        <div><br class="">
          <blockquote type="cite" class="">
            <div class="">On Jun 6, 2018, at 9:53 AM, Vittorio Bertocci
              via Openid-specs-ab <<a
                href="mailto:openid-specs-ab@lists.openid.net" class=""
                moz-do-not-send="true">openid-specs-ab@lists.openid.net</a>>
              wrote:</div>
            <br class="Apple-interchange-newline">
            <div class="">
              <meta http-equiv="content-type" content="text/html;
                charset=utf-8" class="">
              <div text="#000000" bgcolor="#FFFFFF" class="">
                <p
style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"
                  class="">Hi all,</p>
                <p
style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"
                  class="">We have been having issues with renewing
                  tokens via invisible iFrame in our Javascript SDKs in
                  the latest version of Safari - and yesterday's news
                  about ITP 2.0 seem to suggest that the new default on
                  Apple devices will be equivalent to disabling 3rd
                  party cookies, which AFAIK breaks OIDC session
                  management... and/or start displaying dialogs warning
                  the user that they are being tracked at every
                  operation.<br class="">
                </p>
                <ul
style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"
                  class="">
                  <li style="margin-left:15px" class="">Did anyone else
                    experience similar issues?</li>
                  <li style="margin-left:15px" class="">What are the
                    WG's thoughts about whether this calls for a
                    revision of how session works in OIDC?</li>
                  <li style="margin-left:15px" class="">There is one RFC
                    for WebKit that could provide an alternative
                    location for the session, detailed <a
                      href="https://github.com/whatwg/html/issues/3338"
                      class="" moz-do-not-send="true">here</a>. Did
                    anyone consider it? Any insights?</li>
                </ul>
                <p
style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"
                  class="">If the issue is confirmed, that will make use
                  of OIDC session and related token renewal machinery
                  unfeasible on Macs, iPhones and iPads. And without
                  official guidance, that will likely spur a cottage
                  industry of custom solutions. I hope we can come up
                  with guidance that addresses the problem before that
                  happens.</p>
                <p
style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"
                  class="">Thanks in advance for your insights</p>
                V. </div>
              _______________________________________________<br
                class="">
              Openid-specs-ab mailing list<br class="">
              <a href="mailto:Openid-specs-ab@lists.openid.net" class=""
                moz-do-not-send="true">Openid-specs-ab@lists.openid.net</a><br
                class="">
              <a class="moz-txt-link-freetext" href="http://lists.openid.net/mailman/listinfo/openid-specs-ab">http://lists.openid.net/mailman/listinfo/openid-specs-ab</a><br
                class="">
            </div>
          </blockquote>
        </div>
        <br class="">
      </div>
    </blockquote>
    <br>
  </body>
</html>