- From: Kayode Ezike <kezike13@gmail.com>
- Date: Thu, 29 Sep 2022 15:16:27 -0400
- To: Kristina Yasuda <Kristina.Yasuda@microsoft.com>
- Cc: David Chadwick <david.chadwick@crosswordcybersecurity.com>, "public-vc-edu@w3.org" <public-vc-edu@w3.org>
- Message-ID: <CAPXfD67PKkx9QkFR91sm-c29HHp4TaMtvn5Mmk9MHhoCOY1-YA@mail.gmail.com>
Hi David and Kristina, Sorry, I had some issues receiving messages from this mailing list for a couple days, but I see them now. Thanks for the detailed guidance. Cheers, Kayode On Thu, Sep 29, 2022 at 2:20 AM Kristina Yasuda < Kristina.Yasuda@microsoft.com> wrote: > Hi Kayode, > > > > David has already answered but just to add a little color. > > > > > How are holders redirected to their wallet from a QR code? In the spec, > I see that clients can define a new Issuance Initiation Endpoint > (initiate_issuance_endpoint) in its metadata, but it is unclear how and > where this is supposed to be defined for a native wallet. Is this supposed > to be communicated via the wallet_issuer parameter of an authorization > request as defined in section 7.1.3 of the spec? If so, how is this > endpoint exposed in a pre-authorized code flow? > > -> yes, Issuers and the wallets are expected to pre-agree on the Issuance > Initiation Endpoint value using out of band mechanisms, including interop > profiles. For the cases when such out of band mechanisms are not available, > VCI specification defines a static value `openid-initiate-issuance://` that > David’s team is using. However, note that there is no guarantee that all > wallets supporting VCI also support `openid-initiate-issuance://` and > because it is a custom URL scheme, it faces all the usual limitations on > mobile OS (which is a broader problem than only for VCI). > > > > > For the pre-authorized code flow, wallets are not expected to hit the > authorization endpoint. Since this is the endpoint that shares > wallet_issuer with the issuer, how else should the issuer discover this > value in this flow? Is that exchange supposed to occur in an unspecified > offline process? > > -> note that wallet_issuer would work only for the wallets with cloud > components that can do dynamic discovery. per > https://www.rfc-editor.org/rfc/rfc6749.html#section-3.2.1 wallet can pass > `client_id` in the token request, which might be useful if the > authorization server already knows metadata tied to certain client_id. > > > > > Am I correct in my interpretation of the issuer parameter of > the Issuance Initiation request as the OP issuer URL, at which clients can > discover the server metadata (located specifically > at /.well-known/openid-configuration, per the original OIDC Core spec), > including the new Credential Endpoint (credential_endpoint)? > > -> as David said, yes with on-going discussion whether credential endpoint > metadata shall be published in a separate file, so would recommend to keep > an eye on the outcome – this is probably one the last discussion points > before going to the First Implementer’s Draft to give implementers IPR > protection. Implementer’s draft is not a final draft and there are still > some open issues, but for some of those we need more implementation > experience before deciding on a direction. > > > > Cheers, > > Kristina > > > > *From:* Kayode Ezike <kezike13@gmail.com> > *Sent:* Monday, September 26, 2022 7:55 AM > *To:* David Chadwick <david.chadwick@crosswordcybersecurity.com> > *Cc:* public-vc-edu@w3.org; Kristina Yasuda <Kristina.Yasuda@microsoft.com>; > Torsten Lodderstedt <torsten@lodderstedt.net> > *Subject:* Re: OIDC4VCI JFF PlugFest 2 Questions > > > > You don't often get email from kezike13@gmail.com. Learn why this is > important <https://aka.ms/LearnAboutSenderIdentification> > > Thanks for the detailed reply David! Follow-up questions/comments: > > - For the pre-authorized code flow, wallets are not expected to hit > the authorization endpoint. Since this is the endpoint that shares > *wallet_issuer* with the issuer, how else should the issuer discover > this value in this flow? Is that exchange supposed to occur in an > unspecified offline process? > - Could you please share access to the two documents you shared in > your response? > - Thanks for sharing the credentials for the Identiproof portal, as I > have been searching everywhere for that haha! > > Cheers, > > Kayode > > > > On Mon, Sep 26, 2022 at 6:09 AM David Chadwick < > david.chadwick@crosswordcybersecurity.com> wrote: > > p.s. If people want to test the initiate issuance request they can do so > by going to this web site that we have created for this purpose (which we > demonstrated at the end of the last meeting) > > https://idp.research.identiproof.io/ > <https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fidp.research.identiproof.io%2F&data=05%7C01%7Ckristina.yasuda%40microsoft.com%7C79c55b8c784f42bdc01608da9fcf27e0%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637998010439215000%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000%7C%7C%7C&sdata=MgJ7UQRPsFF%2FVD%2FWXd%2FAd%2B3PEshBDVh4r6VRzPJ%2BG1o%3D&reserved=0> > > Use the username: "user" and the pw: "password" to login. This web site > can be used by all three types of wallet: cloud wallets, smart wallets same > device flow and smart wallets cross device flow. > > Kind regards > > David > > On 26/09/2022 10:58, David Chadwick wrote: > > > > On 26/09/2022 10:10, Kayode Ezike wrote: > > Hi there, > > > > Calling on issuers in the JFF PlugFest 2 cohort and practitioners > generally that implement the OIDC4VCI spec as a means of delivering > credentials with the following questions: > > Hi Kayode > > > > > - How are holders redirected to their wallet from a QR code? In the > spec, I see that clients can define a new *Issuance Initiation > Endpoint (initiate_issuance_endpoint)* in its metadata, but it is > unclear how and where this is supposed to be defined for a native wallet. > Is this supposed to be communicated via the *wallet_issuer* parameter > of an authorization request as defined in section *7.1.3* of the spec? > If so, how is this endpoint exposed in a pre-authorized code flow? > > If you read the profile that Spruce and Crossword have written for our > interworking tests, all should be revealed. You can find it here > > https://drive.google.com/drive/folders/1qHI12GeBhl2s1_bSopDIGpkPdjCE5pMO > <https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdrive.google.com%2Fdrive%2Ffolders%2F1qHI12GeBhl2s1_bSopDIGpkPdjCE5pMO&data=05%7C01%7Ckristina.yasuda%40microsoft.com%7C79c55b8c784f42bdc01608da9fcf27e0%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637998010439215000%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000%7C%7C%7C&sdata=BrqtDmgNtNIqq6FI7D3xzsPyxhvE9ufA6ZAClWY%2B2Mk%3D&reserved=0> > > To answer your question specifically the following deeplink is used > > openid-initiate-issuance://?issuer=https%3A%2F% > 2Foidc4vci.demo.spruceid.com > <https://nam06.safelinks.protection.outlook.com/?url=http%3A%2F%2F2foidc4vci.demo.spruceid.com%2F&data=05%7C01%7Ckristina.yasuda%40microsoft.com%7C79c55b8c784f42bdc01608da9fcf27e0%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637998010439215000%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000%7C%7C%7C&sdata=c1MCv2ASvllNj%2FKt7RDu9hmTEDFEVXcEOyEdKL9Knxc%3D&reserved=0> > ....... > > where the issuer parameter points to the issuer and by appending the well > known suffix you can find the issuer's metadata. As to what that suffix > should be, this is still under debate in the OIDC working group. Spruce and > Crossword have published theirs here, and invite other JFF participants to > send their details to us and we will add theirs to this table > > > https://docs.google.com/document/d/e/2PACX-1vQtNazS0EtOO05cR8VM4lCHWtr8KTZRESbiaF3DzpR_JaL3Y540vPjx9d2U4gj7fxMSfi3JRnFQOJKg/pub > <https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdocs.google.com%2Fdocument%2Fd%2Fe%2F2PACX-1vQtNazS0EtOO05cR8VM4lCHWtr8KTZRESbiaF3DzpR_JaL3Y540vPjx9d2U4gj7fxMSfi3JRnFQOJKg%2Fpub&data=05%7C01%7Ckristina.yasuda%40microsoft.com%7C79c55b8c784f42bdc01608da9fcf27e0%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637998010439215000%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000%7C%7C%7C&sdata=ne%2Bt5LdtaBQ79xyr%2FhfsxPAEySgKVj1reEZmBcYX6mI%3D&reserved=0> > > > > > - Am I correct in my interpretation of the *issuer* parameter of the *Issuance > Initiation* request as the OP issuer URL, at which clients can > discover the server metadata (located specifically at > */.well-known/openid-configuration*, per the original OIDC Core spec), > including the new *Credential Endpoint (credential_endpoint)*? > > Yes you are correct, but see above > > Kind regards > > David > > > > Cheers, > > > > Kayode > > > > > -- > > Kayode Ezike | https://ezike.io > <https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fezike.io%2F&data=05%7C01%7Ckristina.yasuda%40microsoft.com%7C79c55b8c784f42bdc01608da9fcf27e0%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637998010439215000%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000%7C%7C%7C&sdata=0lKihvHyS5bMUNb6jXeIxh52fHGeFyWQaOl%2Fp6SqtHI%3D&reserved=0> > > MIT | BS 2017 | MEng 2019 > > Engineer | Writer | Creator > -- Kayode Ezike Principal at I/O by Ayo
Received on Thursday, 29 September 2022 19:16:51 UTC