Draft -04 of the JWS Unencoded Payload Option specification addresses the shepherd comments. Thanks to Jim Schaad for his careful review. The primary change was adding additional security considerations text, including describing when “crit” should be used.
The specification is available at:
An HTML formatted version is also available at:
Proof-of-Possession Key Semantics for JWTs draft -06 addresses the remaining document shepherd comment – adding use case diagrams to the introduction.
The updated specification is available at:
An HTML formatted version is also available at:
Proof-of-Possession Key Semantics for JWTs draft -05 addresses Kepeng Li’s document shepherd comments (other than adding some use case diagrams in the introduction, which will happen soon).
The updated specification is available at:
An HTML formatted version is also available at:
Draft -03 of the JWS Unencoded Payload Option specification addresses the working group last call comments received. Thanks to Jim Schaad, Vladimir Dzhuvinov, John Bradley, and Nat Sakimura for the useful comments. Changes were:
The specification is available at:
An HTML formatted version is also available at:
Draft -02 of the JWS Unencoded Payload Option specification makes these updates:
b64” be integrity protected.b64” Header Parameter value MUST be the same for all of them."b64":false.Thanks for the working group feedback that resulted in these improvements.
The specification is available at:
An HTML formatted version is also available at:
The JSON Web Key (JWK) Thumbprint specification is now RFC 7638 — an IETF standard. The abstract describes the specification as follows:
This specification defines a method for computing a hash value over a JSON Web Key (JWK). It defines which fields in a JWK are used in the hash computation, the method of creating a canonical form for those fields, and how to convert the resulting Unicode string into a byte sequence to be hashed. The resulting hash value can be used for identifying or selecting the key represented by the JWK that is the subject of the thumbprint.
Thanks to James Manger, John Bradley, and Nat Sakimura, all of whom participated in security discussions that led to the creation of this specification. Thanks also to the JOSE working group members, chairs, area directors, and other IETF members who contributed to the specification.
A JWK Thumbprint is used as the “sub” (subject) claim value in OpenID Connect self-issued ID Tokens.
Proof-of-Possession Key Semantics for JWTs draft -04 addresses the remaining working group comments received — both a few leftover WGLC comments and comments received during IETF 93 in Prague. The changes were:
jwk” for symmetric keys when the JWT is encrypted.jku” (JWK Set URL) member.cnf” (confirmation) claim is defined before it is used.cnf“, to represent additional proof-of-possession keys, using the same representation as “cnf“.The updated specification is available at:
An HTML formatted version is also available at:
The former JWS Signing Input Options specification has been renamed to JWS Unencoded Payload Option to reflect that there is now only one JWS Signing Input option defined in the spec — the “b64”:false option. The “sph” option was removed by popular demand. I also added a section on unencoded payload content restrictions and an example using the JWS JSON Serialization.
The specification is available at:
An HTML formatted version is also available at:
The initial working group version of JWS Signing Input Options has been posted. It contains no normative changes from draft-jones-jose-jws-signing-input-options-00.
Let the working group discussions begin! I particularly call your attention to Martin Thomson’s review at http://www.ietf.org/mail-archive/web/jose/current/msg05158.html, Nat Sakimura’s review at http://www.ietf.org/mail-archive/web/jose/current/msg05189.html, and Matias Woloski’s review at http://www.ietf.org/mail-archive/web/jose/current/msg05191.html to start things off.
The specification is available at:
An HTML formatted version is also available at:
The IESG has approved JWK Thumbprint draft -08, meaning that it will now progress to the RFC Editor. Draft -08 added IANA instructions in response to an IESG comment by Barry Leiba.
The specification is available at:
An HTML formatted version is also available at:
JWK Thumbprint draft -07 has been published, addressing a Gen-ART review comment by Joel Halpern. Beyond updating the acknowledgements, the only change was replacing this sentence:
“Only if multiple parties will be reproducing the JWK Thumbprint calculation for some reason, will parties other than the original producer of the JWK Thumbprint need to know which hash function was used.”
with these two:
“However, in some cases, multiple parties will be reproducing the JWK Thumbprint calculation and comparing the results. In these cases, the parties will need to know which hash function was used and use the same one.”
The specification is available at:
An HTML formatted version is also available at:
The editors have published draft-ietf-oauth-proof-of-possession-03, which addresses the working group last call comments received. Thanks to all of you who provided feedback. The changes were:
jwk and jwe confirmation members; the former represents a public key as a JWK and the latter represents a symmetric key as a JWE encrypted JWK.iss, sub, and azp claims.The updated specification is available at:
An HTML formatted version is also available at:
A new JWK Thumbprint draft has been posted addressing the IETF Security Directorate (SecDir) comments from Adam Montville. The changes clarify aspects of the selection and dissemination of the hash algorithm choice and update the instructions to the Designated Experts when registering JWK members and values.
The specification is available at:
An HTML formatted version is also available at:
There’s been interest being able to not base64url-encode the JWS Payload under some circumstances by a number of people. I’ve occasionally thought about ways to accomplish this, and prompted again by discussions with Phillip Hallam-Baker, Martin Thomson, Jim Schaad, and others at IETF 92 in Dallas, recollections of conversations with Matt Miller and Richard Barnes on the topic, and with Anders Rundgren on the JOSE mailing list, I decided to write down a concrete proposal while there’s still a JOSE working group to possibly consider taking it forward. The abstract of the spec is:
JSON Web Signature (JWS) represents the payload of a JWS as a base64url encoded value and uses this value in the JWS Signature computation. While this enables arbitrary payloads to be integrity protected, some have described use cases in which the base64url encoding is unnecessary and/or an impediment to adoption, especially when the payload is large and/or detached. This specification defines a means of accommodating these use cases by defining an option to change the JWS Signing Input computation to not base64url-encode the payload.
Also, JWS includes a representation of the JWS Protected Header and a period (‘.’) character in the JWS Signature computation. While this cryptographically binds the protected Header Parameters to the integrity protected payload, some of have described use cases in which this binding is unnecessary and/or an impediment to adoption, especially when the payload is large and/or detached. This specification defines a means of accommodating these use cases by defining an option to change the JWS Signing Input computation to not include a representation of the JWS Protected Header and a period (‘.’) character in the JWS Signing Input.
These options are intended to broaden the set of use cases for which the use of JWS is a good fit.
The specification is available at:
An HTML formatted version is also available at:
The -01 version of draft-jones-jose-key-managed-json-web-signature tightened the semantics by prohibiting use of “dir” as the “alg” header parameter value so a second equivalent representation for content integrity-protected with a MAC with no key management isn’t introduced. (A normal JWS will do just fine in this case.) Thanks to Jim Schaad for pointing this out. This version also adds acknowledgements and references the now-final JOSE RFCs.
This specification is available at:
An HTML formatted version is also available at:
This JWK Thumbprint draft addresses issues raised in Kathleen Moriarty’s AD review of the -04 draft. This resulted in several useful clarifications. This version also references the now-final JOSE RFCs.
The specification is available at:
An HTML formatted version is also available at:
The OAuth Assertions specifications are now standards — IETF RFCs. They are:
This completes the nearly 5 year journey to create standards for using security tokens as OAuth 2.0 authorization grants and for OAuth 2.0 client authentication. Like the JWT and JOSE specs that are now also RFCs, these specifications have been in widespread use for a number of years, enabling claims-based use of OAuth 2.0. My personal thanks to Brian Campbell and Chuck Mortimore for getting the ball rolling on this and seeing it through to completion, to Yaron Goland for helping us generalize what started as a SAML-only authorization-grant-only spec to a framework also supporting client authentication and JWTs, and to the OAuth working group members, chairs, area directors, and IETF members who contributed to these useful specifications.
The JSON Web Token (JWT) and JSON Object Signing and Encryption (JOSE) specifications are now standards — IETF RFCs. They are:
This completes a 4.5 year journey to create a simple JSON-based security token format and underlying JSON-based cryptographic standards. The goal was always to “keep simple things simple” — making it easy to build and deploy implementations solving commonly-occurring problems using whatever modern development tools implementers chose. We took an engineering approach — including features we believed would be commonly used and intentionally leaving out more esoteric features, to keep the implementation footprint small. I’m happy to report that the working groups and the resulting standards stayed true to this vision, with the already widespread adoption and an industry award being testaments to this accomplishment.
The origin of these specifications was the realization in the fall of 2010 that a number of us had created similar JSON-based security token formats. Seemed like it was time for a standard! I did a survey of the choices made by the different specs and made a convergence proposal based on the survey. The result was draft-jones-json-web-token-00. Meanwhile, Eric Rescorla and Joe Hildebrand had independently created another JSON-based signature and encryption proposal. We joined forces at IETF 81, incorporating parts of both specs, with the result being the -00 versions of the JOSE working group specs.
Lots of people deserve thanks for their contributions. Nat Sakimura, John Bradley, Yaron Goland, Dirk Balfanz, John Panzer, Paul Tarjan, Luke Shepard, Eric Rescorla, and Joe Hildebrand created the precursors to these RFCs. (Many of them also stayed involved throughout the process.) Richard Barnes, Matt Miller, James Manger, and Jim Schaad all provided detailed input throughout the process that greatly improved the result. Brian Campbell, Axel Nennker, Emmanuel Raviart, Edmund Jay, and Vladimir Dzhuvinov all created early implementations and fed their experiences back into the spec designs. Sean Turner, Stephen Farrell, and Kathleen Moriarty all did detailed reviews that added ideas and improved the specs. Matt Miller also created the accompanying JOSE Cookbook — RFC 7520. Chuck Mortimore, Brian Campbell, and I created the related OAuth assertions specs, which are now also RFCs. Karen O’Donoghue stepped in at key points to keep us moving forward. Of course, many other JOSE and OAuth working group and IETF members also made important contributions. Finally, I want to thank Tony Nadalin and others at Microsoft for believing in the vision for these specs and consistently supporting my work on them.
I’ll close by remarking that I’ve been told that the sign of a successful technology is that it ends up being used in ways that the inventors never imagined. That’s certainly already true here. I can’t wait to see all the ways that people will continue to use JWTs and JOSE to build useful, secure applications!
An updated OAuth Proof-of-Possession draft has been posted that address the open issues identified in the previous draft. Changes were:
kid” (key ID) confirmation method.Thanks to Hannes Tschofenig for writing text to address the open issues.
This specification is available at:
An HTML formatted version is also available at:
The latest JWK Thumbprint draft addresses review comments on the -03 draft by Jim Schaad, which resulted in several clarifications and some corrections to the case of RFC 2119 keywords.
The specification is available at:
An HTML formatted version is also available at:
Powered by WordPress & Theme by Anders Norén