W3C Statement about the XRI 2.0 Vote

On 21 May 2008 the W3C TAG (Technical Architecture Group) posted the following statement regarding the OASIS Standard vote on XRI Syntax 2.0 and XRI Resolution 2.0:

In The Architecture of the World Wide Web [1] the TAG sets out the reasons why http: URIs are the foundation of the value proposition for the Web, and should be used for naming on the Web. The TAG has reviewed XRIs twice, once in a previous draft [2] and more recently the XRI Resolution draft of [3] and on both occasions we raised a number of questions with the OASIS XRI TC [4], [5].

The TAG's work in this area is ongoing. For further information see ISSUE-50: URNsAndRegistries [6].

We are not satisfied that XRIs provide functionality not readily available from http: URIs. Accordingly the TAG recommends against taking the XRI specifications forward, or supporting the use of XRIs as identifiers in other specifications.

Tim Berners-Lee and Stuart Williams, co-chairs, W3C Technical Architecture Group

Executive Summary of XRI Technical Committee Response

  1. The XRI TC has multiple times provided detailed replies to questions or comments from the W3C TAG but has never received any response back from the TAG. In fact the TAG's archives (cited at bullet #6 above) mention no further review of our responses.

  2. Although the TAG does not recognize XRI functionality beyond what is available from http: URIs, OASIS XRI TC members do. Most importantly, other standards and leading open source projects also recognize the merits of XRI and are supporting use of XRI in specifications and implementations including OpenID 2.0, Higgins 1.0, i-names, and XDI.

  3. The XRI TC was chartered in February 2003 to address requirements for abstract structured identifiers beyond the capabilities of http: URIs. These requirements were published in the XRI Use Cases and Requirements (June 2003) and are summarized on the Wikipedia article on XRI.

  4. Lastly, XRI TC members feel it would be a serious mistake for the W3C TAG to assert that "URIs are the foundation of the value proposition for the Web" while at the same time effectively saying URI schemes are off limits to innovation. The TAG is not recognizing five years of hard work by many companies, scores of individual contributors, and several open source projects to create a new form of URI that has already proved itself in deployed infrastructure, applications, and open source libraries.

XRI TC Efforts to Communicate with the W3C TAG

The XRI TC has great respect for the W3C TAG and their efforts to define an interoperable and robust architecture for the World Wide Web. The irony of the TAG's position is that the XRI TC has made it an explicit goal since its inception to innovate a new form of URI and service discovery that fully implements the TAG's definition of World Wide Web architecture.

This effort was first reflected in the extensive rearchitecture of the original contributions made to the TC that eventually resulted in the XRI Syntax 2.0 specification. The motivation for these changes was to make XRI fully compatible with URI and IRI architecture as specified in RFC 3986 and RFC 3987. Just as the IRI specification defined a lossless tranform of IRIs to URI syntax, the XRI Syntax 2.0 specification defines a lossless transform of XRIs to IRI syntax. In addition, the XRI TC has always planned to register the xri: scheme with IANA as soon as the specifications reached OASIS Standard status.

Even more effort was expended to engineer the XRI resolution framework so it takes full advantage of HTTP and HTTP(S) URI architecture. After finishing the initial 2.0 specification, two additional years of deployment experience and public feedback resulted in further revisions – many of them motivated by and innovated in cooperation with the OpenID community – to include full support for service discovery for both native HTTP(S) URIs as well as XRIs expressed as HTTP(S) URIs (called HXRIs).

When the W3C TAG posed several questions about this revised XRI resolution specification at the end of its public review cycle in March, the XRI TC took great pains to compose a highly detailed, five-page response explaining how it leveraged key features of WWW architecture. However we receive no further response, and the W3C TAG's archives show that no action was taken to review or reply our response (see Action 124, Action 125, and this email).

The Problems XRI and XRDS Are Solving on the Web Today

A greater irony of the W3C TAG's position is that the XRI 2.0 specifications were submitted for OASIS Standard consideration because deployers were telling the XRI TC that they deserved full OASIS Standard status due to the real problems they are solving in the market today.

We urge OASIS members to review the May 6 webinar given by the XRI TC entitled "What do OpenID, Higgins, i-names, and XDI Have in Common?" that explains why XRI and XRDS have already become key building blocks of the emerging Internet identity layer. Following are five key examples:

OpenID 2.0

Please see the paper "OpenID Discovery with XRI and XRDS", presented at the 2008 IDtrust Symposium, for a detailed explanation of why the XRI 2.0 specifications were incorporated into the OpenID 2.0 specifications. The top three reasons are:

  1. XRDS service discovery solved the key problems of how to describe the OpenID services associated with an OpenID identifier.
  2. XRI i-numbers (persistent synonyms) solve the OpenID recycling problem in a way that preserves the full control and independence of an OpenID user.
  3. Unlike URLs, XRI resolution provides two modes for trusted resolution of any XRI without any knowledge or extra effort by the user.

Higgins 1.0

Paul Trevithick, founder and project lead of the Higgins Project user-centric identity framework, describes it this way:

Further details of Higgins usage of XRI and XRDS are provided in the IDtrust Symposium paper cited above.

I-Names and I-Numbers

XDI.org currently offers XRI i-name and i-number registry services to over 10,000 individuals and businesses via a network of accredited i-brokers. See the XDI.org website, the Global Services Specifications, and the I-Services Specifications.


XDI is a structured data sharing protocol under very active development by the OASIS XDI Technical Committee. Based entirely on the structured identifier syntax and service endpoint discovery capabilities of XRIs, XDI provides an RDF model for universal sharing, linking, and synchronization of data. XDI proof-of-concepts have already been deployed by XDI TC members including ooTao's MyXDI server and the open source XDI4J (XDI For Java) project within Higgins.

Identity Capabilities Description

At the Internet Identity Workshop held May 12-15 2008 (where a third of the attendees identified themselves with i-names), several sessions were devoted to discussing new ways to apply XRI and XRDS to the problem of fully interoperable, cross-system description and discovery of identity capabilities. See this blog post by AOL Chief Identity Architect George Fletcher and this blog post by Axel Nennker, founder of the OpenInfoCard Project.


Although the XRI TC deeply respects the work of the W3C TAG, we believe the market is already demonstrating the utility of abstract structured identifiers and interoperable service discovery. We further believe they advance World Wide Web architecture in ways that are fully compatible with the W3C TAG's vision. We therefore urge OASIS members to support this innovation by voting in favor of the XRI 2.0 specifications:

XriSolvesRealProblems (last edited 2009-08-12 18:07:15 by localhost)