Re: Minutes from yesterdays special topic call

To throw another requirement into the bucket, consider:

  *
What works well for DID *Document* Resolution (suggested renaming) should also work for Verifiable Credential *Document* Resolution. At least from a canonical syntax/semantic patterns perspective.

Michael Herman
Chief Digital Officer
Web 7.0 Foundation

Get Outlook for Android<https://aka.ms/AAb9ysg>
________________________________
From: Otto Mora <otto@mora.cr>
Sent: Thursday, April 9, 2026 8:58:08 AM
To: Decentralized Identifier Working Group <public-did-wg@w3.org>
Subject: Re: Minutes from yesterdays special topic call

Please use this zoom link!!


https://us06web.zoom.us/j/5637387869?pwd=R2lzUXpLUC91VitFajRKRVlpTlphUT09



Otto I. Mora
otto@mora.cr<mailto:otto@mora.cr>


On Thu, Apr 9, 2026 at 10:44 AM Otto Mora <otto@mora.cr<mailto:otto@mora.cr>> wrote:
Hello All,

Please see here for the minutes from yesterday's special topic call:

https://www.w3.org/2026/04/08-did-minutes.html


AI Summary:
The Decentralized Identifier (DID) Working Group met on April 8, 2026, to primarily address outstanding disagreements regarding Pull Request #316 in the did-resolution repository. Chaired by Otto Mora, the meeting brought together key contributors including Stephen Curran, Joe Andrieu, and Will Abramson to reconcile two competing approaches to DID URL dereferencing. The discussion focused on finding a consensus for a generic path-handling algorithm while balancing the need for method-specific logic, with members acknowledging that the small group size made it difficult to establish a formal quorum for definitive decisions.

A major portion of the technical debate centered on the "precedence" of DID method-specific handling versus a generic algorithm. Joe Andrieu raised concerns that prioritizing method-specific logic could undermine interoperability by forcing clients to understand the intricacies of every individual DID method. There was also a significant discussion regarding whether a dereferenced DID URL should return the actual resource or merely a URL pointing to it. Stephen Curran suggested moving method-specific handling further down the priority list in the algorithm to address some of these concerns, though "violent agreement" and confusion over specific implementations persisted throughout the session.

The meeting concluded with a dive into terminology and architectural definitions, debating the roles of the "client," "resolver," and "dereferencer." Participants examined the applicability of the RFC 3986 reference resolution algorithm, with some members arguing that its behavior—such as overwriting paths rather than appending them—might not align with the group’s specific use cases. While no final resolution was reached on the PR, the group identified specific action items for ongoing issues and agreed to continue the technical deliberations in their next meeting.

Otto I. Mora
otto@mora.cr<mailto:otto@mora.cr>