Skip to content

DID Document Metadata, ECMA-262 citation, CIDv1 bytes, INVALID_DID, DID Resolution pin - #340

Open
jintekc wants to merge 5 commits into
mainfrom
spec-change/bundle-1-small-fixes
Open

DID Document Metadata, ECMA-262 citation, CIDv1 bytes, INVALID_DID, DID Resolution pin#340
jintekc wants to merge 5 commits into
mainfrom
spec-change/bundle-1-small-fixes

Conversation

@jintekc

@jintekc jintekc commented Aug 3, 2026

Copy link
Copy Markdown
Collaborator

One commit per issue, none of the changes overlap. Bundled so it is one review pass instead of five.

jintekc added 3 commits August 3, 2026 18:20
The Resolve operation returns didDocumentMetadata with REQUIRED versionId,
confirmations, and deactivated fields, but the DID Document Metadata data
structure never defined confirmations and marked versionId OPTIONAL. Define
confirmations, mark versionId REQUIRED, and extend the example to match.

Also give block_confirmations a defined initial value of 0 in the resolver
state, so resolving a never-updated DID no longer reads an unassigned value
when returning metadata.
The spec uses encodeURIComponent() in three normative capability URN
formats without defining or citing it. Cite ECMA-262, 16th Edition at
each site and add the bibliography entry. Per discussion, the whole
standard is cited rather than a subsection, for consistency with how
other references are cited.
The recipe called 0x00 the multicodec code for raw binary, but 0x00 is
identity; raw binary is 0x55, which the Raw Leaves mandate requires. The
multihash digest-length byte 0x20 was also missing, so the spec-literal
CID was malformed and no IPFS node would recognize it. Correct the bytes
and add a worked example (file bytes in, exact CID out) as a test vector.
This matches what implementations built on multiformats libraries already
emit.
jintekc added 2 commits August 3, 2026 19:44
INVALID_DID was raised in four places but never resolvably linked: the
resolve chapter's references rendered as literal brackets and the
algorithms chapter linked ./errors.html, which has no such section. Per
the issue discussion, DID Resolution already defines the resolution
errors and we should not recreate them, so the link registry now points
INVALID_DID at that spec's errors section and both identifier
algorithms use the reference.
The spec cited DID Resolution v0.3, a version label upstream has
already moved past (the draft became v1 on 23 July 2026). Per the
issue discussion, cite https://www.w3.org/TR/did-resolution/, which
always points at the latest published version, and say "latest
version" in prose instead of naming one. The errors chapter names the
Errors section instead of a section number for the same reason.

Also state the resolver content types: DID Core v1.1 consolidated the
DID document media type to application/did, and DID Resolution defines
application/did-resolution for the full resolution result. The prose
example previously said application/ld+json while the example file
said application/did+ld+json; neither exists in the current
references, so both now say application/did, and the contentType
property is clarified to always record the DID document's own media
type.
@jintekc
jintekc force-pushed the spec-change/bundle-1-small-fixes branch from f88d425 to 49e8a12 Compare August 3, 2026 23:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant