auto-publishing using echidna?

[brentzundel]

I thought that all of the work items adopted by the VCWG had resolved to use Echidna to auto publish, but I'm not seeing an updated version at https://w3c.github.io/vc-status-list-2021/
Is this a known issue?

[iherman]

@brentzundel someone has changed the short name of the document, which probably led to errors.

• The latest editors' draft on the repository refers to vc-bit-string-status-list
• The latest version on /TR refers to vc-status-list-2021

I presume that is the problem, echidna has been set up for a specific short named document. The (painful) changes we made with jose-cose shows what has to be done if we want to get it right. I do not even remember having had a discussion about this.

My advice is to stay with vc-status-list-2021. It is just too much hassle at this point...

cc @msporny @mprorock @OR13 @mkhraisha

[TallTed]

The (painful) changes we made with jose-cose shows what has to be done if we want to get it right ... [which] is just too much hassle at this point

I sympathize with this, but even though URIs (and arguably, document short names) should be opaque (as I often vocalize), there is (and will be, for a long time to come) significant difference between what humans will read into vc-bit-string-status-list and vc-status-list-2021, and that difference will bite many over time.

I think the change is worthwhile.

I have hopes that the recent experience of having done this for jose-cose left a track that can be retraced for bit-string-status-list (and maybe written up for future humans who may need to do similar in/for other WG documents).

[OR13]

Happy to help however I can, but unfortunately there are some things only staff can do regarding name changes.

[iherman]

Yeah, if the WG decides to do this, then we can of course go through.

Just for the records in case we do this:

• see Change short name to vc-jose-cose vc-jose-cose#115 (comment) for what has to be done process wise
• see Rename work item vc-jose-cose#125 (comment) for some more comments on what/how to manage some details

[TallTed]

+agenda?

[iherman]

The issue was discussed in a meeting on 2023-09-15

List of resolutions:

• Resolution No. 1: the new shortname is vc-bitstring-status-list.

View the transcript

1.3. auto-publishing using echidna? (issue vc-status-list-2021#82)

See github issue vc-status-list-2021#82.

Brent Zundel: intro, spec from april, fix for this is straightfoward, we need to go back to call vc status 2021 ... unless we as a group change the record name ...

Brent Zundel: we either change it back or we have a formal resolution to change the name ...
… ivan and the editors ...

Manu Sporny: just need to change the name of the spec ... no longer vc status list 2021 ... use something less generic.

Dmitri Zagidulin: @TallTed - sure, that's business logic. I'm just wondering why you're saying "VCs weren't designed for this". business logic is a huge part of VC validation, it's absolutely designed for it.

Manu Sporny: maybe vc multi status list ... vc stream status list ...
… hesistant to change it today.
… interested in time boxing it.

Brent Zundel: 2 mins.
… if something can come up with a good name in the next 2 mins.

Andres Uribe: I likevc-bit-string-status-list.

Kristina Yasuda: why is this so complicated? why can't we just remove 2021 from the name?

Brent Zundel: that would work.

Manu Sporny: -1.

Gabe Cohen: +1.

Dmitri Zagidulin: +1.

Brent Zundel: would that work for everybody?
… anyone opposed?

Manu Sporny: yes, too generic.

Andres Uribe: 0 (prefer vc-bit-string-status-list).

Manu Sporny: bit string status list.

Brent Zundel: anyone opposed?

Kristina Yasuda: i do think you need status list of "what"?

Brent Zundel: no one wants to die on the hill.
… i'll draft a proposal.

Manu Sporny: there is also ... nevermind.

Kristina Yasuda: what was manu's proposal?

Joe Andrieu: seems like misalignment between two names ...

Ivan Herman: lets not rename the doc in three months ... lets do them both at the same time ...

Proposed resolution: the new shortname is vc-bitstring-status-list. (Brent Zundel)

Manu Sporny: +1.

Gabe Cohen: +1.

Joe Andrieu: +1.

Shigeya Suzuki: +1.

Ivan Herman: 0.

Dmitri Zagidulin: +1.

Andres Uribe: +1.

Brent Zundel: +1.

David Waite: +1.

Ted Thibodeau Jr.: +0.

Hiroyuki Sano: 0.

David Chadwick: +1.

Kristina Yasuda: 0.

Jay Kishigami: 0.

Paul Dietrich: 0.

Resolution #1: the new shortname is vc-bitstring-status-list.

Brent Zundel: not seeing any opposition.
… we are resolved.
… practical level, who is going to fix the spec.

Ivan Herman: who wants to help?

Manu Sporny: i can help.
… i can do the naming changes.

Ivan Herman: i have put a comment into the issue where i summarized what we need to do in the COSE/JOSE change.

Brent Zundel: good conversation ... we still have like 25 issues to look at.
… and 30 mins to do it.
… if the editors have preferences for how to prioritize ...
… take advantage of the people in the room.
… jumping into 81.

Manu Sporny: skip that.

[iherman]

The WD has been published under the new short name, i.e., https://www.w3.org/TR/vc-bitstring-status-list/. A PR has been raised for the proper changes of echidna (see #95). I propose to close this issue as settled.

cc @brentzundel

[msporny]

@iherman auto-publishing isn't working because the publishing token is invalid for some reason:

https://github.com/w3c/vc-bitstring-status-list/actions/runs/7349609343/job/20009850064#step:3:683

https://labs.w3.org/echidna/api/status?id=7c84e63d-317a-45ad-ab2d-17aa48928fee

I believe auto-publishing will work again once the Echidna token is updated/set.

[iherman]

@msporny I have generated a new secret token (the one in the repository was bound to the previous short name...), and refreshed the value in github. Can you test again?

[msporny]

Unfortunately, the Echidna publication process is still failing (same error):

Same error (Echidna logs: https://labs.w3.org/echidna/api/status?id=5f32a89a-36fe-4df9-96e5-bdfcd75e9f4f):

      "token-checker": {
        "status": "failure",
        "errors": [
          "not-authorized"
        ]
      },
      "

 Token Checker Errors:
    not-authorized

[iherman]

Well... no idea then. This is @deniak territory...

Denis, help! :-(

[deniak]

I'm not sure what happened but I updated the token in GH and I re-triggered the action which passes now.

[iherman]

Well... echidna needed your personal touch, @deniak...

Thanks!

[msporny]

This issue has been resolved, closing.