Change control of standards documents about standards-maintenance HOT 5 CLOSED

I note the publication of ConsumerDataStandardsAustralia/standards#50 (comment) and observe that it doesn't answer some of the above questions; in particular there is no description of how the actual text in the standard is changed and whether or not this would follow the expected model of having one pull request created that contains changes to the standard for one decision, linked to the issue and with the opportunity for the actual new text in the standard to be reviewed and potentially clarified/improved before being submitted for final approval.

from standards-maintenance.

As the approval process for the standards is via the Data Standards Chair the model of using decision proposals for changes has been adopted.

Following the approval of a change a branch and pull request is created but this is managed using a private staging repository.

Once the DSB team has reviewed the changes and confirmed that they align with the approved Decision the PR is merged and then the changes are pushed to the public repository.

As a result, while the public repository does not contain the PR and Branch used to create the changes, the commit history still contains all of the changes and can be used for comparison purposes.

from standards-maintenance.

Thanks for the response!

Following the approval of a change a branch and pull request is created but this is managed using a private staging repository.

Once the DSB team has reviewed the changes and confirmed that they align with the approved Decision the PR is merged and then the changes are pushed to the public repository.

I'm not sure I understand the rational for:

1. Why the review of whether the actual text changes reflects the agreed decision is being done in private

2. Why there is no public opportunity for review of the text prior to it being finalised

Whilst I understand that the legal need for review by the Data Standards Chair causes some deviation from the processes you'd expect open standards to follow (e.g. the processes used to agree the FAPI, OpenID Connect and OAuth 2 specs that the Australian standards built on top of), these deviations don't seem to relate to that requirement. Can you expand on the rational for the above lack of transparency please?

from standards-maintenance.

The only activity that is performed internally is the iterative development of the changes. The consultation process and resulting output are all fully transparent and are open to critique. There have been a number of occasions where the text incorporated to adopt a decision was ambiguous or in error and needed to be corrected and this was highlighted by the community. These changes are also transparent.

The actual process for developing the standards artefacts is a choice for the DSB team.

It should be noted that the standards, while openly and transparently consulted upon, are not openly developed. They are explicitly a centrally endorsed and managed set of artefacts.

from standards-maintenance.

@CDR-API-Stream Perhaps @jogu's point could be addressed by putting in a summary of changes into the description of a release, and tagging all our releases in Git? The details would then be visible in: https://github.com/ConsumerDataStandardsAustralia/standards-maintenance/releases

from standards-maintenance.