plehegar
commented
1 year ago [[ The Working Group: [...]
- must not have made any substantive changes to the document since the most recent Candidate Recommendation Snapshot, other than dropping features identified at risk. ]] 6.3.9. Transitioning to Proposed Recommendation
I don't see any feature at risk listed in the last Candidate Recommendation Snapshot. Yet, the change log contains lots of substantive changes.
You need to follow 6.3.8.1. Publishing a [Candidate Recommendation Snapshot](https://www.w3.org/2021/Process-20211102/#candidate-recommendation-snapshot). This contains by reference the following requirement: "must show that the changes have received wide review.". Was there a wide review of the substantive changes done since the last Candidate Recommendation Snapshot?
Document titles, URLs, estimated publication dates
Abstract
Status
Will new features be allowed to be incorporated in the Recommendation?
Link to group's decision to request transition
Changes
Requirements satisfied
Quoting from the main requirements in the WG charter:
The WG has:
The WG believes that fully satisfied the requirement for the clarifications needed in the specification, as well as establishing a comprehensive testing environment.
Dependencies met (or not)
There were no major dependencies.
Wide Review
Horizontal reviews:
See horizontal review tracker.
The WG discussions were conducted through github issues. The group took over the repository used to develop earlier versions of EPUB, and that meant ≈100 open issues at the time the Working Group started (September 2020); some of those were around for years. During the lifetime of the Working Group ≈499 issues were added, coming both from the wider community and the Working Group itself, with the participation of experts outside the group.
Issues addressed
The change log items listing the substantial changes, i.e.:
include references to the issues that were addressed by those changes.
Formal Objections
None.
Implementation
The exit criteria of the Working Group are described in:
Reports of the testing and implementation results can be found here:
There are numerous EPUB 3 Reading System implementations on the market, some major ones are represented on the Working Group (Google, Apple, Kobo, EDRLab, VitalSource, Colibrio, etc.). We have received testing results from all of these participants, and more. Also, the group has numerous contacts with Publishers, as well as accessibility organizations, who have contributed to the CR testing of, e.g., vocabularies.
Finally, there is a strong cooperation between the Working Group and the developers of EPUBCheck, that already has a beta version of an EPUB 3.3 checker. (This is particularly important, because epubcheck is used by virtually all publishers before releasing a new publication.) You can find results of EPUBCheck testing here: https://w3c.github.io/epub-structural-tests/
Patent disclosures
Additional note on PR Transition
The Working Group's charter has a very strong clause on backwards compatibility:
What this means that the usual CR approach, whereby if a feature does not get the right number of implementation per the exit criteria must be removed from the specification, cannot be applied in this case. Instead, the WG has introduced a special category called "under-implemented features" to mark those features in the final publication.
Only one feature was found during testing to require this category: https://www.w3.org/TR/epub-33/#attrdef-dir. We have kept this feature in due to its importance for i18n support.