Hi,
our last release was already in July, and I think we now have a lot of contributions in the dev branches, so I'd like to propose we do a release soon. Now, the question is: Do we only want to release 3.5.5 and 3.6.2 or do we also already want to release 3.7.0 at this point? The CHANGELOG for 3.7.0 currently lists a few changes, but I think the only major change is the added support for mid-traversal E()-steps.[1] Our current plan for the 3.7.0 release however lists quite a number of features that we wanted to implement in that version.[2] We can now either decide to move the features that are not implemented yet (and not expected to be implemented in the next few weeks) to version 3.8.0 or not release 3.7.0 at this point and instead move its target release date to 2023. If we decide to release 3.7.0 at this point, then we should also discuss whether we want to continue supporting the 3.5 line or making 3.5.5 the last release. If we want to continue supporting it, then we have 4 active dev branches (3.5-dev, 3.6-dev, 3.7-dev, and master for 3.8.0). I suggest that we move the target release date for 3.7.0 to 2023 as most of the planned features are not ready yet and because most (all?) graph providers haven't even updated to 3.6 yet which also means that we probably want to continue supporting 3.5. Having to support 4 branches however adds some burden to our development workflow. Any thoughts on this? Now to 3.5.5 and 3.6.2, I think we have quite some contributions ready for these releases, including some updates that fix security vulnerabilities. Are there any open issues / PRs that should definitely be included in these releases? [1]: https://github.com/apache/tinkerpop/blob/master/CHANGELOG.asciidoc#tinkerpop -370-release-date-not-officially-released-yet [2]: https://github.com/apache/tinkerpop/blob/master/docs/src/dev/future/index.as ciidoc#37x---target-22q4
