All three of us showed up to a shorter meeting that dealt entirely with triage. Last week’s new blocker was resolved in the meantime, but we had a few new small ones show up, which we merged fixes for. The last-minute IO::Compress release has needed some oopsie fixes. Another release candidate is inevitable and will be shipping this week.
[P5P posting of this summary]
Once again we were all present.
- A new blocker appeared. It is only just impactful enough to register, but unfortunately not negligible enough to ignore. We will attempt a resolution before we consider other options.
- There has been some more activity in the LLM policy discussion, where we didn’t get around to posting our thoughts last week. We will rectify that.
[P5P posting of this summary]
All three of us attended.
- We accepted a number of last-minute patches provided by Karl to improve our z/OS support stance, since they are strictly limited to that system and largely touch things that were broken there anyway.
- The worrying situation on CPAN Testers for the latest
HTTP::Tiny situation cleared up; it was only caused by smokers-only tests against a live system that was intermittently unresponsive. Consequently we have now synchronized with CPAN, which means every one of the late-in-cycle release blockers we had is now addressed.
- Leon had already begun work on RC1 even before that.
- We touched base on LLM policy based on the feedback we have had on the thread so far. We have been chewing on some thoughts about that which are in draft state; we agreed that we should actually send them out soon.
- We discussed one of the two new PPC PRs, about viral value magic, which Leon and Aristotle realized is more similar to tainting and its limitations than previously understood. No verdict was reached; we will continue this discussion.
[P5P posting of this summary]
This week we had to reschedule to Tuesday in order to attend in full strength.
- Release blocker triage continues and fortunately has not turned up anything new, but unfortunately has not delivered full resolution of already-known issues either: CPAN Testers says that HTTP::Tiny 0.096 is not a shoe-in, as we had hoped. We will have to evaluate the reports more closely.
- We nevertheless intend to begin work on RC0, since it mostly consists of writing the perldelta.
- None of us had time to spend on the LLM policy discussion this week.
- We discussed the two new PPC PRs in very general terms but need to return to them in more depth next week.
[P5P posting of this summary]
We were all present.
- 5.43.11 has so far turned up a few problems, thankfully small. As a result there are new versions of Archive::Tar and HTTP::Tiny to sync, which we intend to merge.
- We decided how to proceed with our schedule: we will wait another week for any other findings in 5.43.11, and if nothing else shows up then we intend to start working on 5.44.0-RC0 at that point.
- We discussed briefly the trajectory of the core team’s LLM policy conversation. For now we continue keeping an eye on the thread.
- We noted PRs to add two new documents to the PPCs repository but didn’t have time to discuss them in this meeting. We intend to do so in the next one.
[P5P posting of this summary]
