Core Team Meeting Notes

Announcements
1

On last Scala Core team meeting we decided to start publishing the meeting notes for each of the meetings. This thread will serve as a place to publish them and will be locked to other comments. If you want to discuss a particular point you can always open a separate topic.

If there are any issues related to the compiler that you want to have on the agenda you can always propose them, but decision to include them will be up to the people in the core team. Most likely we will not include any simple matters and only focus on topics that need an actual decision or discussion.

I will follow up with the most recent meeting notes in a bit.

14 Likes
3

Core team meeting 12.08.2026

Attendees:

  • Piotr Chabelski as Product Manager

  • Martin Odersky

  • Oliver BraÄŤevac as LAMP Representative

  • Wojciech Mazur as Scala Native Representative, Release Officer and CB Maintainer

  • Tomasz Godzik as Coordinator and VirtusLab Representative

  • Lukas Rytz as Scala 2 Maintainer

  • Seth Tisue as Community Advocate

  • SĂ©bastien Doeraene as Scala.js Representative

Guests:

  • MikoĹ‚aj Fornal

Topics:

  1. https://github.com/scala/scala3/pull/26497 who can help out with getting java records pattern matching reviewed for 3.10?

  2. Releases status and discussion.

  3. https://contributors.scala-lang.org/t/proposal-fixing-null-methodofany-under-explicit-nulls/7504 and its PR https://github.com/scala/scala3/pull/26771 (from Seb)

  4. Should we post meeting notes in a public capacity? (from Tomasz)

  5. Should add a section about backporting to LTS? (by default it’s best effort, but for important fixes we could do max effort)

Postponed

  1. SLC: Automatic conversion to Option https://contributors.scala-lang.org/t/slc-conversion-a-option-a/7514/37

  2. https://github.com/scala/scala3/pull/26668 Tighten rules for implicit conversions

Conclusions:

  1. Lukas Rytz will take a look at the PR and review it.

  2. There are a number of potential regressions in 3.10 that need to be looked at. Some of them are expected due to recent changes and fixes, but we will need to make sure none of them are unexpected.

  3. The approach was accepted and the PR can be merged.

  4. We will post the meeting notes on the contributors forum.

  5. We will add a note to the PR template and mention this to the compiler team.

3 Likes
4

Core team meeting 19.08.2026

Attendees:

  • Piotr Chabelski as Product Manager

  • Martin Odersky

  • Oliver BraÄŤevac as LAMP Representative

  • Wojciech Mazur as Scala Native Representative, Release Officer and CB Maintainer

  • Tomasz Godzik as Coordinator and VirtusLab Representative

  • Lukas Rytz as Scala 2 Maintainer

  • Seth Tisue as Community Advocate

  • SĂ©bastien Doeraene as Scala.js Representative

Guests:

  • MikoĹ‚aj Fornal

Topics:

  1. Java Records pattern matching implementation (with Mikołaj)

    1. https://github.com/scala/scala3/pull/26497#pullrequestreview-4936091864

    2. how should we handle java records varargs in pattern matching?

    3. the case of JavaRecordFields annotation

  2. Releases status

  3. Current status of Metals Worksheets and Mdoc being reliant on the internal compiler APIs.

  4. SLC: Automatic conversion to Option https://contributors.scala-lang.org/t/slc-conversion-a-option-a/7514/37

  5. https://github.com/scala/scala3/pull/26668 Tighten rules for implicit conversions

Conclusions:

  1. In the implementation, the annotation solution seems fine. For pattern matching, yes let’s emit an `unapplySeq` so pattern matches read naturally in Scala (accepting the risk that Java programmers will find this odd/different since, Java unpacks it to Array in pattern matching).

  2. We discussed whether the long series of 3.9 RCs implies we ought to be doing anything differently, e.g. w/r/t the community build. Conclusion: no, we don’t see an opportunity for changes here. We are being especially careful since 3.9 will be LTS.

  3. We will introduce a workaround in Mdoc for Scala 3.10.x, but introduce a concept of worksheets into REPL later on. Multiple tools will be able to interact with it and will give us an opportunity to have a stable API.

  4. Pending an official implementation, but overall it most likely will be accepted provided we also move ahead with 5.

  5. There are two questions here, the timetable of introducing warnings and errors, and the longer-term status of `implicit def` itself. We may need a fresh round of community input in order to find out how many libraries are still dependent on `implicit def` (for real reasons, not just inertia). Seb has proposed considering keeping `implicit def`, if it has turned out that `Conversion` isn’t powerful enough. The warning for Conversion will most likely be merged into 3.10 becoming an error from 3.11.

5 Likes
5

Core team meeting 26.08.2026

Attendees:

  • Martin Odersky
  • Oliver BraÄŤevac as LAMP Representative
  • Wojciech Mazur as Scala Native Representative, Release Officer and CB Maintainer
  • Tomasz Godzik as Coordinator and VirtusLab Representative
  • Lukas Rytz as Scala 2 Maintainer
  • Seth Tisue as Community Advocate
  • SĂ©bastien Doeraene as Scala.js Representative

Absentees:

  • Piotr Chabelski as Product Manager

Topics:

  1. Updates

  2. Releases status

  3. Bounded opaque type no longer provides typeclass instance of the bound type · Issue #26849 · scala/scala3 · GitHub - should we fix this?

  4. There are many issues coming out of -> becoming inline - can we fix those for 3.10?

    1. `Predef.->` as extension is always chosen over user-defined implicit class member · Issue #26851 · scala/scala3 · GitHub
    2. Regression in lichess/lila during access to contextual member in Tuple arrow · Issue #26801 · scala/scala3 · GitHub
    3. Macros crash in `Inlines.inlineCallTrace` after changes to `->` desugering · Issue #26796 · scala/scala3 · GitHub

  5. `Tuple` `toList` method and friends with unusable signatures (MT application to singleton types) · Issue #26817 · scala/scala3 · GitHub

    1. based on discussion under Regression: "a match type could not be fully reduced" when flattening a list of tuples · Issue #22022 · scala/scala3 · GitHub, it seems the error is valid/expected?
    2. is this a valid regression or a won’t fix?

  6. Changes to Akka status in the core team.

  7. Scaladex publishing is no longer triggered by Sonatype

Conclusions:

  2. 3.9.0 is currently being published. Work on 3.3.9 is getting underway.

  3. We should try to make Ordering contravariant, which should help in this case, and then run it on the open community build to verify if this wouldn’t break existing libraries. If this proves to be problematic we will need to introduce an explicit Ordering. The change that caused this will not be reverted itself.

    1. #26851 is expected and can be worked around by using an extension method
    2. #26801 is a regression, might be fixed by removing inline from thatparameter
    3. #26796 might be related to b)

  5. The behaviour itself is expected when it comes to match types due to their limitations. We can still improve the methods itself to force a better inference and fix the issue.

  6. Scala 2 is now being maintained by VirtusLab and will be maintained in the same capacity as it was before by Akka. We will publish a short blogpost about it, but in short no dramatic changes are happening from the user perspective.

  7. Most Scaladex reliability problems may have been fixed, but there is still some effort to improve it further. Open ticket or comment on existing ticket if you find any issues to let us know. There is a separate issue of missing libraries, because Sonatype shut down the channel that Scaladex was receiving notifications from. Alternatives are being investigated and we are trying to contact Sonatype.

2 Likes
6

Core team meeting 02.09.2026

Attendees:

  • Martin Odersky
  • Tomasz Godzik as Coordinator and VirtusLab Representative
  • Lukas Rytz as Scala 2 Maintainer
  • SĂ©bastien Doeraene as Scala.js Representative
  • Piotr Chabelski as Product Manager
  • Seth Tisue as Community Advocate

Guests:

Absentees:

  • Oliver BraÄŤevac as LAMP Representative
  • Wojciech Mazur as Scala Native Representative, Release Officer and CB Maintainer

Topics:

  1. Updates

  2. Releases status

  3. Should we release scala-xml and parser combinators with -Yfuture-lazy-vals? Support Java 25 without warnings · Issue #828 · scala/scala-xml · GitHub

    1. Release new major and allow backports to current?
    2. What version to release on 11 or 17?

  4. Time slot conflict for the Core meeting for SĂ©bastien, starting next week

Conclusions:

  3. We will release both libraries under 3.3.8 with JDK 17 and the -Yfuture-lazy-vals flags. This will be a new minor release for both libraries.
  4. We switched the meeting to Thursdays.