---
tags: [meeting-notes]
title: '2024-05-01'
---
# conda-forge core meeting 2024-05-01

Add new agenda items under the `Your __new__() agenda items` heading

- [Zoom link](https://zoom.us/j/9138593505?pwd=SWh3dE1IK05LV01Qa0FJZ1ZpMzJLZz09)
- [What time is the meeting in my time zone](https://dateful.com/convert/utc?t=5pm)
- [Previous meetings](https://conda-forge.org/community/minutes/)

## Attendees

| Name | Initials | GitHub ID | Affiliation |
| ----------------------- | -------- | --------------- | --------------------------- |
| Wolf Vollprecht | WV | wolfv | prefix.dev |
| Jaime Rodríguez-Guerra | JRG | jaimergp | Quansight/cf |
| Isuru Fernando | IF | isuruf | |
| Dasha Gurova | DG | | Anaconda |
| Klaus Zimmermann | KZ | zklaus | Quansight |
| Sylvain Corlay | SC | SylvainCorlay | QuantStack |
| | | | |
| | | | |
| | | | |

X people total

### Standing items

- [ ]

### From previous meeting(s)

- [ ]

### Active votes

- [ ]

### Your `__new__()` agenda items

- [X] (HV) Numpy 2.0 migration
- Now following the same approach as the python 3.12rc migration, using a `_numpy_rc` package.
- https://github.com/conda-forge/conda-forge-pinning-feedstock/pull/5790
- Should be ready to go once [this](https://github.com/conda-forge/numpy-feedstock/pull/314) is merged.
- [X] (HV) Discuss how we might want to follow Diátaxis in the docs, and whether we can agree on a rough [outline](https://github.com/conda-forge/conda-forge.github.io/issues/2164) of how such a website (re)structure would look like
- [x] (JK) Drop CUDA 11.2
- Agreed to drop CUDA 11.2
- [x] (JK) Finishing CUDA 12 migration
- https://github.com/conda-forge/conda-forge-pinning-feedstock/pull/5613
- Discussed CUDA 12.0 & 12.x with x as max
- CI usage
- Most jobs have 11.2 & 12.0 (through migrator) now. So dropping 11.2 and adding 12.x should be roughly the same usage
- Decided to keep discussing in the PR
- [x] (JK) GLIBC 2.28
- Do we have the CDTs we need listed?
- https://github.com/conda-forge/cdt-builds/issues/66
- Would like to add libglvnd
- https://github.com/conda-forge/staged-recipes/pull/25919
- Otherwise this list looks good
- [ ] (HV) Bump `c_stdlib_version` to 10.13 on macOS in global pinning
- In a previous core call we agreed that a linter rule warning about missing `{{ stdlib("c") }}` should be enough to deal with feedstocks that haven't been touched by the piggyback yet.
- The linter rule is now live as of conda-smithy 3.35.
- [ ] (WV) CEPs: Sharded (faster) repodata & OCI registry
- [ ] https://github.com/conda-incubator/ceps/pull/75
- [ ] https://github.com/conda-incubator/ceps/pull/70
- [X] (WV) Bigger Windows runners - prefix would like to sponsor if possible
- [ ] JRG: Related (I think): https://github.com/Quansight/open-gpu-server/issues/31
- [ ] Jaime to compile a list of items for adding a Windows runner to Cirun + admin-requests.
- [ ] Pytorch compilation. Maybe cross compile from Linux?
- [ ] https://github.com/conda-forge/clang-win-activation-feedstock/blob/main/recipe/meta.yaml
- [ ] Would need SDK download instead of relying on the underlying VM.
- [ ] Also, last time this came up, Uwe [warned](https://github.com/conda-forge/admin-requests/pull/949#issuecomment-1963416546) of a "major conda-build bug" with this. I (HV) don't have further context/references than that.
- [X] (SC) Year 2 contract for CZI grant

### Pushed to next meeting

- [ ]

### CFEPs

- [ ]