You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Obviously, this crate has not received much attention from the maintainers for the last two years at least. If the current maintainers don't have the time, energy, or will to spend time on ndarray-linalg, maybe it should be open to other interested maintainers. Not even a PR has been merged for two years, or any dependency updated.
For example, ndarray-linalg is not even compatible with the latest ndarray, which means that projects using both ndarray and ndarray-linalg (for example through a workspace) may have to drag in several incompatible versions of ndarray (0.15 and 0.16) in their binaries.
Would the current maintainers, or the rust-ndarray organization owners, be open to welcoming new maintainers?
Hi @samueltardieu, good that you bring this up. This is also being discussed in some other places.
As a main maintainer of ndarray itself I have chosen to not take on linalg because it would be far more than I have time for. I would welcome new maintainers and if I understand correctly @termoshtt is also looking forward to doing this.
However as a short term measure I would also request that I and @rust-ndarray/release are added as crate owners on crates.io. I can then help - or find help - for making an impromptu release to update for new ndarray. New maintainers for further development would still be needed.
The discussions over in rust-ndarray/ndarray#1272 should be relevant for this crate, among other things the topic of what future linalg support should look like for ndarray. (Of course a crate like ndarray-linalg that acts as an extension, could always continue to exist if it wanted to, even if main linalg support in ndarray changes to use some other library.)
Hi.
Obviously, this crate has not received much attention from the maintainers for the last two years at least. If the current maintainers don't have the time, energy, or will to spend time on ndarray-linalg, maybe it should be open to other interested maintainers. Not even a PR has been merged for two years, or any dependency updated.
For example, ndarray-linalg is not even compatible with the latest ndarray, which means that projects using both ndarray and ndarray-linalg (for example through a workspace) may have to drag in several incompatible versions of ndarray (0.15 and 0.16) in their binaries.
Would the current maintainers, or the rust-ndarray organization owners, be open to welcoming new maintainers?
Ping @termoshtt @jturner314 @bluss @LukeMathWalker
The text was updated successfully, but these errors were encountered: