Andy Ferris

Many data sets (e.g. Unchunked, uncompressed ones) will be memory mapped, in which case the OS is already providing a view of the data on disk. Still, something to cover...

It seems very reasonable to me. Perhaps an `AbstractArray` type which is the full view of the dataset, and use e.g. `SubArray` for subsets/views thereof?

@bkamins I would be tempted to try make these three seperate/orthogonal interfaces for perming different mutations, rather than “levels” or layers with some on top of the others. E.g. I’m...

Yes it’s very interesting how mutating a column behaves somewhat the same as mutating the rows. Of course, you can tell the difference when you have access to the column...

OK - I'm not at all familiar with Zygote. Would supporting it mean (a) adding Zygote as a dependency and (b) adding some new methods to some Zygote functions (like...

Yeah OK, seems quite reasonable to me. (Pull requests are always welcome :wink:)

Yeah, that rotations are just matrices is kinda fundamental here. It sounds like the current `conj` on quaternions is a bug. If you want to do actual quaternion algebra, being...

I had never heard of versor before - thanks. It’s an apt name (I love the word!), but probably less well-known to other users so might hurt in discoverability…

> Do you have any thoughts on this? The exact layouts of the structs isn't important to me. This package predates @c42f and I contributing (as well as StaticArrays), so...

Should it just overload `^`? I think that’s equivalent to your `scale`.