Rafael Schouten

> I'd like to see ArrayInterface move to JuliaArrays first, That would be great. I thought it was a stretch to propose moving array traits from ImageCore to SciML. But...

@timholy an @johnnychen94 I'll put together Morphology.jl and write the PR to depend on it here once the ImageCore.jl traits are sorted. Reading them I'm not sure I can do...

Just checking in on this, and where the ImageCore.jl traits are at now?

It seems pretty annoying for users to have to do anything at all to plot geometries. LibGEOS can't take a dependency on GeoInterfaceMakie.jl?

Is it really too large a dependency to just include here? (Not as a subpackage) MakieCore.jl only depends on Observables.jl (which has no deps). @visr are you thinking this still...

Most of my packages use recursion everywhere. Especially DimensionalData.jl and DynamicGrids.jl. I have a policy against generated because it reduces extensibility. But sometimes you need it anyway when something gets...

Thanks for organising this, unfortunately I'm too busy with travel and deadlines for my modelling packages to help much. But I'll update all my own packages when this is in...

I'm wondering about the version as well, especially given packages are putting upper bounds on everything for auto-registration. My pull-request for Unitful.jl needs an upper bounds before it can be...

Any updates on this @Tokazama? looks interesting, apologies for missing it in april

I'm 100% for using generated functions. It's a guarantee that Julia versions won't break performance.