Matt Bauman

Yes, history and inertia is one part of the story. The other part is that it is practically quite useful and (I think) friendlier. Arithmetic on a zero-dimensional array is...

The best way to evaluate this change would be to try it out. Personally, I don't find the status quo so abhorrent, so I won't be pushing for this change,...

Here's a compelling argument: https://discourse.julialang.org/t/broadcasting-and-pairs-using/24739 In short: ```julia p = "2".=>"two" replace.(["123", "246"], p) ``` vs. ```julia replace.(["123", "246"], "2".=>"two") ```` The first fails, whereas the second succeeds, but the...

Here you go: https://github.com/JuliaLang/julia/compare/mb/true28866 Just: ```sh git clone https://github.com/JuliaLang/julia.git cd julia git checkout mb/true28866 make make testall ```

Just an administrative note: this issue started out describing both broadcast's design as well as the bug in #32122 — and I wrote the commit message there before we really...

Ironically, that's happening because FillArray's `Fill` defines its own broadcast implementation that **_preserves the 0-dimensional array_** as requested here. I agree this is how things _should_ behave, but changing this...

~~Looks like this is introducing a regression that's adjacent to https://github.com/JuliaLang/julia/issues/32366#issuecomment-503826041. `sum([1 0 missing], dims=2)` works on master but is currently broken here (and I don't think is tested).~~ Now...

I think I've addressed the most egregious of the performance issues, but I've only done spot tests so far. Now I need to tackle those stdlibs.

@tecosaur I suspect you meant that comment for `#55351`

It's all of the above. This needs: * Work to get the stdlibs passing tests * Running of PkgEval and examining the results * Fixing the issues @N5N3 flagged *...