Pietro Vertechi

```julia surface(kde, (x,y)) ``` This is just a reminder to self, I was showing some demos of the StatsMakie thing and this one came out funky (in combination with datasets...

Related, always from the discourse thread: > we could probably allow to use `groupedbar(["a", "a", "b", "b"], [1,2,3,4])` to give the same plot as `groupedbar(["a", "b"], [1 2; 3 4])`,...

I also think it does not really belong here, but I am not sure what would be the correct package.

I think that would be odd. Especially because there already is `DataFramesMeta.@with` that does something very similar (but for DataFrames, rather than arbitrary tables). Maybe it'd make sense to put...

I'm facing the same issue in StatsMakie, any idea what a good keyword would be? I'm thinking `shrink = 0.8` but maybe there are better options.

> I don't think they can be controlled seperately within one series. A bigger barwidth implies smaller spacing and vice versa, no? In practice for a single series only one...

I can't reproduce the issue: ```julia julia> t = loadtable("test.txt") Table with 9726 rows, 13 columns: Columns: # colname type ──────────────────────────────────────── 1 is_bad Int64 2 emp_length Int64 3 annual_inc Float64...

I was thinking that in practice, in the large majority of cases, the characters that appear dynamically are numbers, the `.`, their uppercase counterpart, and the times symbol. Could we...

The same functionality of this packages is available on [StructArrays](https://github.com/piever/StructArrays.jl) which works on julia 1.0.

Thanks for taking a stab at this! My main concern is that ideally this would have a lazy implementation (just like `lag`, `lead` normally do), so that [this line](https://github.com/JuliaArrays/ShiftedArrays.jl/pull/37/files#diff-278249179b68fe78888f2d5e18630430R142) is...