Jeremie Knuesel

Regarding the readability by focusing on operations, I think it's not that clear cut: sometimes `_` really helps to avoid meaningless names, but names can often be chosen judiciously to...

> so your argument is that because ordinary lambdas are not playing nicely with pipes we have a design issue in a shorter lambda notation that should not behave differently...

Thanks for the link, I couldn't figure out how to make `set_default_plot_format` do anything. I was expecting something like in Plots.jl, where you can do `plot(..., fmt=:png)` to show a...

Yes, in IJulia I'm currently using `plot(...) |> PNG()` which works fine but must be done for each plot. So it would be nice to have an easy way to...

What about the following? * `color="red"` and `color=:red` refer to DataFrame columns * `color=["red"]` refers to a group label * `color=[:red]` refers to a predefined color * ~~`color=RGB(1,0,0)`~~ `color=[RGB(1,0,0)]` can...

I think my previous idea is too inconsistent with how `aesthetic=[...]` generally works (I didn't realize that `color=["red"]` works like a DataFrame column with automatic cycling). Much better to keep...

Thinking about it a bit, I have a preference for the no-cols version, to me the `cols` argument seems an unnecessary complication in this case, so we have to write...

Ah I see how what I propose can be *more* confusing for some people. Maybe the syntax proposed by @nalimilan is less confusing? In the example that would still be...

DataFrames.jl could also provide its own string macro for convenience, for example `cols"t" .=> [length sum]`

This came up on [discourse](https://discourse.julialang.org/t/how-to-properly-add-two-new-columns-in-dataframe-from-a-function-that-returns-two-arrays/63964) today, with the complication that the ByRow function returns several columns: ```julia julia> df = DataFrame(a=[1,2]) 2×1 DataFrame Row │ a │ Int64 ─────┼─────── 1...