David Anthoff

In terms of goals the two packages are super similar. [IterableTables.jl](https://github.com/davidanthoff/IterableTables.jl) emerged out of the design of [Query.jl](https://github.com/davidanthoff/Query.jl), where the design of [IterableTables.jl](https://github.com/davidanthoff/IterableTables.jl) (namely iterators of [NamedTuples.jl](https://github.com/blackrock/NamedTuples.jl)) forms the core...

You can now query ``TimeArray``s with [Query.jl](https://github.com/davidanthoff/Query.jl), and also materialize query results into a ``TimeArray``. Here are some examples: ````julia using Query, TimeSeries, DataFrames dates = collect(Date(1999,1,1):Date(2000,12,31)) mytime = TimeArray(dates,...

Oh, one last thing: the implementation for the ``TimeArray`` sink is terrible right now, so that is known to be slow and I'll fix it at some point (davidanthoff/IterableTables.jl#11). That...

I just wrote the same thing in a different PR, but given how non-related that PR is it might be worth repeating this here: I'm really not a fan of...

I'm very nervous about any automatic lifting stuff. I would much prefer that we start with the white list approach, i.e. simply add as many methods to functions that work...

Well, one issue is that I haven't seen a solution to the 3VL issue. How would we handle something like `map(i->i & false, X::NullableArray{Bool})`? This is assuming that the `&`...

Maybe the even more relevant example would be `map(i->isnull(i) ? something : somethingelse, X::NullabelArray{T})`. That would also not work, right? I would find it highly confusing if that didn't work......

So in the end the semantics of `map`, a query expression (I assume your example with `@collect` is jplyr syntax?) and a generator or comprehension would all be different if...

> But map(f, x::NullableArray) should probably return a NullableArray when f returns a Nullable anyway. That doesn't break the AbstractArray interface. Agreed. For complete consistency I almost feel `map(f, x::Array)`...

Shouldn't `Nullable(1)==1` just return `true`? `Nullable(1)===1` could still return `false`. If the goal is to follow the C# convention here, then the lifted versions of the equality and relational operators...