Nathan Boyer

I'll call on @pdeffebach and @CameronBieganek to add their thoughts too.

I'm afraid I don't know enough to be helpful with the implementation details. Could you start by supporting more basic types and just print a warning "Validation not yet supported"...

Did `normalize_selection` get more features in another PR or is there really nothing that can be done to get better error checking on Pairs?

The best examples I have are linked in the [discourse post](https://discourse.julialang.org/t/frustrated-using-dataframes/67833) which prompted this issue. I am currently writing a documentation PR for the "Basic Usage of Transformation Functions" section...

One thing that is confusing is that the first argument of the pair is not what is passed to the function. In `source` => `fun`, the `source` is not passed,...

**Sidenote** I did: 1. `] dev DataFrames` 2. Created a new branch `origin/nb/docs_transformations` 3. Edited `docs/src/man/basics.md` 4. Committed my changes to `origin/nb/docs_transformations` 5. Clicked "Publish Changes" in VSCode. But I...

> We potentially might list all classes of valid column selectors. Is this what you think should be changed? That might be a good idea, or at least a link....

I was thinking that `typeof(select(df, source))` might help me construct my `function` correctly in the `source => function` `Pair`, but the output of that command is always just `DataFrame`. Maybe...

> Unfortunately they use a different syntax as filter has to work row-wise as opposed to all other functions. Still - probably you as a user can propose something that...

I think it would be better to keep the update method contained within Julia so that ... 1. juliaup doesn't leave remnants in important system files 2. juliaup doesn't have...