Ideas for future array SRFIs: nominal data John Cowan (12 Apr 2026 04:16 UTC)
Re: Ideas for future array SRFIs: nominal data Peter McGoron (12 Apr 2026 04:56 UTC)
Re: Ideas for future array SRFIs: nominal data Alex Shinn (12 Apr 2026 12:16 UTC)
Re: Ideas for future array SRFIs: nominal data John Cowan (13 Apr 2026 16:24 UTC)

Re: Ideas for future array SRFIs: nominal data John Cowan 13 Apr 2026 16:24 UTC

Thanks, I'm reading the Petersohn article now.  A dataframe algebra as
an extension of the relational algebra is very appealing.

On Sun, Apr 12, 2026 at 12:56 AM Peter McGoron <xxxxxx@mcgoron.com> wrote:
>
>  > This is an idea for a wrapper around SRFI 231 arrays that allows the
> dimensions to be named by symbols, and for each dimension, the
> individual rows/columns/whatever to be named by symbols as well
>
> This sounds a bit like Pandas dataframes. They have some foundation in
> the relational algebra [2][3], but those are 2D. I don't know how much
> it generalized to n > 2 dimensions.
>
> [2]:
> https://mchav.github.io/what-category-theory-teaches-us-about-dataframes/
> [3]: https://arxiv.org/abs/2001.00888
>
> -- Peter McGoron
>