[crispyambulance]

I really like great tables and its cousin in R, gt. You know they're taking a long view when there are photos of CLAY TABLETS and VISI-CALC in the article. Bravo!

Tables, when you really care, are so very difficult to get right. Sometimes you really want to densely compact the data to communicate all details to others who are already deeply invested in the dataset, sometimes you need to remove all but the most essential information to get across a single idea with clarity, and then there are a continuum of variations between those extremes. The problem expands into more dimensions when the media becomes a consideration-- you simply can't (and should not) use the same approach when dealing with pdf's vs html vs a slide-deck. On top of all that, you often have a personal style that you want to get across, or have a style that you need to comply with in some inflexible way.

I like how gt just NAMES the parts of table in their docs (and in the schematic in the article). This is a problem where agreeing on what things are called makes a ton of difference in usability.