The appendix contains loads of statistics in tables. The first few pages are all collected into a rather large and unwieldy table. I'm thinking of breaking these up into smaller tables. With the right title, I can also greatly reduce the amount of repeated text and ditto marks. And I can then add meaningful column headers. This is in the spirit of the Web Accessibility Initiative which SEMOS links to, and allows for better "structural markup". A little editorial license would make it all more understandable for an e-reader.
Example: gather population numbers in a table, "Population of various cities", then in the rows give location, year, source, number. Rather than repeating "Population of" or ditto marks on every line.
Example: gather all the "Density of population to the square mile..." rows into a table with that title, and then in the rows just provide year, location, number.
Where there are still a lot of dittoes, is it better to just repeat the information (copy/paste is cheap)? Just thinking how I'd comprehend it if I was listening to it being read by software.