When is a table not a table?

[sean-fitzpatrick]

Long ago we had some debate over many of the tables in APEX: should they be table/tabular or figure/tabular?

In the original conversion, we did table/tabular. Then it was pointed out that many of the tables (such as tables of values) are not tables as defined in the Chicago Manual. We changed these to figure/tabular.
But this actually violates the schema!

Things are still working -- there is support in the code even if officially it is not allowed.

The question is: should we consider whether tabular might be inappropriate for some of these?

In the Taylor polynomial section I found that we got improved formatting by replacing figure/tabular with figure/p/md. That is, something like a short table of values might work better as an aligned environment in math mode.

Thoughts?

[APEXCalculus]

I need to table this for now as three different areas of life are converging with busyness. (Table it? Get it?)

I've always been a bit out-of-step with the PreTeXt philosophy as I'm more concerned with how things look vs. the portability to other formats. But the latter is undeniably important.

If figure/p/md looks good and is future-proof, that's fine. But truthfully I'll have to ponder this for a while once my plate clears (after Sunday? Tuesday?) a bit. (I guess my favorite option would be for figure to officially support tabular.)

[sean-fitzpatrick]

It's not urgent. Things work as they are, and the only way we know otherwise is the validation report.

(Until I figure out a way to put the vshift content into the publication file, we're going to get all kinds of validation errors anyway.)

There is a small risk that tabular in figure stops working some day and we're forced to deal with this, but until then we can leave it.