R4ds: feedback: link together relational data and tidy data

[franknarf1]

The tidy data paper (which I routinely link to on SO) mentions:

Each type of observational unit forms a table

But this important point does not appear in r4ds's chapter on tidy data. Nor does it appear in the relational data chapter, which opens with (paraphrasing) "You may end up having multiple tables, somehow."

Anyway, so: I'd hoped to see this point covered in the book, maybe bridging those chapters.

Note: saw your request for feedback on twitter; and not being a twitterer, assumed you'd welcome feedback here. Sorry if I'm misusing your issue tracker.

[hadley]

I'm planning on covering this in the 2ed (whenever that happens)

[franknarf1]

Great! Feel free to close this if/when appropriate.

[hadley]

I've been thinking about bunch about the connections lately, and this is likely to lead to substantial changes. I'm going to close because it's very unlikely I'll forget about it.