How can I avoid NULLs in my database, while also representing missing data?

Question

In SQL and Relational Theory (C.J. Date, 2009) chapter 4 advocates avoiding duplicate rows, and also to avoid NULL attributes in the data we store. While I have

Answer 1

So how do you design without NULLS? That was the original question.

It's actually quite easy. You design such that whenever you have to leave some data missing, you can do so by leaving a whole row missing. If a row isn't there, it isn't a row full of NULLs. It just plain isn't there.

So, in the case of "DateOfDeath", we have a table with two columns, namely, PersonId and DateOfDeath. PersonId references Id in the Persons table. If there is no DateOfDeath to be stored, we don't store the row. End of discussion.

If you do an OUTER JOIN between this and the Persons table, you'll get a NULL for the DateOfDeath wherever there was no row. And if you use this in a where clause, you'll get the usual perplexing behavior concerning 3-value logic. If you do an INNER JOIN, the rows for which there is no DateOfDeath will simply disappear from the join.

A design that permits every column to be NOT NULL enforced has been called sixth normal form.

Having said all that, I often allow NULLs in non critical columns. And I don't have a succinct way of telling you how I determine that a column is critical.