Before you try to sell me on some "best practice," I want you to know that I'll test it against Emery's Ironclad Test of Best Practices:

For something to be a best practice, it has to be practiced, and it has to be best.

If there ain't nobody practicing it, it ain't a best practice.

If it ain't best (and how would you know such a thing?), it ain't a best practice.

This is a simple enough test. How many "best practices" pass it?