I'm fine with a discussion on naming conventions. I have one that I use and have shared with many. My objection is that the title of the article leads me to believe I'd be reading an article on "Best Practices for Database Design". While naming conventions are important, they have less to do with design re Codd & Date and more to do with preference. Using prefix/suffix mnemonics for objects, a discussion of underscores, abbreviations or singular vs plural object names has very little to do with normalization, data structure, when to/not to use triggers and a whole host of good design practices.
For example, take singular vs plural in table names. A table containing customer data could be named customer for a single instance of the data or customers as it represents a collection of customer records. If you go with the plural form then, for consistency sake, your CRUD procedures wrapping the table would give you Customers_ins (or ins_Customers), select, delete, update maybe a save (sometimes called upsert). Now the question comes; if I'm only affecting one Customer with any of those actions then why is the base name plural?
As you can see, you've departed from design and gotten into preference. Both are perfectly valid options with rational arguments on either side. Neither should be called a "best practice" because neither is. They are both "common practices".
If the author want's to bring both points of view forward on "common practices" for discussion then good. If, as you state, he's bringing this forward so that people who haven't given serious consideration to this topic may begin, then that's good but, don't tell half the story and and say it's a "best practice".