For the author of this review to say this:
"Carey goes on to state that the human genome contains 20,000 protein coding genes, so I guess that makes 980,000 of the junk variety"
implies that Carey has actually done a poor job of explaining things. The 98% non-protein coding DNA is - by definition - non-gene DNA so there are not "980,000 [genes] of the junk variety".
The book is about junk DNA so surely the difference between coding and non-coding DNA should have been made at the start. That the Reg journo missed this implies that the book is not actually all that well-written....