He must be using chess books written in the 19th century. I'd challenge the guy to provide some specific examples of errors in the opening manuals such as ECO, NCO, etc.
Individual game analysis is another matter. Anything analyzed and published before the computer is a good target for double checking.
I was speaking with a strong chess expert who teaches scholastic chess yesterday who said he had his students go through opening books with the best programs like Rybka. He said the computer proved many of the books so far wrong it wasn't funny. Are there articles on this?
ECO had the line 1e4 e5 2f4 d5 3exd c6 4dxc6 Nc6 -/+ Computers doubt this assessment.