"Availability of tutorials" vs. "searches for tutorials" doesn't affect the point. It's still selecting a subset of programmers. While it's somewhat useful as

"Availability of tutorials" vs. "searches for tutorials" doesn't affect the point. It's still selecting a subset of programmers. While it's somewhat useful as a leading indicator of popularity over the next couple of years, it's less so w.r.t. either the present or the further future. I'm not saying it has no value at all, but it's not clearly a substitute - let alone a superior substitute - for surveys that measure other things.