> but one can't expect the language to turn over, wholesale, each time new evidence of better linguistic behavior surfaces. That was what I thought when I read

> but one can't expect the language to turn over, wholesale, each time new evidence of better linguistic behavior surfaces. That was what I thought when I read the site; some of the evidence was not very broadly researched; what if that happens and suddenly the language has to significantly change? "The used-to-be-evidence-oriented programming language" ?