No? How does that even follow? That a standard should codify common expectations insofar possible (without breaking rigor) is irrelevant to whether you "should" only use standardized languages or not.
He wasn't even able to fully test a theory because he relied on a feature that wasn't even available in older compilers in a completely unrelated system.
So, if you want to ignore these costs, that's fine, I'm not prescribing your actions for you, just sharing an opinion. Do you follow?