LawrenceC comments on Introducing the WeirdML Benchmark