When Herb wrote his free lunch article, I was still a strong C++ believer and I thought C++ could solve the multicore problem. I considered Haskell a toy langua

When Herb wrote his free lunch article, I was still a strong C++ believer and I thought C++ could solve the multicore problem. I considered Haskell a toy language. Then I saw how quickly Haskell moved into the multicore arena with STM, Repa, Accelerate, and how C++ kept stumbling in the dark. After the C++11 fiasco I realized that C++ was a blind alley. Multicore revolution did to C++ what the Chicxulub asteroid did to dinosaurs.