I think the smaller example Matthew Flatt used in a talk about the expander is a better example. From the old expander's conception (many many years ago) only

I think the smaller example Matthew Flatt used in a talk about the expander is a better example. From the old expander's conception (many many years ago) only 2 people commited any changes and one of those only committed smaller instrumentation changes. Since the rewrite (that was about 1.5 to 2x slower) less than a year ago 6 people had made commits. What one could hope for would be some chez scheme interoperability for the really speed demanding parts.