I’ve never heard J.Lam credit that fact. Maybe, hopefully it’s just that I havent listened closely enough.

I’d definitly like to see competing implementations of Ruby on the .net platform.

It’s a more than worthy exercise for an academic institution. Give those DLR boys a run for their (fat) money!

I believe that ultimately there is no need for two different implementations of Ruby on .NET

I can see the desire to have a single reference implementation. I could even be persuaded that the idea of a single common production-quality implementation has major benefits for reusability and interoperability. But, especially having worked at and with Microsoft using compilers of varying levels of quality and conformance, I think it would be a very good thing to have alternate implementations that can be used as development aids. Much as Borland C++ served as a backup compiler for me for many years while Microsoft was still working on a sensible C++ implementation (and having shipped at least one program using BC++ after publicly stating that we “used Microsoft tools in production” due to language-implementation and RTL defects), I would feel much more comfortable as a Ruby developer on .NET if K new that I could use another tool to help me diagnose troubles I was having with IronRuby on .NET. Note that I am not saying that I have those troubles today, just as I would have said that my team was by and large happy with Microsoft’s C++ implementation up until two weeks before we shipped on Borland. But having an alternative compiler/interpreter/adjudicator of the language you are using on the platform you are targeting can and more than once has made the difference between survival and failure for startups and other delicate projects. I would look to someone carrying on, or if necessary forking, Ruby.NET for just that purpose. Maybe it’s not and never plausibly will be “the best” Ruby implementation on .NET; it doesn’t have to be. It’s just there as a crutch for developers, and as something to keep the ongoing IronRuby project “honest”. Performing that role, even if the resulting code was never as fast or small or as the IronRuby equivalent, would still be a service worthy of lyric and epic praise.

PS: I really hope this doesnt hurt Ruby. I use Ruby almost exclusively, so Ruby-on-Net (working, maintained) definitely would definitely be awesome.

Cheers, Wayne