Tuesday, February 05, 2008

Ruby.NET, We Hardly Knew Ye

It comes as not entirely surprising that the Wayne Kelly of the Ruby.NET project from down under has finally given up the ship. Microsoft had already grabbed the spotlight with IronRuby and the DLR. Despite some early gains, it seemed that development momentum had peaked, until Ruby.NET finally made its way up onto Google Code and a "real" open source project.

How, how will John Lam and team proceed without Ruby.NET around? I seem to recall a conversation I had with John when we were at RailsConf, and he was telling me since his M$ cannot look at open-source source code due to patent issues, they were unable to look at the MRI code in order to implement IronRuby. He said they WERE able to look at the Ruby.NET code due to some weird licensing thing. Does this mean they have now had their way with Ruby.NET, and they are done with it?

The DLR is a very cool concept, but it seems like is a lot more than many developers want, particularly polyglot ALT.NET folks who just want to add some capabilities most easily implemented in the Ruby-language, more than they want to switch over to pure Ruby.

Reading further in the mailing list thread, it seems like others in the Ruby.NET project are not quite ready to throw in the towel quite yet... but is it already too late for them?


John Lam said...

Actually we don't really look at other folks' source code unless we're actively porting their work. It's not an effective use of our time.

Instead, we look at source code when we're actually doing something with it, like porting Ola Bini's excellent JvYAML implementation to C#.

Ron Evans said...

I suppose I must have misunderstood you then... although I don't think so. Oh, well, now it must be my long-term memory that is failing.

Of course, having looked at MRI, I can see where that would be far from a productive use of your time anyhow.... the Rubinius test suite is really the new benchmark for everyone, isn't it?