I’ve burned more than enough electrons here talking about dynamic languages and how they’re looming bigger and bigger in the future of software development. The buzz about dynamic languages on the JVM is building too; check this excellent piece by David Kearns, which makes me think I really need to check out Pnuts (also on Java.net) and Rhino.
I’ve also been banging the dynamic-language drums internally and have encountered no real resistance. But if we want Java programmers to use these things (and we do), then we have to give them tools. Around here, tools means NetBeans; Simon Phipps originally had the idea of a dynamic-language module, then he and I pestered people all over Sun and now here we are. Real credit goes to Tim Cramer and Trung Duc Tran from the NetBeans group, and to my manager Juan Carlos Soto for getting behind this and putting some money where our mouths were.
Should Sun Back One? · Here’s an interesting internal debate that’s going on. The name “Coyote” is appropriate because most of these dynamic languages live a lean, mean life on the fringes without much in the way of financial support or organizational infrastructure. If we at Sun wanted to get behind one and, you know, actually pay developers, just like we pay the lead Roller developer, we could really give it some momentum and a chance of becoming dominant on the JVM.
I think this might be a good idea. On the other hand, some other really smart people at Sun argue that we should keep our hands off the market and let the winning languages emerge from the ecosystem. I can see both sides of this question.
Join In! · The code that’s up there on Java.net now isn’t much more than early alpha; it installs on my Mac but a couple of minutes poking around and I’d already filed two bug reports. But we have funding to get quite a bit more work done, so it’s not going to be standing still.
I certainly plan to do some fiddling with Coyote, I want to be able to work on both Zeppelin and my Jython Servlets without switching back and forth between NetBeans and Emacs all the time. Feel free to join in, I think there’s a very high probability that this project is one with legs. We could particularly use help from people who actually use Groovy or Jython; or who want to add support for another of their favorite languages.