I investigated RTalk with interest and exchanged a few emails with Mark Roos. RTalk is nice but has a different goal.

Redline wants to support you and your existing tool chains, editors and environments that exist around the JVM. We want the best of Smalltalk in that environment (Redline), not a Smalltalk environment (RTalk) on the JVM. A subtle but important difference.

- James.

I'm not sure I see how Smalltalk would be very useful without a complete environment? Are you just implementing a Smalltalk(syntax) to bytecode/jar compiler? No support for dynamic execution/compiling code in any text-form? No adding behaviour to gui-object? No highly iterative development?

We are doing almost all of that.

You can already do dynamic execution and compilation. We just need hooks into ides like eclipse etc.

We have no near term plans to do anything GUI wise initially. But we would very much welcome someone taking that on.

The iterative development will be there. We support 'thisContext' which makes supplying a smalltalk style debugger with resumable exceptions, debugger driven development etc possible. It is just a matter of hooking into existing ones and working out any kinks.

We are not trying to be able to boot a full pharo environment while running on the jvm. version 1 roadmap is for integration with tools that already exist in the jvm ecosystem. if others want to run ahead and work on other environments ( including brand new ones ), we will lend a hand where we can.

Hopefully this answers your questions, if not feel free to join the mailing list and ask away in terms of questions. Or email me personally if you want, [email protected] And even if this does answer your questions, feel free to fire away with new ones on the mailing list or to me.

Apparently rtalk will be/is under a MIT licence too -- but so far the only code is for the jvm-side, buried at:

http://wiki.jvmlangsummit.com/Building_a_Dynamic_Language_on...