Slate 0.3 released

Slate 0.3 has been released. Slate is a language similar to Smalltalk with a prototype based object system and multi-method dispatch.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Thanks

However, I'm not sure how interested folks here will be in the system at this point. As of now it's mostly an exercise in how Smalltalk shouldn't really be tied to its original ideas, and also how much integration between some different ideas pays off.

Most of the work we're trying to do involves environments and libraries and so forth. For example, the ideas from A Functional Pattern System for Object-Oriented Design are implemented rather naturally and allow us to perform some heavily stream-based computations ala the fold/unfold algebraic organization of functional programming on just about any datatype or resource. That's just one initial example.

My current project involves bringing in the ideas of Erlang and other languages that make concurrency easy and natural. There's also a user interface architecture project that's been in development for a while to integrate Morphic and CLIM.

<obligatory typing reference>At some point I'm going to be investigating how to type live object environments (the other language I'm a fanatic about is Maude; I'm no stranger). If anyone is interested in the issues involved in this, I'd be happy to discuss the ideas and work together.</obligatory typing reference>

prototype based multi-method dispatch

I think one very interesting element is the blend of multi-method dispatch with a prototype based message passing object system. The only other prototype based system with multi-method dispatch is Cecil.

prototype based multi-method dispatch

That's definitely a major factor to our improvements. I particularly like my partner's slide presentation (it describes OOP as story-telling for a non-technical audience) and thesis on the concept and how we do it (it's moderately different from Cecil).

Why not?

However, I'm not sure how interested folks here will be in the system at this point. As of now it's mostly an exercise in how Smalltalk shouldn't really be tied to its original ideas, and also how much integration between some different ideas pays off.

Sounds great. Why shouldn't it interest us?