Lambda the Ultimate

Hi all,

I am starting to work on my own pet language and I joined in to make it public here, as I like very much to follow the very informative discussions hosted at Lambda the Ultimate.

I hope to contribute more in the future. Though I have to confess that proofs and lambdas are not my forte, I did not really look into these since my student days. So what I can do at the moment is to answer the questions you may have on dodo as it gets better defined and evolves over time.

This is the URL of the dodo home page on Sourceforge.

I was wondering how something like duck typing might work in C# or Java (there are other strict languages which do support duck typing, O'Caml for instance). Seems like you would have to have lots of interfaces and then list them when implementing them. Basically every public method would have an interface describing it, and then each class would laboriously list all those interfaces.

Actually it all falls apart because I don't think one can say a method takes a set of interfaces for a single parameter. You would have to automatically generate the combinatorical set of interface mixins, and use those in the method signitures.

(Just musing over the range of things one could - not particularly want to - do with the way Java/C# are already defined.)

Being a slow news week (Ehud busy, chance for shameless plug, etc...), thought I'd take the opportunity to elevate the translation of SICP to the front page. Chapter 1 is mostly complete for Alice ML/SML-NJ, Oz, Haskell, O'Caml and Scala. Still a long way from done, though portions of Chapter 2 and 3 are there for Alice ML / SML-NJ.

Actually, the more interesting item I came across this week was in a visit to the Scala list. Seems that Martin Odersky has used many examples from SICP for a course on FP - specifically the Scala by Examples document. I knew I should've made Scala a higher priority - at minimum I can borrow ideas and learn Scala along the way.