Open ForNeVeR opened 10 years ago
Well, I personally blame naive backend implementation. But yes, staticity, especially trying to infer function signature from it's usage did played a major part, so a major rethinking and redesign is in order. Though, I'm in favor of scrapping the source entirely and starting from scratch.
Though, I'm in favor of scrapping the source entirely and starting from scratch.
Agreed.
Currently I'm reading about the DLR and what it can offer for us, and after that I'll try to summarize and make a plan for our next steps.
So, looks like everyone is in favor. I'll go purge some source then.
We should integrate the DLR support into our project.
Our attempt to make a statically typed LISP was doomed to failure from the very beginning.
Our LISP dialect should be a dynamically typed but should try to benefit from the compile-time (and runtime) type knowledge when possible (I hope the DLR will help us with that).