Hi Joseph,
Gilad's comments provide links and suggestions to your questions, and also a path through Smalltalk. For what it is worth, let me summarize my personal reasons for playing with Newspeak. From your message, I realize your motivation is different, but I thought sharing this could provide another perspective.
So first of all, I feel a Smalltalk type language is an excellent choice for it's consistent, readable and writable syntax (even if it had no other pluses) .
Being a practicing software person, not a theoretical CS person, over the years I have tortured my head over questions, and sometimes was bitten by (not alone in either :) ) related problems that crop up all over again in our industry. To select just a few of many such questions (simplified wording though):
- Why is (mutable) global state bad?
- Concurrent computation: Are resource-sharing approaches such as threads problematic?
- Reflection and security tradeoffs?
- Can a language that is late bound also have some level of typing mechanism?
- Multiple inheritance issues and are there solutions?
- Are OO and functional languages compatible?
(etc etc .. Many of these problems are actually related on some level, which makes thinking about them harder. They all have program security, writability, and maintainability consequenses).
The reason that drew me to Newspeak many years ago (although only now I am starting to give it more time) is that Gilad et. al. address, directly or indirectly, many of these problems and questions in Newspeak. Newspeak applies solutions that are based on modern research ("modern" may be 40+ years old, but other platforms ignore even that old research), to a wide range of these problems, which makes it interesting and unique (from my personal non theoretical CS understanding) - "principled" as they say.
Well, anyway. Sorry to make this so long, and probably not too relevant :)
Good luck with your quest and experiments.
Milan