First off, you make unwarranted claims, like that your imaginary implementation would be fast. You don't know that and you can't know that. You don't seem to realize that the kinds of language features that enable metaprogramming, also limit the maximum theoretical speed of your implementation.
Secondly, you've failed to distinguish this language from Ruby. You've turned map() into a digraph operator that feeds sideways. Okay? Why do I care?
Spewing out some pseudocode doesn't make people care what you have to say, you need to better explain what makes it different from other languages, not what makes it similar.
I don't care if it's similar to ruby. I'll just use ruby if it's that similar.
What you are doing is somewhat akin to sketching or doodling. It can be an important step towards creating something tangible, but as it has no deadlines and imposes no constraints, it has limited utility. Design is all about making tough choices about what to include, what to leave out, and why.
My suggestion--and I am no expert on writing programming languages or creating anything--is to borrow something from Paul Graham and recast your project as an attempt to answer a question. For example:
Probe is an attempt to answer the question "Does an operator syntax for mapping and folding functions make code easier to read?"
Thanks, appreciate the redirect to reframe as a question.
This is the type of question many developers should periodocially ask themselves, "what do I want in a language?" Our tools deeply affect how we think through a problem.
Guilty as charged. Much of my blog is mental fiddling ;)
I had never before yesterday thought about program language design from a top level, how do I want it to look/feel in the application?
Maybe it is just a flavor or Ruby, but there are some language level features Ruby is missing: implicit loops, smarter interface pattern fitting,smart multidimensional set handling, optimal numeric processing.
Some of my work demands fast turn arounds to think through and iterate on the data processing & algorithms
processing
First off, you make unwarranted claims, like that your imaginary implementation would be fast. You don't know that and you can't know that. You don't seem to realize that the kinds of language features that enable metaprogramming, also limit the maximum theoretical speed of your implementation.
Secondly, you've failed to distinguish this language from Ruby. You've turned map() into a digraph operator that feeds sideways. Okay? Why do I care?
Spewing out some pseudocode doesn't make people care what you have to say, you need to better explain what makes it different from other languages, not what makes it similar.
I don't care if it's similar to ruby. I'll just use ruby if it's that similar.
I hope your next post is more substantial.