> Why do you think it's pointless to learn without knowing how it handles memory?
I think that a non trivial program will have to employ dynamic memory allocation, so if it is not obvious how to do that then it will not be possible to write a complex program. Learning a language that can handle only simple cases may have some uses, but you already have existing tools for that.
Got it. Red does manage memory for you, but no GC yet so can't do long-running procs without care. I've written small Red apps that used 800MB, loading a large dataset. Totally stable but, again, maybe wait for GC if you're going to grind a lot.
Red's complete toolchain, compiler, linker, etc. are bootstrapped entirely in Rebol2. Red is more capable than R2 (though alpha, with features pending), so it's safe to expect that it can handle more than simple cases. In fact, one of the reasons Red 1 was implemented in R2 was to prove that Red would be able to self host. It's possible most of the current toolchain would even work at 1.0, but the team has learned a lot while building this version, and has said on many occasions that Red 2 will only keep the best bits.
All that said, even if you never use it in production, it's a great language to play with and expand your thinking, because it's so different.
I'm waiting for a "Red in Action" book or the like. What I've seen in the Rebol examples hasn't helped me a ton with understanding the Parse dialect...etc.
> Why do you think it's pointless to learn without knowing how it handles memory?
I think that a non trivial program will have to employ dynamic memory allocation, so if it is not obvious how to do that then it will not be possible to write a complex program. Learning a language that can handle only simple cases may have some uses, but you already have existing tools for that.