02:04 <lispmacs> question: in multi-tasking Forth environment, my understanding is that each thread has it's own data and return stack, which makes sense. But what about `allot'ed memory? don't things get mixed up when certain tasks want to roll back the pointer?

08:38 <veltas> lispmacs: I think usually they have their own dictionaries too

08:38 <veltas> There is a system dictionary and user dictionaries, or something like that

08:38 <veltas> Download PolyForth, read manual

23:11 <veltas> Chuck Moore is probably right about comments

23:11 <veltas> Good code does speak for itself

23:13 <neuro_sys> Yeah, more so in high level languages. But when writing assembly code, I comment out groups of instructions that mean something distinct (which can be routines or macros).

23:14 <neuro_sys> Although if they're repeated often and are idiomatic, then I don't. Like in Z80: ld a, (hl) / inc hl / ld h, (hl) / ld l, a

23:16 <veltas> Sparse comments are helpful in assembly, heavy comments are hard to maintain and not useful documentation in assembly or anywhere

23:38 <Zarutian_HTC> depends on what the comments are used for

23:38 <Zarutian_HTC> if they document the why the sure

23:39 <Zarutian_HTC> but if they just restate the same as the code they are worse than useless

23:44 <neuro_sys> Then they're no more than tautology

23:44 <neuro_sys> "Where Forth programmers come from is as mysterious and as much a subject for debate as was in the previous century the source of Nile."

23:48 <Zarutian_HTC> but usually I use comments to stick in url links to reference material for the particular problem the code is written to solve