okay. i think i have a new name for LRDL.
how about "nudl":
native (compiles to native programs)
unified (cpu + gpu)
data (-oriented)
language
but also "nu datalog"
but also a misspelling of the german word "nudel" for "noodle": you can noodle around (prototype) with this language, but it is also is how we derisively call visual dataflow programming, which it is also intended for as a backend.
it's not unsearchable, it deters techbros, it is fun to say, it is logo-suggestive.
typing is performed forward (propagation) as well as backwards (inference), which is why we can omit variable types entirely here.
instead of using control flow, i can now structure the program by observers and insert new handlers into a fitting section.
there's still Frood, as in "this frood really knows where their towel is", which would also take care of a catchphrase i suppose.
"nobros" is also untapped as a name for a programming language. it's great because it sounds vaguely greek, and you think it's something cool like "ouroboros". but then you learn "oh... literally 'no bros'. ok. 😔"
RE: https://mastodon.gamedev.place/@lritter/108243625888645023
four years ago i started this semantic R&D path which is now coming to a close; i thought "why don't i rename LRDL to the name i picked in the beginning", but, guess what, in the meantime several services named "frameloop" have popped up.
i still think the OO's pointing at each other and themselves in the logo is very clever so ... uh... if i come up with a novel name with two O's in it, i can reuse this.
names that put off techbros are especially welcome.
been thinking about a final name for LRDL when the first release is ready.
i'm going through a lot of names. presently i'm liking BraidsL (for Braids Language).
1) it's quite googleable
2) phonetically it sounds like the german word brezel (which sylvia came up with but is taken)
3) makes choice of logo easy for me
4) suggests coders need weaving skills
5) can reuse .b file extension
6) braids are pretty
but i don't think this is it.
okay! since a bunch of bugs have been fixed and evaluation flow got a few upgrades, i can now continue with my SDL3 demo, which presently just opens a window and runs a mainloop:
https://git.sr.ht/~duangle/tukan/tree/master/item/testing/test_sdl3.lrdl
i believe we should have enough to get a GPU interface and render a triangle. this would be next.
and try blocks work! i updated the prime decomposition example to use them; it now also tests more than one prime.
https://git.sr.ht/~duangle/tukan/tree/master/item/testing/various/prime_decomp.lrdl
as user i noticed i'd like to have some kind of if/elseif/else form. and for the compiler, it would also be helpful since it is computationally efficient.
here's one idea how this could work. all blocks form disjunctive cases, but the first case that completes excludes execution of all remaining cases.
LR