@vidak@social.solarpunk.au @vidak
My loosely interpreted lisp-moo-reality is coming together (I guess com-@dump ~= make-load-form is the one missing very fundamental primitive command from my perspective).
I'm thinking about porting your uwu.lisp virtual pets https://git.sr.ht/~vidak/uwu.lisp
into my moonclimb https://codeberg.org/tfw/moonclimb
first to develop porting lisp into a moo-reality-lisp-image,
and secondly as having good-old-fashioned-ai denizens in the moo-reality-lisp-image.

Thoughts?

@vidak
My passage of time is very loosely modelled on VHDL - actions either happen on the next clock cycle (ie they're inserted into a ring buffer of what's continuing cycle to cycle at the moment), but a priority queue also lets you evaluate sequences of actions right away step by step, which I'm identifying with advancing delta-times per action in the vhdl process definition.

@vidak this is different and much more coarse-grained to MOO where user accounts are basically allocated a number of flops per cycle.

@screwtape @vidak
Just thinking out loud, I wonder if you actually need more than two slots in your priority queue: do-now and do-on-next-clock, that sort of thing.

BTW I can't remember if it was Scheme (probably not) or more probably MDL of Zork fame, but some Lisp family language did count virtual clock cycles to task switch without relying on the external operating system.

There have been too many languages in that family, I've lost track.

@dougmerritt
I eventually didn't submit a research masters thesis on formally verified VHDL program generation using lisp quite some years ago (that wasn't how *they* titled the project, to be fair). From my understanding my work and findings were diametrically opposed to what the sponsor and supervisors wanted found ;p sez me!
@vidak
Edit: I clearly owe a lot to that time, and I have regreted that there didn't seem to be a way for me to express that.

@dougmerritt
But actually on topic, yeah, that's why I was trying to refer the problem of what-I-need out to the VHDL standard description of simulations in terms of delta time steps.

I guess the most pure direction in terms of moo would be to get really into *server-hackers (https://github.com/wrog/lambdamoo/branches) and how MOOing shares computation between players. But I would actually like to get much less deep into time sharing, not more deep.

Which is why ^ I was voluntarily command/cycle-ing.
@vidak

@mdhughes did you or one of the other muddlers have a strong opinion on MDL's processor time-sharing? Though I kinda want to only deal with the issue as and when it becomes important (deferred commitment!).

@dougmerritt @vidak