« first day (942 days earlier)      last day (1698 days later) » 

dzaima
dzaima
7:44 AM
made the parser assume that anything it couldn't parse is actually just a strand of the parts it could, which increases the working dfn count to 207/207, but now with a known percentage of failure rate, which brings up the amount of found ⍥s to 33
and a gif of it in action (inside my android IDE which uses dzaima/APL for syntax highlighing because i like colors)
 
dzaima
dzaima
8:11 AM
because this is fun, 14 places where forks could be used:
https://tio.run/##tZXNbtpAEIDvfoq9YQKYmEqpZIme@gK9R0gLXmAFXrveBYJoLk1FCS1V06pSLq3SU6Neeki5cKmUR9kXoTNekxB@3EpVV7Y8s97xfDs/6wYVfSqXSyolbwmixx9II@xKzyI4UNTj9yNbz37q2TyPj8m1fvsLlFOyb@jZZ7TsBYJEcRjRFlU8FM7GKmtDr1lrDFKxqJJKNJaJFFDZ8R5a4RTw2XE4kESwE5UHPIROlHsaXEfCJvFZxITPhCKJBRV@skVnm@YPbG4KksgYoQRYjy9sPf6IkSrpt9Py7ighkOuRmPm9BjMg9SEJuOBBLyB92u2x/UArktT/XWLGF0hwCB@fZ20EfWNEFRlw1U7wyeFmZtZ816xGuyN7QepOzxaw2SOi37witnt4AHr@xQDerzjqfdYorjOgQ1hV8nmLg1PGW23FfAImmI/6ULESmKgwdjIyUINh@axp6kDQuhGimEXetplLXH3@1YYbiYqGzNPTs63YIJygAfBEMQ@44n1GmsIjPpeNb…
 
 
3 hours later…
ngn
ngn
10:45 AM
3
Q: Billiard balls collision

ngnGiven the 2-dimensional positions and velocities of a pair of billiard balls right before impact, calculate their velocities after a perfectly elastic collision. The balls are assumed to be ideal spheres (or equivalently: circles) with the same radius, same mass, uniform density, and no friction....

code-golf geometry linear-algebra physics
^should be fun in apl
 
dzaima
dzaima
11:07 AM
small performance test where the thunk idea could help (dfns has 23 occurences of A,1↓A but no A@1)
 
 
4 hours later…
Adám
Adám
2:47 PM
@dzaima Do you understand why the performance is like that?
 
dzaima
dzaima
2:57 PM
@Adám it needing to copy/move the array twice instead of once?
 
Adám
Adám
@dzaima Or once, instead of in-place change (depending on the array ref-count). However, construction of derived function through operators has a considerable overhead (we're working on that), so the smaller case is slower.
 
 
1 hour later…
Jonah
Jonah
4:20 PM
I am trying to come up with a J solution for "Church Substraction". The problem is this: In J, higher order functions are adverbs and conjunctions. It's easy enough to create "church numerals" using adverbs, eg, ^:2, ^:3, etc. The problem is that to create a church numeral "minus" operator you must create an adverb to operate on two of these. But the object of an adverb must be a noun or verb: You can't modify another adverb. So afaict it can't be done. Am I missing something?
 
 
1 hour later…
Adám
Adám
5:50 PM
@Jonah Couldn't you operate on gerunds?
 
Jonah
Jonah
@Adám I am actually looking into that atm :). So... not sure yet, but maybe
 
Jonah
Jonah
6:43 PM
@Adám I haven't been successful yet, other than some truly ugly ideas that aren't worth pursuing. It looks like the problem is that even if I allow myself to work with gerunds, there is no way to gerund-ify a church numeral, because you can't gerund-ify an adverb, and fundamentally that's what they are (in J). Any ideas you have are welcome.
mmm... looks like I might be able to use 5!:1 and 5!:0 for this...
 

« first day (942 days earlier)      last day (1698 days later) »