IRC channel logs

2024-05-28.log

back to list of logs

<tonyg>hey cwebber it just occurred to me to thank you (again?) for switching canonical-order of terms to encoded byte strings rather than preserves-order. it's a good simplification
<tonyg>it means you can get canonical terms without having to fiddle around with floating-point total ordering etc :-)
<dthompson> https://davexunit.itch.io/cirkoban
<dthompson>^ spritely's entry into the lisp game jam
<dthompson>sokoban meets wireworld! it's a lot of fun!
<dthompson>the first artifact of goblins + hoot, showing off some core goblins features.
<cwebber>\o/
<civodul>dthompson: d’oh! just played (in LibreWolf), it’s amazing!
<civodul>woow
<civodul>now i’m stuck in a room with a sleeping devil
<old>dthompson: was all of this written in hoot?
<old>I don't see where goblin come into this?
<dthompson>old: yes this was compiled by hoot!
<dthompson>goblins is used to model the game objects and their interactions
<dthompson>it's not all of goblins, though. far from it. just the very core part that we've been able to port to hoot so far.
<dthompson>but the game is showing actors and rollback working. two of the most important parts.
<dthompson>imagine what we can do once we have goblins with no compromises 😈
<cwebber>:D
<cwebber>and if you're looking for more to read
<cwebber>wingo has had some lovely blogposts about Hoot recently
<cwebber> https://wingolog.org/archives/2024/05/16/on-hoot-on-boot
<cwebber> https://wingolog.org/archives/2024/05/22/growing-a-bootie
<cwebber> https://wingolog.org/archives/2024/05/24/hoots-wasm-toolkit
<cwebber> https://wingolog.org/archives/2024/05/27/cps-in-hoot
<cwebber>:)
<dthompson>the boot, the boot, the boot of hoot
<dthompson>ACTION goes afk, singing
<cwebber>;)
<old> dthompson: I'm not sure to follow why goblin is useful for a single player game. I though it was for distributed stuff? Maybe I've misunderstood gobling entirely
<old>cwebber: Thanks, I like reading wingo
<ekaitz>cwebber: i need to share some thoughts about wingo 's articles: he is so smart and knows that much that is really hard to follow
<ekaitz>superinteresting, but also "omg, i'm probably losing the 70% of the content"
<ekaitz>does this only happen to me?
<theruran>ekaitz: the whippet GC post read like it was for other GC implementers, and I had some trouble following. The Wasm posts have been really great, but I've had to supplement by reading all the Wasm posts and extra material elsewhere like the Wasm spec.
<ekaitz>yeah...
<ekaitz>i watched the whippet talk also and I think I understood better
<ekaitz>but... omg! this is cool and hard!
<theruran>right :P
<theruran>"compilation is just a graph" "gc is just a graph" ohhhh
<old>ekaitz: I've only had difficulty reading the on-hoot-on-boot because I don't get the pun joke in english I suppose
<old>It is easier to follow the posts with VM and GC knowledge I think
<ekaitz>yeah sure
<cwebber>old: it's powerful for distributed programs for sure
<cwebber> https://dustycloud.org/blog/goblins-time-travel-micropreview/ but it's also a powerful transactional object model in general which supports time travel
<cwebber>And indeed, the undo feature in Cirkoban is implemented on top of that feature exactly
<cwebber>No special game logic needs to be added to support time travel, just wiring it up to the UI
<cwebber>Obviously the superpowers of Goblins is distributed programming, but it turns out to be really cool even for single process programs