Ways to Make Sand (2020) [video] | Hacker News Reader

Ways to Make Sand (2020) [video]

youtube.com

51 pointssurprisetalk13 comments

This is about digitally simulating sand physics, for a video game or something. I thought it would be about manufacturing/processing actual physical sand.

Probably should add "in SpaceTode" to the title.

Also:

Top 9 ways to make BIG Sand

https://www.youtube.com/watch?v=9mbs0sx3z2A

And:

Tourism

https://www.youtube.com/watch?v=nCR9zMU2Q_M

More than mere magical pink sand (the bastard stepchild of gray goo and pink slime), Tourism is a sweeping tour of many different vastly different approaches to visual programming, games, engines, cellular automata, user interfaces, and simulations! Including the Moveable Feast Machine (both JavaScript and C++) and the "SPLAT" language and T2 Tile hardware.

Hello thank you for sharing my 9 ways of making sand. If you have any more ideas please let me know, I'm trying to make a sequel to this where I do 99 ways.

How about an MMPORG where each player is a grain of sand, and gets to first act out by hand and then gradually automate their behavior like Factorio, like programmable dirt. Call it "Ground Up Programming".

No problem! I wanted to share sooner, but I didn't want to stress you out with an influx of traffic right before your SPLASH talk :)

No ideas for now! Looking forward to seeing a video of that aroost demo you showed me at Strange Loop (this is Taylor btw hehe)

Keep up the incredible work!

Todepond is a treasure. Everything they touch is brilliant. This is one of my favorites: https://www.youtube.com/watch?v=Q4OIcwt8vcE

TodePond is such an inspired and inspirational genius! Screens in Screens is exactly the one I was going to recommend too, that will recursively suck you into watching all the rest.

Screens in Screens (ScreenPond) source code:

https://github.com/TodePond/ScreenPond

TodePond's videos:

https://www.youtube.com/@TodePond/videos

Including the latest stuff on Patreon:

https://www.patreon.com/todepond

...Where I heard about her recent talk about CellPond at ACM SIGPLAN in Lisbon:

https://www.youtube.com/watch?v=cBYudbaqHAk&t=6703s

It's a modern ground-up reimagination of visual computing, as foundational as Scott Kim's 1988 PhD dissertation, "Viewpoint", but deeper:

https://www.youtube.com/watch?v=9G0r7jL3xl8

>Demo and explanation of Viewpoint, a computer system that imagines how computers might be different had they been designed by visual thinkers instead of mathematicians. Caution: this is basic research, not a proposal for a practical piece of software. Part of my PhD Dissertation at Stanford University in 1988.

Toward a computer for Visual Thinkers:

https://web.archive.org/web/20230207211417/https://scottkim....

Todepond talked about how she was introduced to visual programming as a child by Stagecast Creator (aka KidSim aka Cocoa):

https://en.wikipedia.org/wiki/Stagecast_Creator

>Stagecast Creator is a visual programming language intended for use in teaching programming to children. It is based on the programming by demonstration concept, where rules are created by giving examples of what actions should take place in a given situation. It can be used to construct simulations, animations and games, which run under Java on any suitable platform.[1]

>History

>The software known as Creator originally started as a project by Allen Cypher and David Canfield Smith in Apple's Advanced Technology Group (ATG) known as KidSim. It was intended to allow kids to construct their own simulations, reducing the programming task to something that anyone could handle. Programming in Creator uses graphical rewrite rules augmented with non-graphical tests and actions.

>In 1994, Kurt Schmucker became the project manager, and under him, the project was renamed Cocoa, and expanded to include a Netscape plug-in. It was also repositioned as "Internet Authoring for Kids", as the Internet was becoming increasingly accessible. The project was officially announced on May 13, 1996. [...]

https://news.ycombinator.com/item?id=33700318

>That's something that Alexander Repenning's "AgentSheets" supported (among other stuff): you could define cellular automata rules by before-and-after examples, wildcards and variables, and attach additional conditions and actions with a visual programming language.

>AgentSheets and other cool systems are described in this classic paper: “A Taxonomy of Simulation Software: A work in progress” from Learning Technology Review by Kurt Schmucker at Apple. It covered many of my favorite systems.

http://donhopkins.com/home/documents/taxonomy.pdf

>Chaim Gingold wrote a comprehensive "Gadget Background Survey" at HARC, which includes AgentSheets, Alan Kay's favorites: Rockey’s Boots and Robot Odyssey, and Chaim's amazing SimCity Reverse Diagrams and lots of great stuff I’d never seen before:

http://chaim.io/download/Gingold%20(2017)%20Gadget%20(1)%20S...

>Chaim Gingold has analyzed the SimCity (classic) code and visually documented how it works, in his beautiful "SimCity Reverse Diagrams": [...]

https://lively-web.org/users/Dan/uploads/SimCityReverseDiagr...

>Another great visual programming language for kids that supported defining cellular automata rules by example and visual programming:

>KidSim (later Cocoa, then Stagecast Creator) Smith, David C., Allen Cypher, and James Spohrer (1994) In KidSim graphical simulations are created via graphical rewrite rules, which also enables a kind of programming by demonstration. The creators argue that most people can use editor GUIs (e.g. paint programs), and can give directions, but cannot program. Their solution is to “get rid of the programming language” in favor of a philosophy grounded in GUI design:

>• Visibility. Relevant information is visible; causality is clear; modelessness. • Copy and modify, not make from scratch. • See and point, not remember and type. • Concrete, not abstract. • Familiar conceptual model. (“minimum translation distance”).

>They choose a symbolic simulation microworld as a domain because it leads to knowing, ownership, and motivation. All objects are agents which have appearances, properties (name value pairs), and rules.

>Programming by demonstration extends to using a calculator and dragging properties around to define conditionals. One of the creators of KidSim, David Smith, was also the creator of another graphical programming environment: Pygmalion.

>Smith, David C., Allen Cypher, and James Spohrer (1994)

>(Then I ran across TodePond's Spellular Automata video and realized I was preaching to the choir! TodePond wrote: "and stagecast creator is a big inspiration to me! I name-dropped it in a demo I did this week :D")

>Wow I did not realize I was evangelizing to the choir! This video by TodePond, is exactly what I was talking about, just much more beautiful than I'd imagined possible:

Spellular Automata:

https://www.youtube.com/watch?v=xvlsJ3FqNYU

Finally to know there is somebody who appreciates Dave Ackley's amazing work with the Moveable Feast Machine as much as I do!

https://news.ycombinator.com/item?id=24157104

https://news.ycombinator.com/item?id=15560845

https://www.youtube.com/watch?v=XkSXERxucPc

Todepond's work with Max Bittker on SandSpiel Studio is also thoroughly mind blowing:

https://news.ycombinator.com/item?id=34561910

https://studio.sandspiel.club/

https://www.youtube.com/watch?v=ifyYITDq1oo

https://www.youtube.com/watch?v=PGTsy79wx4U

I had the pleasure of encountering them in-person at Strange Loop 2023! They gave us an impromptu demo of some of their experiments in the hallway after the last talks had ended and we were cheering wildly with everything they showed us. It was a delight!

this is super delightful; similar to lattice gas automata (lgas)

spacetode: https://github.com/TodePond/SpaceTode (mit license)

lgas: https://youtu.be/Vllt9QLBBs8?t=8m55s https://youtu.be/0xlRPnQGSXg

unfortunately, lga simulations, like sparse rotating pointclouds, don't compress well on video

the graviton part was amazing. I wonder how it would look if you use spacetime rather than graviton.

I think, to make that work you need planet and with mass and a flat space time. And the add the particle of silica etc.. so it falls properly

No comments or ideas, but a question triggered by watching this video:

Two prisoners in soundproof cells are communicating through a shared underground puddle that has a small opening in each prisoner's cell. They communicate with Morse code, by repeatedly poking a stick into the opening that fully fills it. Water being incompressible, it causes the water level on the other side to rise slightly.

Please ignore the fact that sound travels quite well through water. Perhaps the prisoners are deaf.

The question is, what is the maximum bandwidth of this communication channel? Ignore any physical limitations on the prisoner's part with respect to the speed of poking their sticks in, and assume they both have a great sense of timing and can only either poke or not poke; they cannot do small vs large pokes.

Let's say the little underground water tunnel is exactly 1 meter in length, and the perfectly circular openings are exactly 1 inch in diameter, just to piss off both metric and imperial aficionados.

It's my understanding that the latency is approximately equivalent to the speed of sound as that is a function of stiffness and density.

There's some detail missing still. What's the temperature of the water? How long is the puddle (or rather what's the total volume)? What's the surrounding material?

I'd guess the limit is right depending on how much energy you can put into the water before it boils. At least several hundred megabits per second, I reckon -- I'm probably under by one or more order of magnitudes.

Reply on news.ycombinator.com