Lightnews — Scholar-powered news

Piston Developers 🍿🥤

@pistondeveloper.bsky.social

6K followers 3.3K following 9.2K posts

A modular game engine written in Rust https://piston.rs Research branch: https://advancedresearch.github.io Discord (Piston): https://discord.gg/TkDnS9x Discord (AdvancedResearch): https://discord.gg/JkrhJJRBR2

piston.rs

Posts Media Videos Starter Packs

Pinned

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · Nov 10

#StandWithUkraine 🇺🇦 (since Feb 27, 2022)

5 25 190

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1h

Nature, you are very weird!

#physics

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1h

Maybe this is a way to interpret Noether's theorem, that every continuous symmetry of the action of a physical system with conservative forces has a corresponding conservation law.

If symmetries might change over Aeons, then perhaps we're living in a fractal of time?

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 2h

All this is kind of funny, because why heat? Well, something has to carry away the information that gets lost from other processes. So, just by assuming information is preserved, we get a lot of logical implications. Yet, we can't identify these physical phenomena. We have to do experiments.

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 2h

A symmetry in physical laws can not consume or emit energy if information is preserved, since nothing can interact with it. That's per definition. All computational symmetry in systems with serial inputs must emit heat (quantum decoherence).

It opens up an interpretation of non-interactivity.

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 2h

As far as we know, nothing can survive from one Aeon to another in Conformal Cyclic Cosmology. It seems also unlikely that most life forms survive between other cyclic cosmological models. However, is it possible that universes "evolve" by mutating in symmetries? Because a symmetry is not "alive".

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 2h

Nothing can violate the symmetries of physical laws in IBs because it would add or remove information.

In that sense, this framework can be used to reason both about "alive" systems (those that reproduce and need useful energy) and the symmetries in physical laws themselves.

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 2h

There might be some proof systems with parallel inputs that maintain quantum coherence instead of requiring energy.

Actually, the symmetries in physical laws are often algebraic. This is kind of like a negative physical interaction. Symmetries are constraints on physical behavior.

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 2h

All A-Sym systems (Algebraic Symmetry) have C-Sym. So, any proof system with e.g. addition and multiplication is "alive" in some sense that it needs energy to maintain functionality over time.

The notion of A-Sym and C-Sym here is for symbolic distinction (common sense symmetry or commutativity).

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 2h

There is a property of every PI (Physical Information) system that is an IB (Information Boundary) has IO (Input and Output system), that when is has C-Sym (Computational Symmetry), it either maintains Coh (quantum Coherence) or produces DeCoh (quantum decoherence or heat).

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 2h

New alphabetic list for the Path Semantics project:

Alphabetic List of Physical Information Systems

Describes a list of concepts for physical systems, from the perspective of information. Among the most central laws is `C-Sym => Coh | DeCoh`. github.com/advancedrese...

1 1

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 23h

Yeah. Disaster is an understatement.

Reposted by Piston Developers 🍿🥤

#TuckFrump @realtuckfrumper.bsky.social · 23h

Letitia James: 'I will not bow. I will not break'

thehill.com

Letitia James: ‘I will not bow. I will not break’

twp.ai

1 11 77

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

We're now 55.5% on the way to finding minimum primbix of value 20. There are 26 CPU cores working on this search 24/7.

We know this number is less or equal to 294 698 527 560 839.

#primbix #research #advancedresearch

Reposted by Piston Developers 🍿🥤

David Andersen @poetriesbound.bsky.social · 1d

For @pistondeveloper.bsky.social,

This is a fair adaptation of #Oscar_Wilde's masterpiece about vanity.

I hope you enjoy it.

youtu.be/YbFf8dYmxE0?...

The Picture of Dorian Gray - John Gielgud - Peter Firth - Jeremy Brett - 1976 - Remastered 4K

YouTube video by Shakespeare Network

youtu.be

1 1

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

It's part of the research branch. At best, I can say it might be useful for game design. However, I wouldn't say that learning it makes you more effective at game design, or something.

Probably unrelated! 😜

Reposted by Piston Developers 🍿🥤

Dre (they/them) @dragontongue.bsky.social · 1d

Is this related to the game engine project?

1 1 1

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

Maybe that's a kind of weirdness that humans have, that makes us unique in the sense that our way of existing and using symbols is not like what other beings out there who might be intelligent do. It could be something particular to humans.

So, we're weird? I'm OK with it!

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

Some versions of AGI might need deep intuition about language biases to know in detail how humans think.

If we can't build it, then trying to make AGI understand us might be a hopeless project. When you see how these properties affects soundness, it gets very weird and hard to solve.

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

Maybe some versions of AGI don't need any of this stuff. It might function perfectly fine while just ignoring these particular concerns because it's not human and doesn't do the kind of mistakes we do. It doesn't have to organize information the same way our brains do.

Some versions might need it.

1 1

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

We can study language biases. However, there is no need to "fix" language biases in general. We just have to understand limitations and know how to apply language biases properly, so we don't make mistakes.

If we can do that, then I think it's "good enough". But, can AGI do that? I don't know.

1 1

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

That's a common mistake people do in philosophy, that they think what they're working on will be useful in every possible context and make sense in every situation. I think that's a bad idea.

Language are limited for a reason. They have language biases. Otherwise, they wouldn't be effective tools.

1 1

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

Not having to worry about ontology saves us a lot of philosophical headache. We know that Path Semantics is limited in this particular way. So, since we're real human beings, of course we have other perspectives due to ontology.

It would be kind of choking if it covered everything.

1 1

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

The way I see it, the core axiom is the only thing that Path Semanticists study. It's about breaking everything down into something that can be understood from that perspective. That's the way we define "use of symbols".

Since this is just an axiom, there isn't any ontology. That's a relief.

1 1

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

I just don't think the perspectives I have as Path Semanticist are the same perspectives I have as a person. I can tell the difference, because I gotten used to. The Path Semanticist in me is kind of an avatar, about use of symbols from the perspective of the core axiom of Path Semantics.

1 1

Piston Developers 🍿🥤 @pistondeveloper.bsky.social · 1d

Path Semantics is a field where people study how humans use symbols. That's it.

No ontology. Theorem proving isn't the important part. Just use of symbols.

Makes it clear what the field is about. Separates itself from other perspectives. I have other perspectives too, personally.

1 1