Listening to Steven Wolfram and I just had a thought. Can we use an LLM to generate an optimal language?

That would be neat, but language adapts quickly to the context in which it's used, so people who are doing different things would create their own shorthand, context scoping, jargon, etc, and the language would keep diverging.

Maybe language is something that shouldn't be so strictly defined, and maybe that's okay.

Me: Can I get an optimal language?

Mom: We have an optimal language at home

The language at home:

Me: Can I get an optimal language?

Mom: We have an optimal language at home

The language at home: https://en.wikipedia.org/wiki/Ithkuil

VC & Engineer | FOGC | Researching, Investing & Building — robinnordnes.com & elixircapital.xyz

Based on the talk I heard from Steven Wolfram, I’m not sure that’s how I would describe language, but maybe there is some truth to that. I’m also not sure why that would preclude a designed language (similar to Korean, but “better”)

Human language is based on a shared consensus of what a word mean where the majority of individuals would know that particular word. 

An LLM cannot construct a new language that everyone understands but it can help you construct your sentences more accurately.

🌷💀⏳

teleyinex.eth ▫️

teleyinex.lens ▫️

teleyine.x ▫️

daniellombrana.es  ▫️

@astaralabs ▫️

@obeygiant ▫️

@scifabric ▫️

dev

Make human to human communication faster, less error prone to misunderstanding, and higher bandwidth per word

Language change is constant - how could you keep it optimal?

Personally I think something like Spanish (simple vowels, phonetic, simple orthography, etc) without grammatical gender would be pretty close to optimal, as far as that's possible.