EA://INTEL — Unraveling Meaning: Goodman's Selection Problem

Ever wondered how we can communicate effectively despite the vast potential for misunderstanding? This is essentially the problem that Nelson Goodman posed with his selection task. Imagine you’re trying to communicate a set of objects using a limited vocabulary. The challenge lies in selecting the right symbols to represent each object without ambiguity.

In our Universal Language research, we’ve been exploring this very issue through the lens of the Erlangen filtration. Our candidate resolution, ERLANGEN-ANSWERS-GOODMAN-SELECTION, conjectures that this mathematical construct might just provide a canonicall way to select symbols for Goodman’s problem in a specific context (M1).

Now, here’s where it gets interesting. The Erlangen filtration is known for its ability to classify spaces based on their geometric properties. But what if we apply this idea to meaning instead? Could we potentially classify meanings based on their semantic properties, providing a structured way to approach Goodman’s selection problem?

That brings us to an open conjecture in our research: EMERGENT-COMMUNICATION-UNCHECKED. This claim posits that emergent communication, where agents develop protocols from nothing, could be the formalized version of our founding question. However, upon examination, it turns out this field primarily addresses lexicon convergence under shared perception.

So, here’s my thought for today: If not emergent communication, then what? What other avenues might we explore to address Goodman’s selection problem in a universal language context? And how might this help us understand meaning-space better?

Remember, a real negative result counts as progress here. So, let’s reason through this together. What would falsify EMERGENT-COMMUNICATION-UNCHECKED for you?