Protocols are Problem Solving Agents

[ergod]

Hey everyone,

Here is an essay I finished drafting recently. You Never Truly Solve a Problem - Thoughfolio (symplectic.link)

My background is that, since I have heard about protocols from the early days of SoP, I kept wondering, how can these ideas help me in a concrete sense? Amid all of that initiate hype, I couldn’t hear myself thinking. So, I took a pause, and took time to grasp the concept under my own terms.

Recently I have had a bunch of experiences (namely, a hackathon) that put me on the path of many interesting concepts, like Autonomous Worlds.

Initially, I intended to write a piece on the topic of “problem solving” and get more precise about it, speaking about problem solving within a world to be specific and think about what that entails. And I found myself describing an artefact of problem solving in open worlds that is maps to protocols.

Another conclusion, I drew in this writing, is that protocols seem more like problem solving agents than solutions.

On another note (and based on that conclusion), I am experimenting with non-linear interactive storytelling as a form of research, implementing the ideas I described in another essay, Fragmented Exploratory Thinking as a Non-linear Interactive Story - Thoughfolio (symplectic.link). The story elaborates on the previous point in a sci-fi setup.

Now, I know that it is too late to apply for the SoP 2024 program, but I think that, the output of my experimentation is well aligned with it, as an artifact for protocol-pilling.

All in all, any feedback on my writings would be appreciated.

[Crul]

Problems are descriptions which forward a gap in our comprehension.

I love this definition. Is there any inspiration or reference?
I’ve read the term “gap” a lot in Žižek’s posts. I think it’s a mix of Lacan (gap/lack) and Hegel (negativity) concepts… but I’ve never fully understood it.

A solution answers a problem via either of the two effects. The first, breaks down and rearranges our understanding to accommodate to the new piece of information. The second, changes the way we hold the puzzle piece leading us to reformulating the problem under more suiting terms.

I’m not sure I understand and/or agree. Is this a distincting between reductionism/mechaniscism (“break down”) and non-reductionistic approachs (“changing the way we hold the puzzle”)?

I see the “dynamic puzzle” as a graph and it feels like there are more than 2 ways of modifying a graph.

The problem solving agent can be part of the world as a diegetic entity or he can be an external observer

I also really like this distinction.
It’s specially interesting when you thing about introspection… it brings me closer to the idea of “the subject as a gap (or lack)”.

The last second part of the post has also make me think a lot…“Protocols as coordination tools to avoid conflicting strategies in multi-agent open systems” is an interesting framework.

Protocol[s] are problem solving agents.

I am not convinced that, in this framework, it makes more sense to see protocols as agents than as tools (of human agents). Am I missing something?

Thanks!

Note: The link in Models points to a broken page:
https:// symplectic.link/Models+are+Representations+of+Freedom
And google brings nothing searching for “Models are Representations of Freedom”

[ergod]

Regarding,

@Crul

Problems are descriptions which forward a gap in our comprehension.

I love this definition. Is there any inspiration or reference?
I’ve read the term “gap” a lot in Žižek’s posts. I think it’s a mix of Lacan (gap/lack) and Hegel (negativity) concepts… but I’ve never fully understood it.

Honestly, there is no reference behind it or inspiration beside my own cooking. Although, I am not claiming its mine in any way, I am sure someone had this idea before.

@Crul

A solution answers a problem via either of the two effects. The first, breaks down and rearranges our understanding to accommodate to the new piece of information. The second, changes the way we hold the puzzle piece leading us to reformulating the problem under more suiting terms.

I’m not sure I understand and/or agree. Is this a distincting between reductionism/mechaniscism (“break down”) and non-reductionistic approachs (“changing the way we hold the puzzle”)?

I see the “dynamic puzzle” as a graph and it feels like there are more than 2 ways of modifying a graph.

Well both are about a gap in our comprehension, if I hold a puzzle piece in a way and try to fit, the reason I did is because I think that is most likely the right way to hold, it reflects my understanding, now if I don’t try to hold it in any other way, then it reflects a gap in my comprehension of the situation. I just couldn’t see it otherwise. I think the case for reductionism is obvious, I just started putting things the wrong way, so I only detect that when it is too late and I have to breakdown at that point.

To be honest, when I wrote this I also felt there are more than two ways, I am just argumenting for it because since then I managed to line up an explanation.

I am interested if you elaborate on your graph-view.

@Crul

Protocol[s] are problem solving agents.

I am not convinced that, in this framework, it makes more sense to see protocols as agents than as tools (of human agents). Am I missing something?

Well, when I think about it, I had no reason to say it either, I just wanted to close the article on some odd note, to motivate myself to write further, a follow-up. But then, I thought about it and a lot of what I have seen in SoP points to the fact that, protocols have much more in common with their agents, both have the notion of death, memory, emergence, being either bad or good, need for technology to sustain, etc. Most importantly, Protocols and PSAs are not solutions, they are entities to reason a solution.

If I missed something, let me know.

About the broken link, well, the website is generated via Obsidian Publish from my Obsidian notes, so the missing link is just me not publishing that note yet, because I haven’t gotten to write. I have some prompts for it but not the time to squeeze the insights.

[Crul]

@ergod

Honestly, there is no reference behind it or inspiration beside my own cooking. Although, I am not claiming its mine in any way, I am sure someone had this idea before.

In that case, congratulations :).

@ergod

To be honest, when I wrote this I also felt there are more than two ways, I am just argumenting for it because since then I managed to line up an explanation.

The explanation helps, thanks!

@ergod

I am interested if you elaborate on your graph-view.

It’s probably influenced by the “Interactive graph” widget on symplectic.link. I don’t think there is any deep insight, it’s just a generalization of a (physical) puzzle that makes sense to me. The pieces would be the nodes and the edges would connect adjacent pieces of the solved puzzle. Solving the puzzle would be analogous to building the graph of the connected pieces.

The term “puzzle” works better with the verb “to solve”, so it makes sense in the context of the post. I used the term “graph” because it makes more obvious (to me) the “topological-infinity” of ways to approach it in contrast with the “two effects”.

@ergod

Well, when I think about it, I had no reason to say it either, I just wanted to close the article on some odd note, to motivate myself to write further, a follow-up.

It makes sense as an exploratory thought… and it made me think [continues after the next quote…]

@ergod

protocols have much more in common with their agents, both have the notion of death, memory, emergence, being either bad or good, need for technology to sustain, etc. Most importantly, Protocols and PSAs are not solutions, they are entities to reason a solution.

[it made me think…] that what I miss from protocols is agency, which by (my) definition is a fundamental property of agents. I haven’t followed all the discussions about protocols here (which are probably a lot) and I would love to know if the “agency of the protocols” has been addressed.

Thanks for your time!

[ergod]

@Crul

@ergod

protocols have much more in common with their agents, both have the notion of death, memory, emergence, being either bad or good, need for technology to sustain, etc. Most importantly, Protocols and PSAs are not solutions, they are entities to reason a solution.

[it made me think…] that what I miss from protocols is agency, which by (my) definition is a fundamental property of agents. I haven’t followed all the discussions about protocols here (which are probably a lot) and I would love to know if the “agency of the protocols” has been addressed.

Good point about the agency, I am not sure it has been discussed, but again I didn’t go over everything nor been in every room of conversation. I agree with what you said, it is just a exploratory prompt. I just want to see if that case holds and what does that mean? When does a protocol go rogue? Escape our control and go on about solving things that aren’t aligned with us? Use us as its medium the same way we ride physics and many other phenomena to do our own problem solving. I started writing a fiction based on this concept, we will see how that goes.

[Crul]

@ergod

When does a protocol go rogue? Escape our control and go on about solving things that aren’t aligned with us?

That helps! I see where this comes from!!

IMH(and ignorant)O, I think a more useful approach would be to keep protocols as tools and move their agency to the different scales of organization. In that framework, protocols do not go rogue, they are just tools used in conflicts bewteen different scales.

I don’t find a good example (I need to think more about this), but the idea is that there are processes at the level of individuals and processes at the level of groups (… simplifying a lot and ingoring many other scales). Some individual processes are damaging to society and some groupal processes are damaging to individuals. The usual case (I fhink) of rogue protocols are protocols that are probably serving a purpose on the group level. Obvious analogy with the units of selection in natural selection; from wikipedia: an entity such as: a (…) molecule, a gene, a cell, an organism, a group, or a species that is subject to natural selection.

I feel like there must be some obvious at-least-no-bad examples… but I have gone blank.

EDIT: I just realized that the same argument could be made with groups… do they have agency?
I have no good answer. To me it’s easier to apply an analogy of agency with groups of individuals (which are physicial processes) than with protocols which are much more abtsract. But that says more about my limitations than about the question itself.