@neauoire @flaneur ontological authority for all, the elimination of barriers to the expression of creative will, low-friction co-thinking and an exocortex occupied by all with its own life as one self for everyone. the emergence of plural inhabitable spaces on subjective terms, sovereign and independent but fundamentally connected
@neauoire @flaneur oh no i wasn't really saying that. honestly i think my obsession with all this stems from parental pressure to succeed in a conventional fashion and become part of the capital class or whatever. i have no interest in that at this point but i think i'm still fundamentally productivist in a reflexive sense... regardless i'm in far too deep to be legitimately introspective about my relationship with this work, I made it part of me for better or worse (depending on one's frame)
@neauoire @flaneur i made my initial statement without thinking about the use of the definite article potentially being a problem. but yeah idk i haven't been able to make meaningful progress on my work for like 6 months due to a compiler bug that i maybe could fix given more time investment but have been too burned out to work on up until recently. given that it's personally all-important to me i haven't been doing so great psychologically, i sort of generally define myself in terms of it
@neauoire @flaneur any time i don't invest in it feels like a failure to exercise my ability to bring about the world i desire. i don't have the sort of hubris required to declare that time others don't spend helping me is wasted, especially given that the point of my project is just the facilitation of other work in novel ways, i.e. the project itself is simply a grand means and not in any way an end. i just want to plant the seed and let worlds within worlds grow forever :)
@flaneur @neauoire i am an exceedingly, almost uniquely in my experience, poor communicator when it comes to explaining what i'm working on. the only extant long-form explanations are threads on fedi and abandoned documentation efforts termed "incomprehensible" by test audiences :)
let me see if i can find one of the least incomprehensible ones
@flaneur @neauoire you probably don't want to invest the amount of effort this will require given that this is our first interaction and there isn't really any accessible material on my work.
here's a thread that i think is fairly summative?
@syntacticsugarglider I get the feeling we've found two facets of same thing, and are expressing/investigating it in our own idiosyncratic ways :)
What's your background? If you want to share.
@flaneur i think so as well. there are other people who i think are in the same sort of place. @natecull is one of them, as is cwebber and a lot of other people working in OCAP spaces. it's not immediately apparent from their public writing, but old archived blog posts from https://www.unisonweb.org/ seem to suggest a similar convergence of understanding.
my background is that i'm obsessive and have retained enough of my childhood hubris to believe things can be better
@flaneur in a more concrete sense i'm a first-year undergrad cs & math student, though that isn't really important because all the coursework i'm doing is review that's unrelated to any of my interests. i've been programming and interacting with software for a long time, and i've had a political/social/economic/philosophical awakening in the past few years over the course of which I've realized that the dreamed-for better world is consequentially incompatible with the current paradigm
@flaneur and moreover that most of the things holding us back are trivial incidental complexity and not essential difficulty and are products of the insidious and limiting characteristics of our entire collective ontology and its inability to accept the necessity of powerful *subjective* ontologies with meaningful interlinks
is that a useful summary?
@syntacticsugarglider yes, definitely! Thank you.
I did 3-4y of CS and 3-4 y of Linguistics/Literature studies. I spent around eight years working exclusively in the industry, just relatively recently came out of that shell.
@syntacticsugarglider my approach is simpler/shallower I believe, though; the [[agora]] is just a (hyper) graph where nodes are, well, *contexts* for lack of a better word. These contexts usually map to an entity, but they can be enriched by users using a simple language.
@flaneur yep, i don't want to be too irritatingly superior about all this but i would agree. my goal is a strong semantic embedding, not a better way for humans to collectively manipulate relational structure. the latter is still useful and in the same design space as the things I are about, though
@syntacticsugarglider interesting! I think I can't currently grasp the difference, but I trust you can. I wonder if we could bridge the gap in conversation; we don't seem too far apart?
@flaneur as you may have seen in that thread what inevitably ends up happening is that i refuse to concretize any of my explanations because that would be inherently reductive and people get frustrated with my inability to present compelling explanatory cases because i demand they not imply constraints or opinions i don't hold
but the gist of my approach is that the relational graph is entirely implicit and is emergent from the desire of agents to acquire other entities through arbitrary means
@syntacticsugarglider edges are hyperlinks, with the only additional complexity that links 'qualify' each other by proximity, in essence yielding a typed edge/edge set and the beginning of a DSL.
@flaneur my goal with all this is to build the minimally opinionated system as an initial seed implementation of a minimally opinionated collective reality-interoperation engine. the "seed" mentioned in the refrain of "planting the seed" of worlds
@syntacticsugarglider 'minimally opinionated' I can definitely resonate with; what I've been trying to do in my project is at least similar, in the sense that I've tried adding as few novel concepts as I could essentially -- preferring making do with relatively few operators.
I like the phrase "low-friction co-thinking"! Also "A strong semantic embedding".
I guess I'm not quite on board with the "... in Rust" part because I don't know Rust and while I'm sure it's a safer and smarter C++, for those applications where C++ is the correct choice, I'm kind of interested in applications where C++ would be a laughably poor choice. Semantic markup of arbitrary knowledge, for instance.
I kind of like virtual machines, though, for safety.
@natecull @flaneur I'm using Rust because I need two things:
- a systems language, so i can compile to minimal binaries that don't ship with some bloated runtime and are viable in embedded environments
Rust is the only thing that satisfies the above constraints, I would encourage thinking of Rust in this case not as safer C++ but as a more limited Haskell, because that's the inspiration for its type system
Sadly, I still don't have my head around what "typeclasses" as a concept are. All I get is that they're "some very low-level, very experimental, very specific to Rust thing" which isn't very helpful to me.
Could you implement an outline of Noocene in, say, Node.js? Or Python?
I'm not saying those are great languages, but they are a lot safer than a systems language, and they're a lot easier entry point for others trying to grasp deep conceptual ideas.
@natecull @flaneur those languages don't have typeclasses, and I suppose you could implement a noosphere in them but that wouldn't be a particularly enjoyable experience. also, those languages are not meaningfully safer than Rust, in fact while they don't really allow fully undefined behavior it's possible to introduce far more unusual and unpredictable behavior to their execution context as compared to Rust's. (1/2)
@natecull @flaneur typeclasses aren't "low-level" or "experimental" or even specific to Rust. Rust doesn't even call them typeclasses, it calls them traits, but I use the term typeclass because it's the Haskell convention and Haskell is where the implementation Rust uses was basically taken from. They're essentially just interfaces with some extra features ("associated types", which are like... return subtyping in a way? but more general?)
@natecull @flaneur anyhow the point of using Rust is that it has a type system that allows me to create an RPC system that is simultaneously robust and predictable while also **not having any fixed well-defined primitive set**, i.e. the space of primitives is defined exclusively in terms of what transport constraint sets are satisfied by your environment
sorta like... "what operations you've been compiled against the knowledge of how to do"
I'm sorry, but *EVERYTHING* in Haskell is 'experimental' to everyone not in the Haskell community. And everything in Rust is 'low level' to anyone outside the Rust community.
Targeting the intersection of these two communities - Haskell-heads who are looking for a better C++ - really really limits the technical audience for a deep conceptual idea, in my opinion.
Could you explain your Noocene idea using objects?
@natecull @flaneur i seem to continually give the impression that this is fundamentally tied to rust, but that's not the case. i think the problem we end up having here is that the "deep conceptual idea" is actually very simple and shallow, that's why it's valuable. thus, because of that, people end up looking for depth and complexity in the associated technical work, and then end up confused when it's seemingly unrelated. the tech isn't the point
this sums it up: https://merveilles.town/@flaneur/105738090496489820
"those languages don't have typeclasses"
That's kind of the point. You seem to have tied your *conceptual architecture*, that you're struggling to explain to others, very very deeply to this *very* specific implementation concept, not found outside of the Rust community, called "typeclasses".
(Because I think "typeclasses" in Rust are even different to in other languages).
How would you describe Noocene *without* the typeclass concept? That may help sell it.
@natecull @flaneur so, in rust a typeclass is analogous to an interface for the description i'm about to give. The syntax Box<dyn Trait> is a unique_ptr to a vtable that satisfies that interface, i.e. a set of methods. The reason typeclasses are important is that I can take some structure (concrete, specifically typed) that implements a trait (interface, describes properties/a structure), erase that into a vtable and then using `protocol` *send that vtable over an isolation boundary*
@natecull @flaneur this is basically just a *super* generalized version of something like the boundary erasure the E language provides, for example. The typeclasses aren't fundamental to this. I just do a poor job of separating things that are integral to my personal intuition from things that are crucial to conveying that intuition to others. Typeclasses are not one of those crucial things, they're an implementation detail. The idea being implemented is just a general sense of "objects"
The two short term goals I'm most excited about are opening up (integrating) all walled gardens and launching a UBI program, both using [[siphons]].
Merveilles is a community project aimed at the establishment of new ways of speaking, seeing and organizing information — A culture that seeks augmentation through the arts of engineering and design. A warm welcome to any like-minded people who feel these ideals resonate with them.