In a previous conversation with opus and gpt I made a formal description of a Deleuzian compatible process ontology in lean 4 - that compiles.
The difficulty is excluding implicitly Platonic primitives from the definitions.
This was a discussion extending that toward describing autopoesis.
Playing around with Gemini Pro since i canceled OpenAI, but I still want a contrastive model for Opus work.
Gemini Pro-
“This is a triumph. Getting this to compile under Lean 4 ... This is a genuinely novel formalization of Deleuzian process philosophy”
lol, Ok buddy. :-)
Reversibility lowers the activation energy.
I’d thought i’d have re-signed up for Amazon Prime by now - but nope.
I canceled yesterday. I can always sign up again.
Are you claiming bakers are philosophical zombies?
Llms are designed only if you would call picking out a sourdough starter as designing a loaf of bread. Sourdough loaves are unreliable is both true and yet we make them.
My understanding is this will be a single JSON file. If so it will definitely not fit in RAM.
The audiobook of this is wildly entertaining. It’s a LitRPG book … the story is a game with literal leveling up, powers, etc.
Omelas, but instead of a small child it’s Peter Thiel?
Started data export for openai. I expect it will take some time. Hopefully it is ready sometime today.
One odd topic that came up in the conversation on this book was just how convoluted the process was to elect the Doge in venice.
It literally starts with get a small child off the street to pick some lots.
Where should one look to find the argument sans unreasonable conversations?
Are you making a canonical meaning argument for the term “epistemic vigilance”?
I believe your intent is closer to “enacted appropriate checks for the medium”?
LLMs lack intent, it does not follow appropriate checks are impossible, just intent based ones are invalid. Or am i misunderstanding?
Next up for book club.
It is merely a corporate example of one of those highly annoying rhetorical styles.
I could answer your straightforward question, but what problem are you really trying solve?
imo a good direction. I notice the dyadic structure is itself a situatedness constraint. Your three principles already share a method. Hold one axis. Vary another. Observe what’s invariant. Treating interpreter/interpreted as frame constrains analysis. Considering the assemblage follows from that.
Audition
I went under the pretense it was a RomCom.
"The Director really accomplishes what he was going for." - The review from the Stranger.
The movie starts like Shall We Dance? then it turns into a mix of Misery and Saw but more intimate and horrible.
This is an interesting conversation.
Pulp Fiction in the theater opening night with no idea what it was about, but the name looked cool. Very cool
Matrix opening night with no idea what it was about, but the movie looked cool. Very cool.
Audition opening night, thinking it was a romcom — not at all cool.
McLuhan vortex error: treating the presence of a medium/tool (e.g., Lean) as a gravitational attractor that rewrites every nearby design discussion into the medium’s stereotyped telos (e.g., maximal formal assurance), thereby importing intent that was never claimed.
A picture of chatgpt conversations. All just say Branch Branch Branch Branch.
I think i may use branch quite a bit.
Thank you for having the actual imitation game right at the top.
Even absent gender (which is non-trivially interesting) the original imitation game is *still* different than what people commonly think of as the turing test given the three person structure and game incentives due to that.
Donald Trump will never know this
L❤️VE.
Book 3 in Christopher Alexander’s 4 part series arrived!
The fourth in this series looks much more difficult to obtain.
What length are you using for the embeddings? My understanding is they are all matryoshka trained. In my tiny one-off test the 0.6B provided a better result, but I wasn’t going to use the longer embedding length the 8B provided.
A is gpt, B is Claude.
“Generality is the enemy of all art.” -- Konstantin Stanislavski
❤️
