3.13 std libs only
+++
ollama on the same machine as the python app
that's it
platforms: Win11 + Ubuntu
go get ollama and run it
- ollama pull gemma2
- ollama pull nomic-embed-text
- ollama serve
navigate to the cognosis folder
run python3 -m venv venv or simply python .\__init__.py or python ./__init__.py -h
(see: /cognosis/experiments/ and/or [[Morphological Source Code]] (tag in the kb))
I have started to learn quantum field theory after getting an undergrad appreciation for differential equations and quantum electrodynamics but primarily classical mechanics and Maxwell's electrodynamics because I think that string theory is a convenient model for agentic motility or my proposition that a language model building a robot and then writing code for that robot; which then goes on to do some impactful thing in the real world; is 'proof' of spooky action at a distance and 'entanglement' or the 'wave function collapsing' is involved; a previously, STRICTLY, not digital phenomenon. I rationalize this as a 'shape' of information. A 'space' of scale-invariant field theories. Or a Bayesian topology. A fundamentally quantum, stochastic process, which is only partially reversible - since observation collapses state and 'evolves' the system. Effectively 'branching' a "new universe" if you want to think of it in the Copenhagen-interpretation but I do like these really complex and difficult to learn string theories, better, than Many Worlds; which is something that I could intuitively understand and appreciate long before finally groking quantum electrodynamics. Anyways, the 'shape' is what I call the 'valuedness' of morphological, cognitive computation. Upon a substrate of binary digital 0s and 1s emerges a scale-invariant 'space'; a stage set for actors of all different kinds, and special effects and lighting, for anything whatsoever. Because I truly believe there to be some sort of shape, or form, that is quantifiable, which represents becoming, complexity, emergence, language, math, consciousness, and the wave function; and it is the infinite set of reals between 0 and 1 which don't exist in this universe at all, so to speak, instead being encoded as mere 0s and 1s (as binary/machine code), but which, indeed, contain all of the infinite complexity of the cosmos! Inertia and information forming the measurable residue on it's body that can not exist in our universe, our 3+1D spacetime. I call it the "morphology" of information. And I posit that it extends into the realm of logic, maths, evolution, and all possible/relevant nomenclature because it IS a multi-layer competency function that excels in existing in myriad forms throughout spacetime. Another proof I give is the 'shape' of the no-copy c-standard library method of software architecture on Von Neumann Architecture. It's always the same! Demonstrably-so. One could take programmers unfamiliar with C, or convention, and they would eventually 'settle-on' the same exact architecture as the initial nerds did 40 years ago, because there IS a morphology of code defined by the cost of information access and thermodynamics! I rationalize this with Landaur's theorem and what I am researching I call quantum informatics which states the fundamental 'cost' of information in-terms of constants of the universe and I position Maxwell's Demon, itself, as the 'thermodynamic head' of all morphological computation. I call that "Morphological Source Code". Or the spontaneous ability for entropy to 'become' into something that didn't exist, before. When Maxwell's Demon, rather than observing the natural, non-relativistic, flow of information past it's head, ejects the current state as heat, fundamentally observing and collapsing the systems wave function (and forming the quantized 'shape' of information/computation).
In modern computational paradigms, we face an ongoing challenge: how do we efficiently represent, manipulate, and reason about data in a way that can bridge the gap between abstract mathematical models and real-world applications? The concept of Morphological Source Code (MSC) offers a radical solution—by fusing semantic data embeddings, Hilbert space representation, and non-relativistic, morphological reasoning into a compact and scalable system. This vision draws from a wide range of computational models, including quantum mechanics, data-oriented design (DOD), and human cognitive architectures, to create a system capable of scaling from fundamental computational elements all the way to self-replicating cognitive systems.
In MSC, source code is represented not as traditional bytecode or static data but as stateful entities embedded in a high-dimensional space—a space governed by the properties of Hilbert spaces and self-adjoint operators. The evolution of these stateful entities is driven by eigenvalues that act as both data and program logic. This self-reflective model of computation ensures that source code behaves not as an immutable object but as a quantum-inspired, evolving system.
Key aspects of MSC include:
- Hilbert Space Encoding: Each unit of code (or its state) exists as a vector in a Hilbert space, with each vector representing an eigenstate of an operator. This enables non-relativistic transformation and morphological reasoning about the state of the system.
- Stateful Dynamics: The system evolves based on the application of operators, where state transitions can be understood as quantum stochastic processes—functions of time that collapse into a final observable state.
- Self-Adjoint Operators: The computation is inherently tied to symmetry and reversibility, with self-adjoint operators ensuring the system's unitary evolution over time, similar to quantum mechanical systems.
At the heart of the Morphological Source Code framework lies the principle of semantic vector embeddings—a novel way of organizing and representing data such that it can be directly processed as executable, stateful code. These semantic vectors map to eigenvalues and self-adjoint operators within Hilbert space, thus opening up a path to reasoning about code with the rigor of quantum mechanics.
By aligning the structure of source code with quantum information dynamics, we create an environment where computation itself becomes morphologically meaningful—where every operation on the system has inherent semantic meaning encoded in its structure, both at the operational and theoretical levels.
MSC does not merely represent a computational process, but instead reflects the phase-change of data and computation through the quantum state transitions inherent in its operators, encapsulating the dynamic emergence of behavior from static representations.
MSC allows for lightweight indexing and retrieval of semantic context embedded within the code itself, optimizing performance for resource-limited hardware while maintaining meaningful inference in local contexts. The system supports data embeddings where each packet or chunk of information can be treated as a self-contained and self-modifying object, crucial for large-scale inference tasks. I rationalize this as "micro scale" and "macro scale" computation/inference (in a multi-level competency architecture).
By applying MSC, we can encode game entities as morphological objects where state transitions happen in an eigenvalue space that dynamically evolves based on interaction within the game world. Memory layouts are optimized for cache locality, ensuring fast processing of game mechanics that are inherently state-dependent and context-aware.
Leveraging cache-aware bulk transformations in MSC allows for the efficient manipulation of data states across distributed systems. The system's predictable memory access patterns combined with semantic indexing enable high-performance in mission-critical applications.
One of the most exciting applications of MSC is its potential to model agentic motility—the ability of an agent to navigate through spacetime in a relativistic and quantum-influenced manner. By encoding states and transformations in a higher-dimensional vector space, agents can evolve in multi-dimensional and relativistic contexts, pushing the boundaries of what we consider computational mobility.
The semantic embeddings of data ensure that each component, from source code to operational states, maintains inherent meaning throughout its lifecycle.
By mapping MSC to Hilbert spaces, we introduce an elegant mathematical framework capable of reasoning about complex state transitions, akin to how quantum systems evolve.
By embracing data-oriented design and cache-friendly layouts, MSC transforms the way data is stored, accessed, and manipulated—leading to improvements in both computational efficiency and scalability.
MSC acts as a bridge between classical computing systems and quantum-inspired architectures, exploring non-relativistic, morphological reasoning to solve problems that have previously eluded purely classical systems.
The true power of MSC lies in its potential to quantize computational processes and create systems that evolve and improve through feedback loops, much like how epigenetic information influences genetic expression. In this vision, MSC isn't just a method of encoding data; it's a framework that allows for the cognitive evolution of a system.
As we look towards the future of computational systems, we must ask ourselves why we continue to abstract away the complexities of computation when the true magic lies in the quantum negotiation of states—where potential transforms into actuality. The N/P junction in semiconductors is not merely a computational element; it is a threshold of becoming, where the very nature of information negotiates its own existence. Similarly, the cognitive event horizon, where patterns of information collapse into meaning, is a vital component of this vision. Just as quantum information dynamics enable the creation of matter and energy from nothingness, so too can our systems evolve to reflect the collapse of information into meaning.
Through Morphological Source Code, we are charting a course that blurs the lines between classical and quantum computation, epigenetics, and self-replicating cognitive systems. This approach unlocks new possibilities for data representation, computational efficiency, and semantic reasoning—creating a system that is not only efficient but alive with meaning and purpose.
MSC offers a new lens for approaching data-oriented design, quantum computing, and self-evolving systems. It integrates cutting-edge theories from quantum mechanics, epigenetics, and cognitive science to build systems that are adaptive, meaningful, and intuitive. In this work, we don’t just look to the future of computation—we aim to quantize it, bridging mathematical theory with real-world application in a system that mirrors the very emergence of consciousness and understanding.
Morphological Source Code, Data-Oriented Design, Hilbert Space Representation, Quantum Stochastic Processes, Eigenvalue Embedding, Game Development, Real-Time Systems, Cache-Aware Optimization, Agentic Motility, Quantum-Classical Computation, Self-Replicating Cognitive Systems, Epigenetic Systems, Semantic Vector Embedding, Cognitive Event Horizon, Computational Epigenetics, Computational Epistemology.
The ontology of objects assumes that there are elementary objects, called “particles,” out of which all more complex objects—and therefore the whole of reality—are constituted. Similarly, the ontology of relational agency assumes that there are elementary processes, which I will call actions or reactions, that form the basic constituents of reality (Heylighen 2011; Heylighen and Beigi 2018; Turchin 1993).
A rationale for the primacy of processes over matter can be found in quantum field theory (Bickhard 2011; Kuhlmann 2000). Quantum mechanics has shown that observing some phenomenon, such as the position of a particle, is an action that necessarily affects the phenomenon being observed: no observation without interaction. Moreover, the result of that observation is often indeterminate before the observation is made. The action of observing, in a real sense, creates the property being observed through a process known as the collapse of the wave function (Heylighen 2019; Tumulka 2006).
For example:
- Before observation, a particle (e.g., an electron) typically does not have a precise position in space.
- Immediately after observation, the particle assumes a precise position.
More generally, quantum mechanics tells us that:
- Microscopic objects, such as particles, do not have objective, determinate properties.
- Such properties are (temporarily) generated through interaction (Barad 2003).
Quantum field theory expands on this, asserting that:
- Objects (particles) themselves do not have permanent existence.
- They can be created or destroyed through interactions, such as nuclear reactions.
- Particles can even be generated by vacuum fluctuations (Milonni 2013), though such particles are so transient that they are called “virtual.”
At larger scales:
- Molecules in living organisms are ephemeral, produced and broken down by the chemical reactions of metabolism.
- Cells and organelles are in constant flux, undergoing processes like apoptosis and autophagy, while new cells are formed through cell division and stem cell differentiation.
In ecosystems:
- Processes such as predation, symbiosis, and reproduction interact with meteorological and geological forces to produce constantly changing landscapes of forests, rivers, mountains, and meadows.
Even at planetary and cosmic scales:
- The Earth's crust and mantle are in flux, with magma moving continents and forming volcanoes.
- The Sun and stars are boiling cauldrons of nuclear reactions, generating new elements in their cores while releasing immense amounts of energy.
In this framework:
-
Condition-action rules can be interpreted as reactions:
{a, b, …} → {e, f, …}
This represents an elementary process where:
- The conditions on the left ({a, b, …}) act as inputs.
- These inputs transform into the conditions on the right ({e, f, …}), which are the outputs (Heylighen, Beigi, and Veloz 2015).
Agencies (A) can be defined as necessary conditions for the occurrence of a reaction. However, agencies themselves are not directly affected by the reaction:
A + X → A + Y
Here:
- The reaction between A, X, and Y can be reinterpreted as an action performed by agency A on condition X to produce condition Y.
- This can be represented in shorter notation as:
A: X → Y
While an agency remains invariant during the reactions it catalyzes:
- There exist reactions that create (produce) or destroy (consume) that agency.
Thus, agencies are:
- Neither inert nor invariant.
- They catalyze multiple reactions and respond dynamically to different conditions:
A: X → Y, Y → Z, U → Z
This set of actions triggered by A can be interpreted as a dynamical system, mapping initial states (e.g., X, Y, U) onto subsequent states (e.g., Y, Z, Z) (Heylighen 2022; Sternberg 2010).