The Dot Theory

This website offers an understanding of a theory of epistemology. An explanation, and interpretive framework for modelling and acting under uncertainty, available for testing and adoption under conditions.

Author’s note:

This website presents, for peer review, a novel and unusually abstract idea on the modelling of realism. Its value only emerges through the reader’s own interpretation and reconstruction, rather than from the artefact alone. In other words, novel combinations of existing scientific premises are presented, which the reader will be required to reinterpret. As a literary device, the approach to structure and content displayed across this site is necessary to fulfil its role and duties as a non-prescriptive scientific programme. While every effort has been made to keep it accessible through programme-aligned examples, structured narratives and blog posts, it may not feel immediately personable or recognisable as it stands as an individual and novel work. This, sadly, is unavoidable, and I call for your tolerance. It does not aim to convince or sell, but, as a website, is simply a standalone piece of intellectual work in epistemology and the logic of philosophy of natural science that I have had the pleasure of publishing.

First publication: September 2024, Dot theory, a work in Epistemology - formal epistemic modelling

Dot Theory*, an introduction

What it is:

This website is a singular curated unit, and functions as the working environment for a continuous piece of research, locally known as the Dot theory. It is designed to formulate an architectural standard for representational completeness across domains, whether as data or information. This makes this website both a) a philosophical proposal with practical outcomes, b) a practical proposal with philosophical implications and c) a complex object inviting potential study. The purpose of publishing it as a website is to make it freely available for scrutiny, discussion, adoption and testing.

The core framework is presented in its most stable form as a scientific programme. It is demonstrated through its primary explanatory tabs and papers, some highlighted as links in the text, while other accessible materials on the site explore its interpretation, morality, formalisation, and application.

What it does:

Beyond its position in epistemology, no one paper in this website will present a singular final physical formalism as more than a toy suggestion. Many forms of physical formalism that express the idea of the Dot theory can be found across the website. From physics to Ai, via law and ethics, they are repeated in various modes and fashions across the site for clarification. Some components may be provisional and are intended to support development rather than serve as final statements.

*In this work, the limited scope for the term ‘theory’ in ‘Dot theory’ refers to its use as an epistemic modelling framework describing how humans structure, represent, and stabilise experiential inputs into forms that become real to them and recordable within shared systems. Dot theory is using the word ‘theory’ as meaning:

a framework that models how knowledge, interpretation, and meaning are constructed and stabilised

More precisely:

  • Like Bayesian epistemology, it treats knowledge as something structured and updated under constraints

  • Like work in Cognitive Science, it treats perception and understanding as active interpretation, not passive reception

  • Like Phenomenology, it recognises that experience is always context-mediated

  • Like constructivist approaches in Sociology, it acknowledges that shared reality is stabilised through systems and records

  • Like Cybernetics, it emphasises feedback, stability, and error correction

  • Like Ludwig Wittgenstein’s language games, it treats meaning as arising from use within structured contexts

  • And like Artificial Intelligence, it is concerned with how representations are built and used operationally

The key difference with the existing scientific paradigm is this:

Dot Theory is not primarily a theory within one domain, but an attempt to provide a common structural architecture that can describe all of these processes in the same terms.

In other words:

it tries to abstract what these frameworks are doing at a structural level and express that in a unified form, centred on the relation Ψ = (ψ, μ)

In that sense, its use of “theory” is closest to:

  • a meta-framework

  • or a general epistemic modelling architecture such as cybernetics, language games and Artificial Intelligence.

Dot theory does not derive its consistency from empirical law-likeness, but from the formal constraints governing the transformation it defines. When implemented, it behaves as a rule-bound system producing consistent outputs under specified conditions.

How it does it:

As a whole, this site aims to be a coherent, scientifically structured, philosophical proposal and research programme for a unified theory of interpretation and truth. It introduces the idea of an epistemological framework and a toy system (GIA) that models how humans construct and evaluate representations of reality under constraint. The work is described semi-formally and illustrated for elucidation rather than full formalisation, but is intended to function as a formal epistemic object in its entirety.

Technically, the architecture is generative at the level of calculation, not control. It enables autonomous construction of context-conditioned predictions, while the scope and legitimacy of those calculations are constrained by the consent-relationships encoded in the contextual structure μ (for those unfamiliar with the annotation, explanations of terms are included in individual papers).

In short:

The underlying idea is straightforward: many systems of representation rely on reduced models that omit contextual structure. In some cases, this omission is harmless; in others, it leads to systematic failure. Under what conditions is a reduced representation sufficient, and when must contextual information be explicitly included?

The work on this site proposes developing a theoretical data architecture capable of supporting such predictions, assuming an optimised method for accessing sufficiently rich data. It is in this sense that it is an aspiring, and as yet unconfirmed, theory of the modelling of representations under constraint. The practical approach centres on AI agents whose models are constructed through query-dependent selection of relevant information, rather than on a fixed, mixed, and monolithic dataset.

End of executive introduction

Life is all about finding out that:

  • life is real

  • life is a game

  • that game has rules

  • reality is the record of it

  • you can not play a game in it

  • you can not play a game and not finish it

  • the game you play is the one you most understand

  • you are living less, the less playfully you play your game

Extended Introduction and Conceptual Foundations

What this site is:

A formal epistemological research programme in the philosophy of science that proposes a context-explicit modelling framework, with exploratory extensions into AI governance and institutional design.

In other words, Dot Theory, as an abstract object in formal epistemology (the study of knowledge and its structure), is presented here as a research programme in philosophy of science with one specific technical aim:

To investigate whether certain current physical and computational formalisms are representationally incomplete with respect to contextual and observer-conditioned information, and, if so, to propose an architectural extension that is mathematically coherent, reduces to standard theory in appropriate limits, and yields either empirically discriminable predictions, or demonstrable improvements in modelling outcomes.

This is not:

An interdisciplinary manifesto or physical theory. It is a scientific programme intended to be evaluated through formal definition, derivation, and, where applicable, empirical relevance, within a novel formal framework. Its innovation is minimal but hypothetically pertinent. Unusual, yet non-trivial in its potential impact.

The writing on this site is occasionally personal in tone because I think motivation matters. The logic, however, has to stand without rhetoric across all branches and can be found as follows, via a series of links on this website for publication.

For internal navigation:

Internal links to

The core programme claims in one paragraph:

Physical theories map observations into state representations and then evolve those states to generate predictions. Dot Theory asks whether certain classes of contextual variables that affect modelling and measurement are being treated implicitly or discarded entirely in standard representations. If and where such variables can be formalised as auxiliary structure, an extended state representation may logically be warranted.

The programme requirement (of which the physics programme is a manifestation) is strict: any extension, therefore, must be consistent, must preserve required symmetries unless explicitly justified, must recover the standard formalism as a limiting case, and must generate at least one clear empirical discriminator.

What Dot Theory does not claim:

To keep the site accountable, it is important to set boundaries up front.

Dot Theory, as a central claim and programme, does not, on its own, claim to:

  • replace Quantum Mechanics or General Relativity

  • “solve” unification by assertion

  • derive consciousness from physics, or physics from consciousness

  • establish a universal ethic

  • offer conclusions without derivations

It merely offers toy examples in relation to them for further instrumentation, instruction, development, familiarisation and maturation. As an object, it single-mindedly offers introduction and framing, not prescription. These emerge on an individual system basis. You will find conjectures and programme proposals here. Where something is conjectural, it is labelled as such. Where something is (semi-)formal, it is presented with such definitions, assumptions, and conditions.

Where to start:

If you want the programme-level overview first:

  • Project Overview: what is being claimed, what is not, and what would count as success or failure. This does not claim that reality is projected, but that its experience, as it is available to us in science, is contextually interpreted under constraint. With that precise understanding, expansions into other areas, such as experimental physics, healthcare, legislation, and social policy, readily note that new constructive, safe, and ethical optimisation and resource-management options become feasible and available.

If you’re into law, ethics and want the constitutional and institutional extension of this interpretation:

  • Informational Constitutionalism: a structural argument about procedural access to evaluative information in computational governance. Combined with an understanding of the implications of implementing Normative Generative Architectures in real-world applications, it offers a deep, well-defined understanding of the operator logic of Dot theory.

If you want to expand on aspects of human motivation and experimentation within a disciplined scope:

  • Happiness & Health: how representation, feedback, and agency relate to wellbeing under partial observability, without metaphysical inflation.

If you want technical material on logic in philosophy of natural science through the Dot theoretical interpretation:

Blog posts between linked pages remain as working notes and drafts, along with links to individual papers for elucidation and narration. They are exemplary of the core programme pages only.

AI is implicit in the program and is discussed variously as a tool, used as a tool, and recognised only as a tool.

A minimal formal orientation to the technical reader:

When notation appears on this site, it is used in a restrained and orienting sense:

Let ℋ denote a conventional state space and let ψ ∈ ℋ represent a standard system state.

Across this site’s papers and posts section, various toy applications of Dot Theory explore whether an extended representation Ψ = (ψ, μ) ∈ ℋ × ℳ is warranted, where ℳ denotes a space of contextual or structural metadata.

This is a representational proposal, not presented as an assumption. The validity of any specific choice of ℳ, and of any dynamics defined over ℋ × ℳ, must be demonstrated by developments in the project’s extensions rather than presupposed. However, this work only features a framework that makes this action possible and invents a space for it to occur from existing components, thereby redefining and using them as familiar objects. Such is the limit of the scope of this project.

To the reader:

Engagement with this framework is complex but does not require formal or mathematical treatment. The notation serves as a formal philosophical orientation within the discipline of philosophy of science, particularly for readers familiar with such structures and allows for technically robust statements where needed. The core ideas, however, can be understood conceptually without it. Where mathematical language is used elsewhere on the site, including references to matrices or compatibility conditions, it reflects one possible formal expression of the same underlying idea: that representation depends not only on state, but also on the structure under which that state becomes admissible. For readers less familiar with such notation but interested; AI tools may assist in interpreting and exploring the work.

Publishing it here makes it readily available for analysis, discussion and adoption under adequate conditions. Your participation is essential.

Closing:

I built this site for a possibly niche, but emerging cross-disciplinary audience with a formal interest in epistemology, particularly those working at the interface of physics, artificial intelligence, ethics and the philosophy of science, where questions of representation, modelling, and interpretation are becoming increasingly central to human activity and civilisation.

It is at that intersection that I stand to invite serious critique on how we currently describe, build and legislate information systems. That may well not be you, but it affects you and might be worth considering regardless. This is my invitation to you.

If the framework is found to be useful and adopted, it will be because it improves modelling under stated assumptions and survives confrontation with data. If it fails, it should fail clearly. Comments are open and welcome in all papers individually.

Either way, the work is better for being tested.

Thank you for reading. For a good next read from here: https://www.dottheory.co.uk/logic

Stefaan

We are here to experience the world we create