Plog n2
A formal entry in the Plog series, part of the Archival series.
Plog 1 can be found here: https://www.dottheory.co.uk/paper/playable-worlds
These different models for retaining validated information as a function in a long-existing series of experimental and observation-logging objects. This websites cites prior experiments with these here:
https://www.dottheory.co.uk/paper/observed-effects-of-framework-onboarding and https://www.dottheory.co.uk/paper/captains-log
MetHaMatrix
DOT THEORY / CONSTITUTIONAL PHYSICS RESEARCH PROGRAMME
Navigational Tools and the Burden of Ontology
Plog — Entry II | 16 August 2026
A second thought follows from the distinction between imagination, visualisation and ontological rendering:
Perhaps scientific tools should remain navigational until something further has been earned.
A mathematical formalism helps us navigate relations.
A coordinate system helps us navigate location.
A probability distribution helps us navigate expectation.
A model helps us navigate observations and possible outcomes.
A simulation helps us navigate consequences under stated assumptions.
An ontology does something stronger. It proposes what there is.
The difficulty begins when the success of the navigational tool is treated as evidence that the ontology suggested by the tool must possess substrate-like reality.
That transition is extraordinarily easy to make because a successful scientific model becomes increasingly difficult to distinguish psychologically from the world it models.
The better the map becomes, the easier it is to forget that it remains a map.
The virtue of not knowing
This suggests an interesting challenge for science.
Science is exceptionally good at subjecting proposed observations and theories to demands such as:
What is the evidence?
What does the experiment discriminate?
What does the calculation actually establish?
What alternative explanations remain?
What would falsify the proposition?
But perhaps the same discipline should be applied to the ontological interpretation of the scientific model itself.
If the available experiment establishes A while several ontological descriptions O₁, O₂, O₃ … remain compatible with A, then the scientifically justified conclusion may simply be:
A established
ontology underdetermined
That is not scientific weakness. It may be scientific precision.
The challenge is therefore not necessarily to eliminate ontological uncertainty. It is to justify exactly how much ontology the evidence can presently carry.
A scientific tool should remain navigational until its transition from navigation to ontology has acquired an independent warrant.
The tool may still suggest ontology. It may inspire one. It may make one extraordinarily compelling. It may even supply the structure from which a testable ontology eventually emerges.
But suggestive power and ontological entitlement are not the same scientific object.
Applying scientific scepticism to science
There is a curious asymmetry here.
Scientific method rightly subjects incoming claims to scepticism. Data are interrogated. Explanations are compared. Assumptions are exposed. Measurements acquire uncertainty. Conclusions are bounded.
Yet once the resulting object is called a scientific theory, some of that restraint can disappear.
The theory becomes the language through which the evidence is subsequently interpreted, and its conceptual furniture can quietly begin to inherit the authority of the observations that support the theory.
A computational object becomes a physical object.
A formal dimension becomes a physical dimension.
A state space becomes somewhere reality occupies.
Information represented by the theory becomes information physically stored by the universe.
A possible history becomes a physically traversed history.
A useful ontology becomes the ontology.
Again, any one of those transitions may eventually be warranted.
The question is simply why the transition itself should be exempt from the scientific scrutiny applied to everything entering the theory.
If science asks its objects “How do you know?”, perhaps it should occasionally ask the same question of its own interpretations: “How do I know that this is what my successful representation means physically?”
Navigation need not be realism
There is nothing weak about describing a model as navigational.
On the contrary, navigation can be extremely powerful.
A map can predict where we will arrive without being made from the landscape.
A coordinate system can locate an object without being a constituent of the object.
A mathematical state space can calculate the evolution of a physical system without requiring every feature of the representation to possess a one-to-one physical counterpart.
Scientific models can therefore be physically consequential without every representational component being physically literal.
predictive reality
≠
representational reality
≠
ontological identity
A model may reliably predict events that are unquestionably real. Its calculations are themselves physically instantiated acts. Its measurements correspond to real observations. Its success is therefore not imaginary.
Yet the ontology used to make those calculations intelligible may remain one of several admissible renderings.
There is no contradiction in this.
Information physics makes the problem particularly interesting
Information physics makes this question unusually difficult because “information” crosses several representational registers very easily.
a mathematical quantity;
a relation between possible states;
a property of a communication channel;
a computational object;
a thermodynamic quantity under specified operations;
a physically instantiated pattern;
something known by an observer;
or, in stronger proposals, something assigned fundamental ontological status.
Those uses may be related. They need not be identical.
The scientific question therefore cannot merely be: “Is information real?” The word real is doing too much work.
A more discriminating set of questions might be:
In what operation is information being identified?
What physical transformation corresponds to that informational quantity?
Which properties survive a change of representation?
Which properties are consequences of the modelling apparatus?
What experiment distinguishes an information-bearing description from an information-substrate ontology?
That does not weaken information physics. It may be exactly what allows its stronger claims eventually to become more precise.
The current scope of the Information Physics Institute makes this question especially interesting to me. Its programme brings together physical, informational, computational and cosmological approaches rather than requiring “information” to occupy only one representational register. That diversity creates an opportunity to ask not only what information can explain, but what different informational descriptions are entitled to mean.
Ontological restraint as a result
Perhaps we have been culturally inclined to think that a theory which declines to tell us what reality ultimately is remains incomplete.
Quantum mechanics has made that discomfort particularly visible.
But perhaps there are circumstances in which:
ontology underdetermined
is itself a scientific result.
If two representations produce the same discriminable observations, then the absence of evidence separating their ontologies does not necessarily create an obligation to choose one.
The absence may itself tell us something:
the presently available operation does not carry sufficient information to adjudicate that distinction.
That is not an assertion that no underlying distinction exists.
It is a statement about the jurisdiction of the present evidence.
This makes scientific restraint itself informational.
The inability to distinguish O₁ from O₂ is not “nothing”. It tells us something about the information presently available to the observer and the discriminatory capacity of the operation used.
The navigation principle
I would therefore provisionally state the idea as:
Scientific representations should be treated as navigational instruments within the jurisdiction they have empirically earned. Ontological properties suggested by those representations require an additional warrant before being attributed to the represented world as substrate properties.
This retains imagination.
It retains speculative physics.
It retains mathematical invention.
It retains simulations, holographic descriptions, multiversial modelling, information ontologies and radically unfamiliar physical hypotheses.
What it removes is only the automatic inheritance:
the model successfully contains X
⇏
reality therefore literally contains X in the same representational sense
And perhaps that is particularly important where several apparently incompatible models each demonstrate empirical efficiency.
Their coexistence need not force us immediately to construct an ontology large enough to make all of their pictures simultaneously literal.
It might instead tell us:
We have several powerful navigational tools. We have not yet earned the additional operation that determines which aspects of their imagery belong to the substrate itself.
That is not the end of the scientific inquiry.
It identifies where the next one begins.
Research note
This entry follows the preceding Plog observation concerning the transition from imagination and visualisation to ontological rendering. It was generated through correspondence within the Information Physics Institute concerning quantum interpretation, informational descriptions, observer relations and the physical status attributed to mathematical objects.
The references to IPI describe the intellectual setting from which the question arose; the position expressed here is my own developing interpretation within the Dot Theory / Constitutional Physics research programme and is not presented as an institutional position of IPI.
Working proposition: indistinguishability of ontologies under an operation is information about the operation.
Stefaan Vossen ⊙
This Plog is accompanied by visual artwork entitled Laborynth: www.dottheory.co.uk/paper/MethaMatrix
