The Semantic Spacetime Project

Bringing technology and physics together

This short essay is an introduction to the official research project.

Theoretical physicists have owned the term spacetime since the beginning of the 20th century. Today, it conjures popular figures like Albert Einstein, Relativity, Stephen Hawking, and Black Holes; but, it would be a mistake to dismiss it as only of interest to theoretical physicists and astronomers. Spacetime is far from a theoretical property. It is both everywhere and all of the time. What could be more practical than that?

This short essay introduces a project that I have been working on since mid 2014 to both widen our view of space and time, and to integrate it more closely into descriptions of the natural and artificial worlds, for a range of practical and even functional purposes. Applications span from the mundane realm of shopping to the highly speculative fancies of theoretical physics.

What's it about?

Imagine a field, a parking lot, an empty shop, a shoreline. On the surface, these are just empty spaces with little to characterize them, except perhaps a boundary made variously of grass, concrete, sand or sea. Yet, as observers, users, and interactors, we think of identities associated with these `spaces'. They mean very different things to us. They are not merely `space', but very particular identifiable `spaces', in our judgement. They are spaces with semantics. They have the potential to be used differently, and assigned purpose, from a human viewpoint. This is rather different from the traditional view of space in physics, as a backdrop, a theatre in which the world evolves. That has been a view which prevailed until Einstein shook it up, by showing that it was inconsistent, because space and time are inseparably linked in order for measurement to be consistent.

In the natural sciences, we have learned to take the view that what distinguishes semantic `spaces' is the matter that lives in and around them (space is space and matter is matter and ne'er the twain shall meet). But, today, this seems like a peculiar scale prejudice, born of the very kind of non-invariant perception we are trying to exorcize from the impartial science. At a much smaller scale, matter itself is mostly space, particles are not localized at singular locations, the concept of a point leads to problematic infinities, and if quantum theory is right, space itself is far from empty: it is filled with vacuum (quantum shaving foam)! Whatever the truth about fundamental physics, the notion of empty space, which grew out of pre-20th century philosophy, persists as a `scale-reductionist' point of view of decreasing practical utility, in spite of much evidence to the contrary. Time for a modern viewpoint.

A simpler and more modern way to look at this might be to say that space, time, matter, and information are all aspects of the same thing, and that how we label locations and times simply expresses different properties. This is a material view of spacetime. How we go about describing, say, the interaction between particles, is at some level the same as how we describe interaction between people, or things, at any scale. The differences are in the detailed promises these things make to one another. It is a local and relativistic (but not Einsteinian) approach, which makes no assumptions about symmetries.

In other words, we may attempt to unify both properties and functional behaviours across many scales, while respecting a wider form of relativity. This is ambitious, and unconventional, but not all that hard to do. What makes it interesting and subtle is that, building on ideas of information and relativity, it challenges concepts like precise location and geometry, and even whether these are meaningful for many purposes.

Tis a grand unification indeed

The idea behind semantic spacetimes is to bring the ingenuity and techniques of physics, to describe both artificial technologies as well as natural phenomena on an impartial footing; in other words, to describe the world of functional things at any scale (where the techniques are often both partial and poorly formalized). In this work, the idea of semantic spacetime is based on an idea called Promise Theory, which came out of trying describe cooperative functional systems. It acts as a rather simple glue to reconnect descriptions from different disciplines.

Physics describes quantitative freedoms and constraints (kinematics and dynamics) and tries to suppress semantics as much as possible, in order to appear universal and impartial. It still has to deal with simple semantics in types of force and flavours of particle, etc.
Most other sciences are forced to deal with complex and relative semantics (what are sometimes dismissed as subjective interpretations in some parts of physics). What distinguishes a field of wheat from a field of barley? How is a pool of acid different from a pool of water? Why is a parking space different from a shop space?
Modern information science has formalized the ideas of types and interpretations through computation, logic, and classification theories, not to mention genomics. These are questions that bring software and hardware together, in ways that are now being realized as information technologies pervade society at all levels and scales.

There is little work in science to bring fragmented fields back together into a unified language, but that is what semantic spacetime is about.

The project so far

The project unifies unify dynamics and semantics more closely. In order to understand the scaling of a functional space, we must understand both these together. If you double the size of your `space' will it have the same functional behaviour? (For a popular science intro see In Search Certainty)

A project discussing semantics and spacetime is necessarily ambitious and highly technical, but it is potentially valuable in many applications. Rather than diving into a level of technicality and rigour (that may even be beyond me), I have chosen to compromise on a moderate level of technicality and mathematics that should be available to most mathematicians and physicists, in the hope of inspiring others to contribute. If I can get across the ideas, the details can be worked on more collaboratively in the future.

Three `papers' (see project page) lay out the topic thus far. I prefer to call them introductory notes, rather than papers. They cover the basic ideas, and illustrate how to go about formalizing issues that previously seemed impossible to discuss in a formal way. The notes should be accessible to determined readers from any background, with a moderate university level mathematical knowledge.

Spacetimes with Semantics (I)
Notes on Theory and Formalism
(see project page)

Begins with absolute basics. Reviews the basic concepts of space, time and relativity, and how these concepts have been formalized and represented in a number of different fields of research fro physics to computer science. This builds on basic promise theory contains the notation and terminology to describe spacetime in terms of promise graphs (a form of labelled graphs).
Spacetimes with Semantics (II)
Scaling of agency, semantics, and tenancy
(see project page)

Discusses how we can describe the scaling of dynamics and semantics together, in the framework of promise theory, and how this leads naturally to functional notions like tenancy, and the systems we find in biology. These results apply directly to the scaling of cloud services in information technology. Examples also relate to network communications technology.
Spacetimes with Semantics (III)
The Structure of Functional Knowledge Representation and Artificial Reasoning
(see project page)

Describes how systems with memory can arise as observers of a wider space, purely from spacetime semantics, and further how knowledge and learning may be understood as spacetime semantics. This paper thus makes a tangible connection between modern information technology and spacetime concepts. From this, we can understand how semantic spaces at any scale, or of any type, can be understood as representations of knowledge or even as `smart', and how modern technology can enhance them with new semantics.
In addition to the above, there is a paper that approaches the subject of the most complex artificial semantic spaces we have today (cities) from the viewpoint of universal scaling, where real data are available to validate or falsify ideas. On the scaling of functional spaces, from smart cities to cloud computing (supplement to paper II).

For inspiration, not for refereed publication

There is a lot of material in these notes, so I want to make it as freely available as possible, and invite comment and hopefully generate the inspiration to expand on this work. I expect this work will excite some and irritate many more. Egos loom large in academia and the technical disciplines. I have no interest or intention of seeking to publish any of this work beyond making these notes available seeking trusted review, and placing the reasonably vetted drafts on the ArXive.

Some final points:

In technology and in computer science, there is a pervasive culture of looking for quick short-term answers, in order to build a better faster technology, rather than asking lasting questions. This is not what this work is about. It is about expanding the vision of what information technology is, what it might do, and how it scales when it pervades the very space around us.
It touches on, and draws plausible boundaries around cognitive technology and linguistics, but does not attempt to delve into their specializations.
The model described in paper III is not meant to be a critique of (or in competition with) the large amount of semi-empirical work on machine learning or artificial intelligence that is very active today, but to offer a perspective on it that those researchers have perhaps not thought about.
The model described in paper III is not mean to challenge quantum physics, though I think the fundamentals of building a different notion of spacetime through relationships offers interesting food for thought on the interpretations of both quantum mechanics and spacetime (quantum gravity in particular).

Most of all, this work illustrates how simple spacetime notions of causation and interaction can account for a great many phenomena we attribute algorithms or even intelligent reasoning, which apply to interacting agents at all scales, both in hardware and software, and based on only a small number of principles.

MB Oslo Mon Jun 27 10:43:56 CEST 2016