The Paradox of Multiple Autonomous Entities in a Procedural Environment

Echoes of a Dead Species

As we know, PISES is a fully procedural piece of software. This means that environmental detail is only generated when an environment is traversed. At program-start, the galaxy consists of a number of Solar System Frames, which are skeletal data structures containing just enough information to create a persuasive galactic view.

Inside a solar system frame, there exist a number of World System Frames. Much like their solar counterparts, these frames are skeletal data structures that contain only trace information about planetary bodies.

The only way to trigger the generation of a more defined world system is to enter the planetary scope, at which point we will generate more (not all) of the planet’s detail.

The software user may be the one to trigger the generation of a new world, but they are not the only entity in PISES which may be autonomously exploring new worlds. A procedurally generated species in PISES may very well be conducting explorations of its own, most likely at some time in the past (more on this later).

The results of these two explorations can sometimes result in paradoxical experiences for the software user. In essence, we may run into a situation where the software user might catch a glimpse of two alternate contradicting timelines of the galactic saga.

Rolling Back History to Generate a New World

When an observer (human or otherwise) visits a new WorldSystem, PISES must roll backwards in time to the moment of the world’s creation and begin generating the world’s history from there.

The reason we revert history is because it is impossible to instantaneously generate a world in all of its multifaceted detail, with a deep and interesting timeline (at least not in the “simulative” way we want to in PISES), without creating geological, ecological, atmospheric systems (to name a few) that evolve over time.

In PISES, we want to simulate, even if it’s a notional simulation, the evolution of a world across billions of years. We want to create situations where a spike of geothermal activity at T-200MY might yield an incremental change in a world’s magnetic field, which might allow a primordial atmosphere to form at T-198MY, and in turn allow microbial life to begin producing oxygen by T-197MY, and so forth, all laid out on a timeline of geological scales that eventually, a software user will be able to explore in the visualizer. In order to create such detailed environmental interactions from scratch, we need to roll back and time and begin these processes at the time of world creation.

For this reason, usually when a species is generated in PISES, it’s almost never in the present. It is at some point millions or billions of years in the past: long dead by the time the software user shows up.

In fact, while I in my own wanderings have discovered species in the present, I have never once discovered an intelligent civilization in PISES that was not extinct – so long extinct, in fact, that any molecular evidence of their civilization would have, at the time of my observation, been long since consumed and reclaimed by the engines of geology and heliology.

The Retrohistorical Paradox

The retro-active generation of ancient civilizations is almost always a benign and harmless process. However, occasionally PISES will spit out a spacefaring civilization. When we say spacefaring, we mean: a civilization capable of visiting new worlds, and triggering a procedural generation event.


  1. The Software User visits World A and triggers the generation of a detailed World System. For the sake of example, let us say that this world is a barren, lifeless world with few features of note.
  2. The Software User visits an “adjacent” world in the same solar system and triggers the generation of another detailed World System. Let us call this World B.
  3. At some point in the past (Let us say T-100MY), an advanced and spacefaring civilization arises on World B. At T-100MY+4000Y, this civilization visits World A, nearly a billion years before the software user visited World A. World A is, much like the Software User experienced, a barren and lifeless world with few features of note. So far, the universe of the user and the universe of the generated species are still consistent.
  4. The spacefaring civilization makes some impact on World A. Perhaps the civilization decides to terraform World A, and makes dramatic changes to its geological and ecological history.

The World A that the software user experienced moments ago is now no longer consistent with the timeline of the galaxy. If they were to re-visit world A, they would at a minimum see the blip in its geological history that the spacefaring civilization effected, 100 million years ago. At a maximum, they might experience an entirely different world. We have experienced a paradox, and have glimpsed into two parallel, inconsistent versions of the same universe. That’s a fancy way of saying PISES has a really complicated bug.

Bridging the Universes

How do we reconcile the discrepancies between the procedural experiences of the user and of the civilizations of PISES?

The short answer is that, while PISES remains procedural and can produce autonomous entities, there is really no good way to solve this problem. We can’t generate all detail up front, the program’s startup time would take hundreds of hours and require many terabytes of storage, at least.

One theoretical mitigation of this problem would be to generate all world systems, just enough to determine if they are going to produce any sort of civilization. If the world is clearly not going to produce a spacefaring entity, abandon generation and wait for the software user to show up until we finish.

If the world looks like it might produce a spacefaring entity, continue generation until the world (or species) disqualifies itself.

This of course brings us to the next challenge: at what point do we decide that a planet will not give rise to civilization, or that a civilization will never become spacefaring?

Some worlds, surely, we can throw out quickly: small, rocky, tectonically dead planetoids like pluto, charon, etc. But other worlds are more ambiguous. Mars, billions of years ago, is thought to have had liquid water and notable seismic activity. Who is to say that some exo-mars might not experience a reverse process: perhaps wandering into a favorable orbit, or having its tectonic engines jumpstarted by some sort of massive impact?

Other worlds still, like Venus, prove even more mysterious. Scientists have begun to speculate that phosphene in the Venusian atmosphere may in fact indicate the presence of microbial life. The conditions for life are always expanding, and who is to say that civilization is required for life to become spacefaring?

In a recent PISES run, a Type-II Gas Giant (A sort of gas giant supporting water clouds) managed to become home to airborne phytoplankton in its uppermost atmospheric decks. The world’s exosphere extended so far out into space that its innermost moons were in fact enveloped by the exosphere (not unlike the Earth and its own moon) – which means this phytoplankton, by virtue of its extreme habitat, was in fact spacefaring! How would this Phytoplankton fare on these moons – could it take hold there, survive there, become a terrestrial organism Perhaps even make an impact on the ecology of these moons? Unfortunately, this unique situation was not accounted for by PISES, but if it had been, that phytoplankton would have needed to trigger the generation of three new worlds.

Even more ambiguous is where we might choose to disqualify a species from ever becoming spacefaring, and to halt their generation. Surely, humankind has veered perilously close to having their generation halted numerous times. Where do we draw the line? The eve of nuclear war? The moment Donald Trump becomes elected? Perhaps we have actually been halted for eons, completely oblivious, hanging frozen in time, waiting for some sad observer to wander into our corner of the universe and catalyze the final leg of our fateful timeline – or to grow tired of this universe and simply close the software.

The takeaway is that there is really no good way of disqualifying a world or a species from being spacefaring or not. This solution may catch some paradoxes, but surely will not catch all.

The good news is that producing a “retrohistorical paradox” in PISES is extremely hard to do and almost never happens. When it does, I have only ever seen it contained to one solar system. So for the time being, I choose to turn a blind eye to this cosmically horrifying aspect of PISES and continue working on my orbital propagation system and on raising my Icospherical vertex limits.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s