Project Persephone

Virtual exovivaria are exovivaria simulations - virtual worlds. Such worlds incur no costs except computational ones. They carry no risks except perhaps for social ones. With these advantages, they can serve Project Persephone in several ways - some functional, others recreational. The aim is make a world worth talking about, if not dozens of them.

To be useful, however, parts of these virtual worlds must be limited: they must match what's physically possible in telebotically maintaining an ecosystem in Earth orbit. In all other respects, however, users and developers would be free to do anything that's possible in CGI software. If true exovivaria ever come into being, virtual exovivaria will probably continue to have uses for the Project.

Practical uses

Virtual exovivaria can be classrooms, laboratories, playgrounds and places to begin the test of citizenship for true exovivaria. In these roles, they will complement another approximation of true exovivaria: terrestrial prototypes?. However, unlike terrestrial prototypes, virtual exovivaria will be able to simulate some aspects of life and telebotic operations on orbit that can't be prototyped on Earth.

Among their uses:

• Market testing: Virtual exovivaria could cover a wide range of designs. They could have different gravities, ecosystems, sizes, layouts and maintenance problems. The ones that attract the most game-players per launched (virtual) pound could be the best sources of ideas. Terrestrial exovivaria prototypes could be based on the more successful virtual ones, at least insofar as their success doesn't depend on low gravity or microgravity?. And real exovivaria would offer better user experiences if game- and ecosystem-designers designed them with a full understanding of gravity (low, variable, or even negligible) as a design parameter. Until actual exovivaria are launched, populated, and in full swing, the best guesses about what they should be like for users will require interpolating between virtual exovivaria and terrestrial prototypes.

• Training: There will always be new users who aren't yet skilled enough (or trusted enough) to be allowed to telebotically operate in a real-world prototype, much less in a real exovivarium. In a virtual exovivarium, these "newbies" could build needed skills, at low cost to themselves and the Project. Serious games are already used in job training, to help reduce costs and build teams. Many of these serious games are video games, virtual "microworlds". A virtual exovivarium could be a lot of fun, but also be a serious training ground for learning to perform the vital functions of real exovivaria.

• Human resource 'filtering': Any virtual world is likely to be visited by Griefers - those who are intent only on making trouble. Prototype exovivaria and real ones in orbit can be damaged (including the killing of living things) even without malicious intent. In particular, with launch costs being high, exovivaria could represent several thousands of dollars per pound of sunk costs just for transportation to LEO. They may contain equipment for managing orbital debris damage that saboteurs could also use to disable an entire exovivarium. It's important to catch this sort of user before they can add further risks to such a high investment. The most accessible virtual exovivaria could serve as honeypots - places where abusers could be attracted, identified, isolated and blocked.

• Procedural rehearsals: Exovivaria will not be all fun and games. There will be repair and rescue operations too. Orbital debris strikes are likely to be the biggest issue. Repairs and rescues will need to be choreographed. It would be wise to rehearse such operations in simulations, before trying them out in terrestrial prototypes (where mistakes could cost between tens and thousands of dollars) or in launched exovivaria (in which a totally disabling error could sink an investment of millions of dollars). The simulations might use more advanced computational resources, going to a level of physical detail not used in game play, in order to bring the more subtle risks to the surface. In general, more detailed simulation is more expensive simulation. The high costs and risks for prototyping exovivaria (not to speak of orbiting real ones) would make ever-more-detailed simulation worth the cost.

• Arts and cultural projects: There should be virtual exovivaria where users can create without fear, without risk to physical assets, and without undue costs. Exovivaria will have art. Exovivaria could even be art - indeed, this may end up being the most important thing about them. Good art requires taking risks. It's better if the risks are taken only with an audience's ability to appreciate art rather than with hard-won real-world assets. Virtual exovivaria can also be playgrounds and laboratories for social diversity, not least in governance - which of course can also be dangerous.

Practical limits

Virtual exovivaria will be different from most virtual worlds in important ways. (But see the caveats about avatars and augmentations, below.) The differences all relate to the goal of eventually hosting vital societies (virtual) and ecosystems (real) in orbit.

• Game physics - will need to match the artificial (or zero-) gravity of a proposed (or existing) exovivarium. In making artificial gravity in a rotating structure of modest size, Coriolis effects could be significant. There is also the Eötvös effect if there are fast-moving animals or telebots. Gravity will also vary depending on closeness to the axis of rotation.¹ Some of the games that users invent will be kinetic, and until there are exovivaria in orbit, the only way for the better players to keep their skills honed would be in accurate simulation.

• Variable realism - the paying users would initially want nearly photorealistic rendering -- they will want to know "what it's really going to be like." However, the serious players and the paid worker-avatars would tend to prefer a more schematic view. This would save on (simulated) satellite downlink costs (and computational costs), but it should also boost productivity. Some telepresence? studies have shown that workers are less distracted and feel greater "flow" (immersion) when shown only what's required for the job being done in the remote scene.

• Limited zoology/botany - Creatures and plants will be based solely on real species that actually die and decompose (or get eaten). Reproduction won't magically happen any sooner than normal mating/seeding patterns and gestation permit. Even species extinction will be possible.² Virtual animals should only be able to move as they could in the real world - albeit a real world with low gravity or micro-gravity. New species either can't be added, or can only be added at high expense.

• Variable interaction bandwidth - real exovivaria will probably be in low-earth-orbit (LEO). Such orbits are good when there are ground stations in contact with an exovivarium, because the speed-of-light delays will be imperceptibly short. However, it also means there will be times when there is no direct link with the ground - the Earth itself is in the way. For some given notional orbit, the simulated communication with the simulated environment would be degraded by a (likewise, simulated) round-trip delay through a geosynchronous satellite. The link could even be available only sporadically, as is the case with some LEO satellites, if it can't relay its messages to another satellite that's more Internet-accessible.

• Limited materials - Virtual worlds don't have any necessary limits on raw materials. And devices in those worlds can do magical things, if desired in the game. In an exovivarium, however, raw materials and components will be limited to what was initially supplied on virtual "launch". Resupply - if it's available at all - will be infrequent, expensive and possibly error-prone. Any simulated devices will have to behave as they would in a real-world exovivarium.

• Limited territory - Making new territory in most virtual worlds mainly means buying the needed computing power to allow a population within it to be animated, although the costs of rendering what's designed in the new space are also a factor. In exovivaria, making new territory will be either impossible (because the physical geometry of the simulated exovivarium is fixed) or very expensive (because it's expensive to put more things in space to make more surface area.)

An augmentable reality simulation

Within these simulated limits, however, almost any degree of "augmented reality" overlaid on the physical simulation should be permitted. Let's say someone wanted to pose their avatar seated on a simulated grasshopper perched on a simulated twig, waving a custom-designed flag, perhaps to "stake a claim" on some contested land. What's to stop them, if the API? permits it? Needless to say, users should have access to similar privileges in real terrestrial prototypes of exovivaria as well, and in any actual (i.e., orbiting) exovivaria, though only at higher prices, prices that some Project-internal market determines.

To be sure, the augmentations couldn't directly cause anything in the simulated physical environment. A user, through an avatar, should be able to direct a simulated telebot to pluck a leaf, but the avatar itself couldn't do the plucking.

Apart from this limit of "ghostliness" in the augmentations, however, allowing users to draw on any MMORPG? concept could be very useful in market-testing. Let's say that the augmentations grow to the point where they begin to cloud the simulated exovivarium. They might even reach the point where they reduce user interest in maintaining the ecosystem being simulated. That bad trend would be an excellent game-design diagnostic: it would say that the reality you're trying to simulate is not compelling enough, and that an exovivarium must have more of something, or else it will be abandoned.³

A future after the first orbit

If and when exovivaria become real, virtual exovivaria will probably still be useful. They may even become more popular than ever.

To own some real estate in a real exovivarium, perhaps bossing a few telebot and avatar crews to add value to one's property in orbit ... this should be the social pinnacle of Project Persephone society. But pinnacles invite climbers. Virtual exovivaria could form the foothills and lower slopes.

People need to start somewhere in any society, and in Project Persephone, many of them won't start with much money at all. Some will even start with the idea of making money in exovivaria that others have paid for - whether virtual, terrestrially prototyped, or orbiting. For the chance of getting to the point of making things happen in orbit (whether for fun or profit), many people wouild move first through the virtual exovivarial worlds, learning the ropes and the ethics. Then they can move onward to terrestrial analogues, learning yet more. After climbing a naturally-forming ladder of ascent, they can hope to eventually earn the right to do real work and own real things within real exovivaria. After all, real exovivaria are likely to remain relatively scarce for quite a while. Scarcity can sometimes do more than enhance the perception of value. It can also induce a sense of value into the steps and stages required to qualify for, or afford, the experience that's scarce.

¹ All three of these effects could be doubly significant if telebots are capable of hopping, as has been proposed for robots on the Moon and on Mars: Fraser Cain, "Hopping Microrobots", Universe Today, Dec 9, 2005 ⇑

² And necessary, in the case of otherwise-unmanageable pests or pathogens that snuck in somehow - scenarios that it would be irresponsible not to simulate. ⇑

³ The virtual world users' knowledge that it's a simulation, where things can't really die, could make any such diagnosis a lot less meaningful. But that's an important purpose of the more expensive terrestrial prototypes (where real living things can die): to calibrate the virtual worlds against real biophilia. Some interplay between the two design spaces should be expected. ⇑