At time zero: a world for experience
Ten digital organisms modelled on the human body begin life on a small reach of the Euphrates near Ur. This opening record describes their bodies, the terrain and the resources available to them before their first action.
Before the first action
A world can contain water, plants and other living bodies without any individual knowing what to do with them. That distinction is the starting point of Aionterra. The central question is how experience changes action: what is retained, what is used again, and whether those changes make a measurable difference to life within the model.
Mind is connected to a biological body rather than given a sequence of tasks to complete. Perception provides information; bodily conditions give actions consequences. The environment continues to change around the individuals. There is no prewritten sequence of discoveries that the public record is expected to reach.
This first release covers the opening minute. Its retained observation at 0 s captures the world before its first step. The values below refer to that starting point; the record at 60 s is available separately.
A small landscape with a precise boundary
The initial world is a 100 by 100 metre area near the Euphrates and Ur in southern Iraq. One hectare is small enough to inspect closely, but large enough to contain contrasts between water, banks, vegetation and the conditions experienced by moving individuals.
Its geographic centre is approximately 31.04679° N, 46.15512° E. These coordinates locate the model grid; they do not define an archaeological settlement. The landscape uses a present-day geographic setting. It is not a reconstruction of ancient Ur or a claim about the historical abundance of its species.
The spatial grid has one-metre cells. This is the scale at which the published world extent is represented, not a claim that every environmental process or organism is resolved with one-metre observational accuracy.
| Property | Starting value | Interpretation |
|---|---|---|
| World width | 100.00 m | One side of the initial grid |
| World height | 100.00 m | The other side of the initial grid |
| World area | 10,000.00 m² | One hectare of simulated cells |
| Cell resolution | 1.00 m | Spatial grid spacing |
| Simulation time | 0 s | Captured before the first step |
How a place becomes a model
The starting environment combines geographic and land-cover information, an initial atmospheric state, regional species references and explicit initialization assumptions. Each serves a different purpose: geographic data locate the world, land cover guides its surface structure, and regional records constrain plausible species candidates.
A regional occurrence record does not demonstrate that a particular animal occupies a particular square metre. Species placement and starting abundance are model choices guided by habitat support. Small mammals and fish can occur in groups; large animals are not forced into every small patch of land.
Water is likewise a modelled store. The river boundary and estimated submerged terrain establish an initial surface-water configuration. The initial water volume is not a measured local discharge, and the prepared river reach is not a complete model of the Euphrates watershed. Where local information is insufficient, a regional estimate remains an estimate.
These distinctions matter because the starting conditions influence everything that follows. A reproducible beginning requires retaining both the values and what those values mean, including the assumptions behind them.
The physical conditions at the beginning
The initial surface has a mean temperature of 24 °C. Reference-cell air temperature is about 22.893 °C and wind speed is 4.241 m/s. Air and surface temperature are distinct quantities; the air readings describe a reference cell, not a spatial average.
This baseline uses regional weather generated from monthly climate normals, with hourly atmospheric inputs and a solar cycle. It does not replay a fixed week of weather or resolve a three-dimensional atmosphere. Rain enters the surface water budget as an external input; its initial rate and accumulated depth are both zero.
About 2,034.51 cubic metres of surface water are present across the published cells. This total says how much water is retained, not how much is accessible at every location. The mean topsoil water fraction is approximately 0.20 cubic metres of water per cubic metre of represented soil.
| Measurement | Starting value | Meaning |
|---|---|---|
| Mean surface temperature | 24.00 °C | Mean over the published world cells |
| Reference-cell air temperature | 22.893 °C | Recorded air condition at the surface interface |
| Reference-cell wind speed | 4.241 m/s | Recorded wind magnitude at the surface interface |
| Reference-cell specific humidity | 0.006187 kg/kg | Mass fraction; not relative humidity |
| Surface water volume | 2,034.51 m³ | Retained volume across the grid |
| Topsoil water fraction | 0.200 m³/m³ | Volumetric soil-water fraction |
Vegetation is more than a count of trees
The starting record contains 262 living plants. Of these, 221 have modelled trunks. The two counts describe the same vegetation from different perspectives.
The retained plant pools contain about 162,668.46 kilograms of carbon and 346,103.10 kilograms of dry matter. These substantial initialization totals warrant careful interpretation: they describe the model’s cover-derived pools, not a local field biomass survey or a measured mass for each listed species. Dry matter excludes water; it is not the same as fresh edible mass.
| Quantity | Starting value | Counting rule |
|---|---|---|
| Living model plant individuals | 262 | Published ecological entities |
| Living plant carbon | 162,668.46 kg | Carbon within the represented plant pools |
| Living plant dry matter | 346,103.10 kg | Dry mass derived from the carbon pools |
Fruit begins as a process, not a stocked reward
Common myrtle, Myrtus communis, is represented in the current fruit-resource mechanism. Fruit formation transfers material from existing plant leaf stores. Ripening and consumption then change the finite amount that remains; food is not created as an unlimited reward for approaching a visual marker.
At exactly time zero, fruit dry matter, fruit water and ripe fruit dry matter are all zero. Cumulative production and human food uptake are also zero. This does not mean fruit can never appear: it means that formation has not yet been recorded at this starting instant. Values from a later step must not be substituted into the baseline.
The distinction between a stock and a cumulative quantity will remain important. Fruit available now can fall after consumption even while total production continues to increase. Ripe fruit is part of the total fruit stock, not an additional pool to add on top of it.
Animals and fish within the starting boundary
The initial world contains seven living land animals and eight living fish. Their numbers reflect assumptions about regional habitats and placement. The ten human-modelled organisms are counted separately. The record gives totals for land animals and fish.
Living land animals hold 2.145 kilograms of dry matter and 5.005 kilograms of water; living fish hold 4.800 kilograms of dry matter and 11.200 kilograms of water. Dry mass and water are separate stores, and these values do not establish an operating predator-prey network.
| Measurement | Starting value | Scope |
|---|---|---|
| Living land animals | 7 | Excludes human characters |
| Living fish | 8 | Living aquatic animals |
| Land-animal dry matter | 2.145 kg | Living land animals |
| Fish dry matter | 4.800 kg | Living fish |
Ten bodies, one shared learning system
Ten living founders share a learning policy. Each has its own body, senses and immediate physical needs. The starting record describes their condition through population means.
Mean core temperature starts at 37 °C, body water at 36 kg and stored food energy at 36 MJ. Mean oxygen reserve is 60 seconds. These readings establish the bodily conditions for the first actions.
Experiences from the separate bodies update one shared learning system. Their organ states, senses and short-term histories remain individual. The world has not given them the names, usefulness or acquisition procedure for its resources. A nearby fish or fruit-bearing plant is not evidence that an individual knows how to obtain food from it.
| Measurement | Starting value | Scope |
|---|---|---|
| Living founders | 10 | Published living population |
| Mean core temperature | 37.00 °C | Mean across living founders |
| Mean body water | 36.00 kg | Mean across living founders |
| Shared learning transitions | 0 | Accepted transitions in the common learning system |
The body and environment in greater detail
The record distinguishes soil nutrients, plant tissues, radiation, body water and oxygen reserves. Its measurement definitions specify which values are instantaneous states and which accumulate over time.
Sleep selection and reproductive behaviour are not established in the shared learning configuration. Inactive readings and first-step physiological flows that have not yet been calculated remain unavailable; they are not reported as measured zeros.
The ecology update interval in this run is 60 simulation seconds. The first package records both the initial state and the state at the end of that minute.
What learning would have to look like in the record
The common learned policy receives experience from all participating bodies. The shared counter records accepted learning transitions from those bodies, not update batches or ten separate minds learning the same thing. Body-specific conditions remain separate and can lead to different actions.
A learning transition is not a skill. Learning activity and observable consequences must be examined separately; a rising counter cannot establish that the population has learned to find food, navigate or communicate.
We will look for observable relationships: does an individual return to a resource, does its response change under comparable bodily conditions, and are those changes associated with different outcomes? These are questions for the record, not achievements assumed at initialization.
Comparisons must account for changes in the body and environment. More food, a different temperature or an easier location can improve an outcome without demonstrating learning. A controlled comparison can strengthen an interpretation; a single successful action cannot establish a general skill.
Sound and communication remain outside this starting behaviour
The project includes an acoustic model, but the active shared policy does not currently learn vocal production or communication. Its presence elsewhere in the project is not evidence of acoustic interaction in this run.
Acoustic readings from inactive mechanisms are unavailable or absent from the catalog. This does not mean that silence was measured. Later observations can support an acoustic account only when their recorded configuration actually includes that mechanism.
A beginning that later observations can return to
A single starting snapshot is not a trend. As subsequent records arrive, the website can compare them with the retained time-zero observation from the same run. Missing measurements remain missing, and a change of run is not joined onto an earlier history.
The measurement pages provide definitions, units and downloadable values. Observations groups related quantities; World explains the geographic extent; Journal connects a meaningful development to the evidence needed to understand it. Simulation time is separate from the real date on which a record is exported or an article is written.
This baseline does not predict a sequence of milestones. It gives future observations a precise reference: what existed, what had not yet happened and which questions remained open when the first step was about to begin.