Phases: Life on a Disk

Wrote for Phases. 4 days late.

My head hurts and I'm driving myself crazy again. Had a miserable day for no good reason. I hate how I can make myself scared of anything so easily. Fear is the mind-killer. Perfect love casts out fear. Remember those things.

---

Caution: This story contains links to pictures of cute animals. Advise activation of So-Cute (© Algene Tan) alarm.

I discovered the Creatures series back in 1997. At that time, one of my family’s favourite group activities was window-shopping in computer outlets. I was going through a phase of fascination with artificial life (and Star Trek, but let’s not talk about that) in my mid-teens, and so I liked the idea of a game that let you play with creatures that actually reproduced, foraged, played, and had brains, unlike the abhorrent Tamagotchis then popular. They could even talk to each other! Much to my disappointment, my family’s computer could only run it at snail’s pace.

Computing power was what really allowed the explosion of artificial life as a field of study – creatures created as software objects, following rules that while simple, could produce surprisingly life-like behaviours. However, people were experimenting with it even before that, such as John Conway, the inventor of the aptly named Game of Life in 1970. He and his students “ran” it on big sheets of graph paper with board game pieces. The rules were as follows:

• if a cell was surrounded by less than two other cells, it would die from exposure.
• if a cell was surrounded by more than three other cells, it would die from overcrowding.
• if an empty square was surrounded by exactly three other cells, it would become a cell.

You can try it for yourself here, or if you’d rather do it the original way, any Reversi or checkers board will do. (Hint: nice way to waste time during maths.)

From these few rules emerged populations that grew, crawled around like amoebae on a glass slide, and then – but not always - died. Sometimes oscillating shapes formed that would cycle through the same pattern over and over again. One of the stranger shapes that emerged was a glider, an arrangement of five cells that after a period of four time-steps, would reorganize in the same shape – but having moved across the board. Over time, Life fans made discoveries about the world of Life and clever patterns, including this “humongous” spaceship.

Around that time other people began playing around with computers to see if they could write instructions for the computer that would be like the instructions that nature lays down for the growth of plants. A living plant decides the angle of its stem, where it puts out branches, the shape of its leaves and so forth by hormones activating parts of its DNA saying things like, “We’ve got a bud here giving off hormone X, so don’t start another bud so close, start further down.” So again, a living thing that exists in a certain state which changes over time according to rules. Imagine a system where the letter a is replaced with ab, and b is replaced with a. Start with a seed b:
b
a
ab
aba
abaab
abaababa
abaababaabaab

Now, imagine that instead of letters, each a was a stem and each b was a branch. Here’s an example. One of the best-known ways of doing this was with L-systems, which operated in a way similar to the above example. Not only did this generate very pretty pictures, it also showed another way in which life could be simulated on a computer – not the mad scramble of microbes, but the meditative unfolding of leaves to the sun.

As computers got more powerful, the simulations that could be made became more complex. After cellular automata like Life came programs that tried to reconstruct the behavior of actual animals moving about in real time such as ants, fish, and birds. Craig Reynolds’s Boids are an example. Each Boid tries to maintain a certain distance from its neighbours while keeping close to the flock and going in the same direction, as real birds tend to do. The result is a flock that swoops around obstacles as smoothly as seagulls. Similar principles began to be used in the 1990s to animate large swarms of moving objects, such as the wildebeest stampede in The Lion King.

This brings our history of a-life up to my old love, Creatures. My family’s old computer could just barely run it, albeit at a snail’s pace, so I dropped it impatiently until I had my own laptop in college and bought Creatures 3. One of the reasons for this game’s popularity is that it can be played on two levels – either as a cute virtual pet game, or as a tool for experimenting with both artificial life (through the smaller animals in the game, which can be edited in the game’s CAOS language) and artificial intelligence. The behaviour of the Creatures isn’t controlled from the top down by rules – it’s determined from the bottom up by their brains, each of which has several hundred neurons which fire according to inputs from each other and the environment. (Multiply this by several Creatures, add a couple hundred small animals and machines, and keep in mind that the game world takes place in real time, and you’ll see why it doesn’t run well on a 386.) You can download Docking Station for free, which is a fully functional mini-world using the same engine as C3.

This sort of thing – the general population getting access and gaining interest in tools that mathematicians, computer scientists, ecologists, botanists, and neuroscientists have been playing with for the last few decades, is going to become more popular as biology takes over as the dominant science in the public eye, what with bird flu and stem cells and genetic engineering in the news all the time. The next game I’m hoping to be able to dive into as demiurge is Will Wright’s Spore (caution, Flash-heavy site). Wright was the man behind Sim City and a lot of the other Sim games, too. Soon he’ll be coming out with a game where you, the player, can guide the evolution of a

I’ve learned something from playing Creatures – if you’re truly interested in something, you have to give it the freedom to grow and learn on its own rather than manipulating it into what you want it to be. Even if this means listening to a Creature complain it’s hungry for five minutes with a pile of carrots on the other side of the screen. Otherwise you might as well be feeding a Tamagotchi and picking up its poo.

A-life and other simulation games, like the abovementioned Sim City, are different from others where there’s a concrete goal, like a mission objective or a score. You play not as an actor, but as a creator. You experience joy when your creatures are eating well and reproducing; frustrations when they don’t follow your plans, to the point of being suicidally stupid; despair when you realize something is unexpectedly wrong and a disaster is going to befall those little sprites on the screen. It’s upsetting, even though they’re not real and you could kill them all with a couple of mouse clicks if you liked. As with all games, there’s an emotional investment in the game, but with this kind it’s deeper, because the characters are both under your control and out of your hands. There’s an affection for them; if you play enough, maybe even a love.