Artificial life was defined by Langton (in [3]) as "the study of man-made systems that exhibit behaviors characteristic of natural living systems".
There are two goals behind the conception and study of artificial life.
The first one is to try and understand more about life itself. It is possible
for a biologist to study life-as-it-is, by observing the organisms present
around us. But biology is very much limited to observation. When a physicist
might observe the world, create a theory to explain its workings, and then
begin making experiments, modifying the conditions to see if the theory
remains valid, a biologist has very few possibilities to experiment with
life. Furthermore, even observation is limited to the life we know. A biologist
cannot observe life as it existed 10 million years ago, or as it will exist
in 10 million years on Earth, for example. Fossils are a poor substitute
to live specimen. To be fully able to study life, a biologist would have
to be able to create worlds and observe life on them during a very long
time.
Creating new worlds with different life forms is perhaps not so impossible
as it may seem. A computer simulation of these other worlds would be easy
to create and to manipulate (at least much more so than a set of new planets).
And life could be generalized to other forms than the one we know (based
on carbon and water), further enlarging the field of study for biologists.
The beginnings of artificial life led to a lot of questions about the
nature of life itself. To create life, quite often in a medium rather different
from the usual one (software for example), one must be able to recognize
life.
A list of properties that are characteristic of life as we know it was
established (see [3]), but the boundary between
life and "not-life" remains imprecise. It is hoped that further
research and experimentation with artificial life forms will bring a more
precise, more comprehensive definition of life.
The second goal of artificial life is somewhat more practical. It would
be an attempt to use the new (life) forms created by artificial life for
technical applications.
These applications may exist in very different fields. Artificial life
has been applied to three main media: "wetware", or carbon-based
life-forms, hardware, and software.
The practical applications for wetware are the most evident. Through genetic
engineering many new species of plants and animals (to a lesser degree)
have been created by scientists. Amongst hardware applications of artificial
life was a feasibility study for the NASA about the possibility of building
self-repairing and relatively independent space probes. Software makes
for easy manipulation. Possibilities for this medium range from the optimization
of existing software to the creation of an artificial brain.