It is hoped that the more complex and dynamic environment presented by the network will challenge evolution with a more complex problem than the basic string copy of the single node installation. This should provide selective forces to initiate an increase in complexity. Once a significant impulse in the direction of complexity has occurred, the hope is that selective forces arising from interactions among the digital organisms can lead to an auto-catalytic increase in complexity.
It appears that this is what happened in organic evolution. In the Amazon region, there are rain forests on white sand soils, where the physical environment consists of clean white sand, air, falling water and sunshine. Embedded in this physical environment is the most complex ecosystem on earth, the tropical rain forest. In this ecosystem there are hundreds of thousands of species. These do not represent hundreds of thousands of adaptations to the physical environment, but rather, most of the adaptations of these species are with respect to the other living organisms that they interact with.
Life transforms the environment, such that the living component of the environment comes to predominate over the physical environment, after which most evolution involves adaptations to other living organisms. Thus the complexity of the living component of the environment comes to greatly exceed the complexity of the physical environment that it is embedded in.
The first evolution observed in Tierra was the origin of ecological interactions, which were based on adaptation to the presence of other digital organisms in the environment. Thus this dynamic has been present in Tierra from the beginning. It is hoped that with the help of some impulse towards greater complexity, this dynamic can lead to a large spiraling upwards in complexity.
Returning to the case of network Tierra, consider that if we introduce a single network algorithm into the network, which surveys conditions and makes a decision about where to migrate, it may happen that initially all digital organisms within the net use the same algorithm and make the same decisions, and attempt to migrate to the same location. This would cause the chosen location to be flooded, and it would no longer be a desirable place to be. The digital organisms need to develop strategies to distribute themselves in a rational manner throughout the network. In essence, this would require the evolution of some kind of social behaviors, which would at least be a kind of flocking or anti-flocking behavior. However, this problem of distribution on the net might be solved best through a diversity of strategies, rather than by a single very smart strategy.