Starting about 30+ years ago, we developed neural networks that simulated neurons by connecting individual programs through interprocess communications where each program represented an individual neuron. On many attempts over the years, we discovered machine limitations on the number of processes that can run concurrently on a single processor.
WIth the advent of 64-bit machines and the use of the Internet, we were able to overcome the limitations for the past. 64-bit machines expand the number of processes that can run on a single processor considerably and using UDP, we are able to create asynchonous communications between individual processes.
Providing the Connectome Framework, we can allow anyone to create an instance of a connectome, connect sensory data to the instance and allow the user to define the output. The classic approach is robotics where sensory data is collected by robotic sensors (touch, light, IR, etc) and the motor output are physical motors.