30th
On KPN
The Kahn Principle states that each node in an
asynchronous deterministic network computes a continuous function
from input histories to output histories, and the behavior of the net-
work can be characterized as a least fixed point. Fairness plays a vital but implicit role: the Kahn Principle is only sound when network execution is assumed to be (weakly) fair. Kahn’s model
does not extend easily to non-deterministic networks, since the obvi-
ous generalization to continuous relations on histories is not composi-
tional . Previous attempts to model non-deterministic networks have sought to remain faithful to Kahn’s spirit by retaining some form of continuity assumption; these approaches typically apply only to a limited class of network and do not deal adequately with fairness
