| C. Laneve and B. Victor. Solos in concert. Full version of [10], submitted for journal publication, February 2001. |
....replicated input [20] We cannot use the same machinery because the solos calculus has no pre x operator. Instead, we use a de nition of the replication operator which is closer to a realistic implementation. We begin by showing that nested replication may be attened into non nested replications [11]: Theorem 5 (Flattening) ex) P j Q (y) ex) P j y e z) j ( e w) y e w j Qf e w=ezg) where e z = fn(Q) and y and e w are fresh. In view of this theorem there is no substantial loss of generality to only consider non nested replication. Therefore we from now on adopt the ....
....by fusing the nodes they connect. This interpretation lifts to the dynamics: cut elimination is exactly the edge edge reduction in SDs. The converse connection is complicated because solo diagrams are more expressive than Proof Nets. Nevertheless, it is possible to show that well sorted SDs (see [11]) with nodes occurring at most two times in the graph correspond to a superset of Proof Nets (the proof structures a collection of nets that may possibly be unsound [6] When boxes come into the picture the above relationship is more tenuous. The reason is that boxes are used di erently. In ....
C. Laneve and B. Victor. Solos in concert. Full version of [10], submitted for journal publication, February 2001.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC