R. Gerth and A. Pnueli. Rooting UNITY. In Proceedings Fifth International Workshop on Software Specification and Design, Pittsburgh, Penn., May 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....Guards could determine the whole sequence of the computation, even sequentializing it by having guards of the form step = i, but the intent of the model is rather to use the weakest guards, and therefore say the least, about how the pieces are to be fitted together. This idea lies behind UNITY [29, 49, 95, 168], and an alternative that considers independence of statements more: action systems [12 15] UNITY (Unbounded Nondeterministic Iterative Transformations) is both a computational model and a proof system. A UNITY program consists of a declaration of variables, a specification of their initial ....

R. Gerth and A. Pnueli. Rooting UNITY. In Proceedings Fifth International Workshop on Software Specification and Design, Pittsburgh, Penn., May 1989.

.... way as Sanders [14] soundness and completeness are derived from an embedding of the Unity logic into Gerth Pnueli s temporal logic de ned for a class of programs called SLP (single location programs) and proved sound and relatively complete for the operators U (weak until) and 3 (eventually) [8]. In order to provide a formal background for proving inter reducibility of proofs of unless and U , and leadsto and 3, we de ne a uniform semantical framework for both Unity and SLP programs and show that for each Unity program there exists a semantically equivalent SLP program. Our paper is ....

....to the program semantics. The determination of a suOEcient set of axioms and rules relies on a clear characterization of program properties and the role of the initial conditions on them. To prove soundness and relative completeness of the system SI , we explore results of Gerth and Pnueli [8] that consists of a sound and relatively complete system SGP for the SLP programs. We proceed as follows : rst, we translate Unity programs into SLP, then we de ne an embedding of proofs in SI into proof in SGP and conversely. Unity programs form a subclass of SLP programs Theorem 12. For ....

[Article contains additional citation context not shown here]

R. Gerth and A. Pnueli. Rooting UNITY. In acm, editor, Int. Workshop on Software Speciøcation and Design, pages 1119, 1989.

....true, it is necessary to ensure that q becomes true. Once p and q are both true, q must remain true until p is falsified; otherwise the progress obligation is not discharged. 7. 2 Relationship to UNITY Theory A number of completeness proofs for the UNITY logic have been given in the literature [15, 16, 17, 9]. Althougth the underlying logic described in Section 2 differs in several ways from the UNITY logic, the differences do not affect completeness. The most obvious difference is the extra inference rule, PSP (which stands for ProgressSafety Progress) This rule is derived as a metatheorem in the ....

Rob Gerth and Amir Pnueli. Rooting UNITY. In Proceedings of the Fifth International Workshop on Software Specification and Design, Pittsburgh, May 1989.

....proof steps, and is (relatively) complete, even though it contains fewer proof rules than other temporal logics. This suggests that Unity is simpler and that proofs are more straightforward, since one has fewer necessary tools to work with. The soundness and completeness of Unity is discussed in[24, 18, 10, 14]. 2.2 The Boyer Moore Logic and Prover In a mechanically verified proof, all proof steps are validated by a computer program called a theorem prover. Hence, whether a mechanically verified proof is correct is really a question of whether the theorem prover is sound. This question, which may be ....

Rob Gerth and Amir Pnueli. Rooting unity. In Fifth International Workshop on Software Specification and Design, volume 14(3), pages 11--19. ACM Sigsoft Engineering Notes, 1989.

....of algorithms, to achieve a safe design, if not a systematic one; make it easier to reason about distributed systems by proposing an efficient (i.e. abstract and concise) representation of such systems. In this purpose, we use the Unity formalism [2] which provides a sound formal framework [7, 8] and appears to be well suited to describing distributed systems. Our contribution provides an abstraction of interactions between sites. The construct we propose is intended to be in the spirit of Unity minimalism: it solely aims at expressing what is according to us the essential feature ....

Rob Gerth and Amir Pnueli. Rooting UNITY. In Proc. fifth Int. Workshop on Software Specification and Design, pages 11--19, May 1989.

....and that Unity is less expressive than full first order temporal logic. Unity provides proof rules for taking large formal proof steps, and is (relatively) complete, even though it contains fewer proof rules than other temporal logics. The soundness and completeness of Unity are discussed in [5,9,11]. To see how we use Unity to model storage jamming, let us return to our example jammer, now modeled in Unity logic as program wj. We begin by describing the file system as an array of files and the user input as a tuple that contains the intended word processor operation and the name of its ....

GERTH, R. and PNUELI, A. Rooting UNITY. In Proceedings of Fifth International Workshop on Software Specification and Design, ACM SIGSOFT Engineering Notes, 14, 3, 1989, 11-19.

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