Naive denotational semantics. Information Processing 83, Proc. IFIP Congress'83, Paris 1983, ed. R. Mason, 345-355, North-Holland 1983.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....required. To cope with the problem Scott developed a theory of reflexive domains. It was noticed, however, that the need for solving type equations like (1) does not arise when it comes to consider the denotational semantics of most modern programming languages, like Pascal, Modula, C, etc. In [BT83] it was argued that in most cases one can adopt the naive view rather than the mathematically sophisticated machinery of reflexive domains. One has to abandon, however, type definitions like (1) in which the non continuous operators appear in the essentially recursive way. 1.2 Naive Denotational ....

....in the essentially recursive way. 1.2 Naive Denotational Semantics In the MetaSoft project we have faced the problem of defining a naive denotational semantics of the kernel language of the system. From this perspective the MetaSoft project provided a testing ground for the ideas presented in [BT83]. The preliminary results of this experiment are presented in [BBP90] These results justify expectations expressed in [BT83] at least to some extent. One can indeed give a set theoretical semantics for a functional language with a syntactically restricted form of recursive definitions of both ....

[Article contains additional citation context not shown here]

Andrzej J. Blikle and Andrzej Tarlecki. Naive denotational semantics. In Proceedings IFIP Congress 1983. North Holland, 1983.

....are denotational , by which we mean the compositionality of semantics. We do not use, however, the classical concepts of Oxford style denotational semantics, such as reflexive domains, cf. Scott 76] or information systems, cf. Scott 82] Our semantic domains are just sets as suggested in [Blikle, Tarlecki 83] To some extend MetaSoft is a descendant of VDM (cf. Bj rner, Jones 78] Bj rner, Jones 82] or [Jones 86] VDM is a system where set theoretic denotational models of software are developed in a constructive, continuation free style. On the other hand, however, MetaSoft has been largely ....

Blikle, A. and A. Tarlecki. "Naive Denotational Semantics". Information Processing'83 (Proceedings IFIP Congress 1983), North Holland, 1983.

....a label is an effect message. The responsibility of propagating effects rests with the composition function for statements and other glue functions. When a label reaches the top level, an exception handler that corresponds to our administrator effects the appropriate jump. Blikle and Tarleki [4] showed that by using the VDM technique, denotational specifications can use sets instead of domains and partial functions, instead of continuous ones. The Vienna School has not applied this technique to the specification of other imperative language constructs [D. Bj rner; letter, June 5, 1990] ....

Blikle, A. and A. Tarlecki. Naive denotational semantics. In Proc. IFIP 9th World Computer Congress: Information Processing 83, NorthHolland, Amsterdam, 1983, 345-355.

....of which are assigned to objects of the language as their meanings. A naive attempt might have been as follows: V alues = Algebras j Pow (V alues) j V alues e V alues Clearly, this leads to serious foundational problems, as the recursive domain definition involves heavy recursion (cf. BT 83] and hence cannot have a set theoretic solution (even assuming that we consider here a set al..gebras of algebras built within a fixed universe) However, since the formalism we introduce is not intended to cater for any form of self application of functions or non well foundedness of sets, the ....

....of the language can easily be seen to form a hierarchy indexed by types (see Section 7. 4) Thus, we can define a corresponding cumulative hierarchy of sets of values, and then define the domain of the meanings of objects as the union of sets in the hierarchy, much in the style of [BKS 88] see [BT 83] where the idea of using hierarchies of domains in denotational semantics is discussed in more detail) Another, less constructive , possibility is to work within a fixed universal set of values of objects containing the set of all algebras [Coh 81] 7.3 Proving satisfaction We are ....

Blikle, A. and Tarlecki, A. Naive denotational semantics. Information Processing 83, Proc. IFIP Congress'83, Paris, ed. R. Mason, 345--355, North-Holland 1983.

....of which are assigned to objects of the language as their meanings. A naive attempt might have been as follows: V alues = Algebras j Pow (V alues) j V alues e V alues Clearly, this leads to serious foundational problems, as the recursive domain definition involves heavy recursion (cf. BT 83] and hence cannot have a set theoretic solution (even assuming that we consider here a set al..gebras of algebras built within a fixed universe) However, since the formalism we introduce is not intended to cater for any form of self application of functions or non well foundedness of sets, the ....

.... of the language can easily be seen to form a hierarchy indexed by types (see Section 6) Thus, we can define a corresponding cumulative hierarchy of sets of values, and then define the domain of the meanings of objects as the union of sets in the hierarchy, much in the style of [BKS 88] see [BT 83] where the idea of using hierarchies of domains in denotational semantics is discussed in more detail) Another, less constructive , possibility is to work within a fixed universal set of values of objects containing the set of all algebras [Coh 81] 5 Proving satisfaction We are interested ....

A. Blikle and A. Tarlecki. Naive denotational semantics. Information Processing 83, Proc. IFIP Congress '83 (ed. R. Mason), Paris. North-Holland, 345--355 (1983).

....of which are assigned to objects of the language as their meanings. A naive attempt might have been as follows: V alues = Algebras j Pow (V alues) j V alues e V alues Clearly, this leads to serious foundational problems, as the recursive domain definition involves heavy recursion (cf. BT 83] and hence cannot have a set theoretic solution (even assuming that we consider here a set al..gebras of algebras built within a fixed universe) However, since the formalism we introduce is not intended to cater for any form of self application of functions or non well foundedness of sets, the ....

....of the language can easily be seen to form a hierarchy indexed by types (see Section 7. 4) Thus, we can define a corresponding cumulative hierarchy of sets of values, and then define the domain of the meanings of objects as the union of sets in the hierarchy, much in the style of [BKS 88] see [BT 83] where the idea of using hierarchies of domains in denotational semantics is discussed in more detail) Another, less constructive , possibility is to work within a fixed universal set of values of objects containing the set of all algebras [Coh 81] 7.3 Proving satisfaction We are ....

Blikle, A. and Tarlecki, A. Naive denotational semantics. Information Processing 83, Proc. IFIP Congress'83, Paris, ed. R. Mason, 345--355, North-Holland 1983.

No context found.

Naive denotational semantics. Information Processing 83, Proc. IFIP Congress'83, Paris 1983, ed. R. Mason, 345-355, North-Holland 1983.

No context found.

Naive denotational semantics. Information Processing 83, Proc. IFIP Congress'83, Paris 1983, ed. R. Mason, 345--355, NorthHolland 1983.

No context found.

Andrzej Blikle and Andrzej Tarlecki. Naive Denotational Semantics. In R.E.A. Mason, editor, Information Processing 83, pages 345--355, IFIP, 1983.

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