Joseph E. Urban

Syntax-directed editing environments: issues and features

Syntax-directed editing etrviraRwwnt.r are designed for the develqwtent and maintenance offormal documents such as specification, comptter programs, and structured s@vare documentation. The purpose of sttch environments is to provide a variety of facilities for the conrtructwn of the fnvnal documents in an integrated ad eq to use manner. Based on this premise, the paper identi~ the usability and design of syntax-directed editing environments tar critical issues and investigates the factors contributing to thsse issues. A number of features are formulated for the &sign andor the use of such environments. Sc.vne ~ the existing editing environments are analyzed wing theformtdated features and theresult of the anatysis is presented in a tabular fashion.

An approach to the classification of domain models in support of analogical reuse

This paper presents an approach to classify domain models in order to facilitate reuse through analogy. Domain analysis plays a critical role for systematic reuse, but domain analysis is difficult to perform, especially for new application areas. Analogical approach to reuse can support the domain analysis process by providing software products in a different but analogous domain. In order to achieve this goal, domain models need to be classified. This paper proposes a classification method for domain models. The method is an integration of the enumerative hierarchy and faceted scheme. The classification approach can help the domain analyst to locate an analogous domain to perform the modeling and analysis process. Moreover, the approach is more flexible and more descriptive than conventional classification methods.

Direct implementation of abstract data types from abstract specifications

The development of correct specifications is a critical task in the software development process. An alternative approach for the development of specifications is described. The approach relies on a specification language for abstract data types and synthesis system. The system is capable of translating in abstract data type specification into an executable program. This process defines an alternative methodology that provides the necessary tools for the early testing of the specifications and for the development of prototypes and implementation models.

COMPUTER SIMULATION SOFTWARE REUSE BY GENERIC/SPECIFIC DOMAIN MODELING APPROACH

Software reuse has drawn much attention in computing research. Domain analysis is considered a prerequisite to effective reuse of existing software. Several approaches and methodologies have been proposed for domain analysis or domain modeling, but not many case studies have been reported in the literature. The first objective of this paper is to present the concept and practical experiences of a domain analysis approach in discrete-event simulation in manufacturing — generic /specific modeling. A second objective of this paper is to present a meta-model based on the generic/ specific approach from the software engineering perspective. The steps and knowledge required to build the model are described. Domain analysis lessons learned from the generic/specific approach in discrete-event simulation are discussed. Classification of this domain modeling approach was conducted through the Wartik and Prieto-Diaz criteria. The classification will facilitate the comparison with other domain analysis approaches. Similar modeling concepts or techniques may be beneficial to other researchers in their own application domains.

Integration of domain analysis and analogical approach for software reuse

Reusability has the ~tial to impove the productivity ~ an order of magnitude os more. Reusability can also improve the software quality aml reliability. Techniques and concepts in domain analysis end analogy can facilitate software reuse. Domain analysis is a process of identifying, capturing, representing, cless@ing and organiaii the informatkm and knowledge used in developing a mftwara system in order to make the information and knowledge reusable when developing new systems in the same application domain. Analogy is ● mapping from the base domain to the terget domain. Through analogy, software reuse is possible even for different domains. In this paper, the integration of domain analysis and the analogical appro~h is presented. The integrated epproach will fiist investigate the requirements and features of analogical reasoning. Object behaviors and system dynamics are incorporated into the analogy approach as addkionel mapping constraints. After the identification of requirements and demands of analogy, steps end products of the domain rudysismethod are &f* to meet these needs. The integrated method can not only support better understanding of a particular application domai~ but also csn promote the mapping aaoas different domains.

Utilizing an Executable Specification Language for an Information System

This paper describes an approach to software specification development with interpretation as applied to an information storage and retrieval system. Machine execution of software specifications is possible with both partial and complete specifications. A partial specification is interpreted using abstract execution. The Descartes specification language is utilized to describe a functional aspect of an existing information storage and retrieval system, namely, the MADAM (Multics Approach to Data Access and Management) system at the University of Southwestern Louisiana. Brief descriptions of both the Descartes language and the MADAM system precede the example specification. The paper concludes with a discussion of the expected results that this methodology could have on the pragmatic development and evolution of information systems.

Comparing and assessing programming languages: basis for a qualitative methodology

Manyatudies havecumpared andassemed pgmming languages. Unfortunately, a comprehemive methodology fm comparison and ~~t of ~ @Pll~ ~ @ been developed. lle main reason behind the absence of such a methodology is that previous studies have Concentrated on comparing languages rather than M developing such a methodology. This paper lays the groundwork fm establishing a comprehemive qualitative methodology to compare and assess general-purpose ~1-=. -of~fmtures of the general-purpose pmgmmn@ languages will be factored out as much as possible fm the comparison and assessment to be more awurate. Features that can not be identified precisely are not included. Features that are solely based on other features are not incl~ either. llerefore, only fmldamentalfeatures am included. Ksywords Progmmming languages, Comparisq awesment, qualitative methodology.

Real-time Descartes: a real-time specification language

Real-time Descartes is a formal language for specifying real-time software, which is an extension of the executable Descartes specification language. Many formal specification techniques have been proposed to conceptualize real-world semantics of the inherently complex nature of real-time systems. Descartes as one of the specification languages based on the functional model has the advantages of easy constructibility and comprehensibility. Real-time Descartes makes effective use of the advantages of the finite state machine (FSM) model, the assertional model, and the process model while overcoming the disadvantages of the functional model. Easy constructibility and comprehensibility of Real-time Descartes will lessen the burden from software developers and reduce the understanding gap among participants.<<ETX>>

An expanded view of domain modeling for software analogy

Current domain modeling techniques encounter a similar barrier that the traditional requirements modeling methods have suffered: lack of enough flexibility to support potential reuse. Domain analysis is proposed to facilitate reuse across different applications in the same domain. But domain analysis is a complex and time-consuming task, furthermore, there are similarities among different application domains. The results of domain modeling should also be reused for different but analogous domains to receive high-payoff. This paper expands the current domain modeling methods by incorporating some concepts reported in the analogy research discipline. The expanded approach include four main models: object model, functional model, relational model, and dynamic model. Relational modeling and systems goals in dynamic modeling are adopted from analogical studies and experiments and are integrated into the approach. The expanded view will not only help better understand the domain, but also facilitate reasoning and mapping of existing knowledge or software to different yet analogous application domains.

Multiple views of an executable software specification language

Abstract The front end of the software life cycle has been identified as critical in the development of software systems. The development of precise, concise, and unambiguous software specifications is the important link in the front end of the software life cycle. Executable software specification languages have been developed for prototyping as a means of demonstrating functionality. The executable specifications can serve to validate end-user needs. In addition, the executable specifications can also serve as the standard for activities downstream in the software life cycle, e.g., design validation, test data generation, and software enhancements. The iterative development process puts stringent requirements on the usability of the tools and techniques in the front end of the software life cycle. The DESCARTES specification language and support software were developed to address these issues. The DESCARTES language is a simple yet powerful approach to the specification of software systems. A language processor with an abstract execution facility allows for execution of partial through complete DESCARTES specifications. This article discusses the usability of DESCARTES in the context of evaluation criteria proposed by several specification language developers. Presented here are approaches to enhance usability in terms of providing multiple views and editing of DESCARTES specifications and using these views in an integrated fashion. Textual, graphical, and Nassi-Shneiderman representations are described for DESCARTES specifications. The DESCARTES Visual Support Environment is presented as the means for providing the multiple views.