Steven S. Popovich

Intelligent assistance for software development and maintenance

An environment is described, called Professor Marvel, that provides early error checking and answers questions about the program under development. The environment has a certain understanding of the systems being developed and how to use tools to produce software. It aids individual programmers and helps coordinate programming teams. The key components of Marvel are a database that stores data represented as objects, as in object-oriented languages, and a model of the development process that imposes a structure on programming activities. Marvels support of insight and of opportunistic processing is discussed at length, as is the handling of side effects. A sample session is described.<<ETX>>

A bi-level language for software process modeling

The authors present a multi-user implementation of a bi-level process modeling language (PML). Most process modeling formalisms are well-suited to one of two levels of specification, but not both. Some concentrate on global control flow and synchronization. These languages make it easy to define the broad outline of a process, but harder to refine the process by expressing constraints and policies on individual tools and data. Other process formalisms are inherently local. It is easy to define constraints, but far from straightforward to express control flow. Combining global and local formalisms is proposed to produce bi-level formalisms suitable for expressing the enacting large scale processes. The new PML is called the activity structures language. The activity structures language integrates global constrained expressions with local rules. Its implementation on top of the Marvel rule-based environment is described.<<ETX>>

Inquire: predicate-based use and reuse

There are four fundamental aspects of use and reuse in building systems from components: conceptualization, retrieval, selection and correct use. The most important barrier to use and reuse is that of conceptualization. The Inscape environment is a specification-based software development environment integrated by the constructive use of formal interface specifications. The purpose of the formal interface specifications and the semantic interconnections is to make explicit the invisible semantic dependencies that result in conventionally-built systems. The important ingredient provided by Inquire in conceptualization, retrieval, selection and use is the set of predicates that describe the semantics of the elements in the interface. These predicates define the abstractions that are germane to the module interface and describe the properties of data objects and the assumptions and results of operations in a module. Use and reuse of components is based on a components ability to provide needed semantics at a particular point in a system. It is the purpose of Inquire, the browser and predicate-based search mechanism, to aid both the environment and the user in the search for the components that will provide the desired predicates that are required to build and evolve an implementation correctly.<<ETX>>

MELDing Multiple Granularities of Parallelism

We are developing an experimental programming language, MELD, that suppons a range of concurrent styles by supporting multiple programming paradigms at multiple levels of granularity. MELD integrates three granularities of parallelism: macro dataflow among statements within a method or among methods for flne grain concurrency, synchronous or asynchronous message passing among local or remote objects for medium grain concurrency, and transactions for large grain concurrency among users.

A metalinguistic approach to process enactment extensibility

We present a model for developing rule based process servers with extensible syntax and semantics. New process enactment directives can be added to the syntax of the process modeling language, in which the process designer may specify specialized behavior for particular tasks or task segments. The process engine is peppered with callbacks to instance specific code in order to implement any new directives and to modify the default enactment behavior and the kind of assistance that the process centered environment provides to process participants. We realized our model in the Amber process server, and describe how we exploited Ambers extensibility to replace Ozs native process engine with Amber and to integrate the result with a mockup of TeamWare.

A flexible rule-chaining engine for process-based software engineering

We present the design of a new rule-based process engine that generalizes previous systems to support process enforcement, automation, guidance, monitoring, delegation, planning, simulation, instrumentation and potentially other applications. Our approach is fully knowledge-based, tailored by knowledge regarding the process assistance policies to be supported as well as the process definition.<<ETX>>

An object-based approach to implementing distributed concurrency control

Distributed concurrency control has been implemented by representing in-progress transactions as simulated objects, called transaction objects, that use normal message passing facilities. The implementation of an optimistic mechanism has been completed using transaction objects and a two-phase locking mechanism has been designed. The tradeoffs made and lessons learned, dealing with transactions both on objects and as objects, are discussed.<<ETX>>

AN ARCHITECTURAL SURVEY OF OBJECT MANAGEMENT SYSTEMS

Much work has been done in the last decade in the related areas of object-oriented programming languages and object-oriented databases. Researchers from both areas now seem to be working toward a common end, that of an object management system, or OMS. An OMS is constructed similarly to an OODB but provides a general purpose concurrent object-oriented programming language as well, complementing the OODB query facilities. In this paper, we will define several different types of object systems (object servers, persistent OOPL’s, OODB’s and OMS’s) in terms of their interfaces and capabilities from the viewpoint of how these support the requirements of cooperative information systems. We will examine the distinguishing features and general architecture of systems of each type in the light of a general model of OMS architecture.

Experiences with an environment generation system

The authors report on research experience using the Gandalf environment generation system as a prototyping vehicle for the Inscape environment. A Gandalf-based environment consists of four parts: a structure editor kernel, which is simply linked into each executable, a set of grammar tables describing the language to the kernel in terms of its abstract syntax, one or more concrete syntax views, and a collection of action routines written in the extension language, ARL. Positive aspects of the research included experimentation, incremental evolution, multiple views, the coupling of semantic and editing actions, and the use of domain-specific facilities. Negative aspects consisted primarily of problems with presentation and object management.<<ETX>>