Mike Stark
Goddard Space Flight Center
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Mike Stark.
ACM Sigada Ada Letters | 1987
Ed Seidewitz; Mike Stark
An object is an abstract software model of a problem domain entity. Objects are packages of both data and operations of that data (Goldberg 83, Booch 83). The Ada (tm) package construct is representative of this general notion of an object. Object-oriented design is the technique of using objects as the basic unit of modularity in systems design. The Software Engineering Laboratory at the Goddard Space Flight Center is currently involved in a pilot program to develop a flight dynamics simulator in Ada (approximately 40,000 statements) using object-oriented methods. Several authors have applied object-oriented concepts to Ada (e.g., Booch 83, Cherry 85). It was found that these methodologies are limited. As a result a more general approach was synthesized with allows a designer to apply powerful object-oriented principles to a wide range of applications and at all stages of design. An overview is provided of this approach. Further, how object-oriented design fits into the overall software life-cycle is considered.
Journal of Systems and Software | 1993
Shyam R. Chidamber; Robert L. Glass; Al Goerner; Mary Beth Rosson; Mike Stark; Iris Vessey
Abstract This article summarizes a workshop organized and conducted by the authors on April 1 and 2, 1993, under the auspices of Portland State University and the Oregon Center for Advanced Technology Education. This workshop provided a uniquely comprehensive and realistic look at the state of the art and practice of object-oriented programming. Issues discussed at the workshop and summarized here include the naturalness of object-oriented technology, reuse potential, programming support and evaluation tools, and the cost of converting to an object-oriented design technology. The findings should be of interest to researchers, organizations, and individuals who are exploring and/or using the object-oriented approach to software development. The article concludes by offering some recommendations for the use of objectoriented technology.
Journal of Systems and Software | 1993
Mike Stark
Abstract This article examines the premise that object-oriented technology (OOT) is the most significant technology ever examined by the Software Engineering Laboratory (SEL). The evolution of the use of OOT in the SEL “Experience Factory” is described in terms of the SELs original expectations, focusing on how successive generations of projects have used OOT. General conclusions are drawn on how the usage of the technology has evolved in this environment.
washington ada symposium | 1991
Ed Seidewitz; Mike Stark
A parameterized software system is one that can be configured by selecting generalized models and providing specific parameter values to fit those models into a general design [Stark 1990]. This is in contrast to the top-down development approach where a system is designed first, and software is reused only when it fits into the design. The concept of parameterized software is particularly useful in a development environment such as the Goddard Space Flight Center Flight Dynamics Division (FDD), where successive systems have similar characteristics.
tri-ada | 1989
Mike Stark; Eric W. Booth
The reuse of software holds the promise of increased productivity and reliability. Experience has shown that making even the slightest change to a “reused” piece of software can result in costly, unpredictable errors [Solomon. 19871. For this reason, the Flight Dynamics Division (FDD) of Goddard Space Flight Center (GSFC) is concentrating effort on developing “verbatim” reusable software Components with Ada, where verbatim means that no changes whatever are made to the component.
ACM Sigada Ada Letters | 1997
Sharon Waligora; John W. Bailey; Mike Stark
This paper presents the highlights and key findings of 10 years of use and study of Ada and object-oriented design in NASA Goddards Flight Dynamics Division (FDD). In 1985, the Software Engineering Laboratory (SEL) began investigating how the Ada language might apply to FDD software development projects. Although they began cautiously using Ada on only a few pilot projects, they expected that, if the Ada pilots showed promising results, the FDD would fully transition its entire development organization from FORTRAN to Ada within 10 years. However, 10 years later, the FDD still produced 80 percent of its software in FORTRAN and had begun using C and C++, despite positive results on Ada projects. This paper presents the final results of a SEL study to quantify the impact of Ada in the FDD, to determine why Ada has not flourished, and to recommend future directions regarding Ada. Project trends in both languages are examined as are external factors and cultural issues that affected the infusion of this technology. The detailed results of this study were published in a formal study report [1] in March of 1995. This paper supersedes the preliminary results of this study that were presented at the Eighteenth Annual Software Engineering Workshop in 1993 [2].
conference on object oriented programming systems languages and applications | 1994
Mike Stark; Ed Seidewitz
For the past five years, the Flight Dynamics Division (FDD) at NASA’s Goddard Space Flight Center has been carrying out a detailed domain analysis effort and is now beginning to implement Generalized Support Software (GSS) based on this analysis GSS is part of the larger Flight Dynamics Distributed System (FDDS), and is designed to run under the FDDS User Interface / Executive (UIX). The FDD is transitioning from a mainframe based environment to FDDS based systems running on engineering workstations The GSS will be a library of highly reusable components that may be conjigured within the standard FDDS architecture to quickly produce low-cost satellite ground support systems. The estimates for the first release is that this library will contain approximately 200,000 lines of code. The main driver for developing generalized software is development cost and schedule improvement. The goal is to ultimately have at least 80 percent of all software required for a spacecraft mission (within the domain supported by the GSS) to be configured from the generalized components. Domain Analysis
conference on object oriented programming systems languages and applications | 1993
Steven D. Litvintchouk; Mike Stark; Brad Balfour; Mohamed E. Fayad; Bernard Rosenfeld
The Software Engineering Laboratory (SEL), sponsored by the NASA Goddard Space Flight Center, was created to investigate the effectiveness of software engineering technologies. At SEL, the initial experience with object-oriented technology has been positive. However, the successes achieved thus far took seven years to achieve. In this panel, we will explore how typical the SEL’s experience is of large organizations doing large software developments.
washington ada symposium | 1987
Mike Stark; Ed Seidewitz
Archive | 1992
Linda Landis; Sharon Waligora; Frank E. McGarry; Rose Pajerski; Mike Stark; Kevin Orlin Johnson; Donna Cover