Adaptive Erkennung von Software-Entwurfsm?ngeln

Zusammenfassung Die Qualit?t von Software kann je nach Anwendungsgebiet an unterschiedlichen Kriterien gemessen werden. Für grosse Software-Systeme spielen u. a. Kriterien wie Wartbarkeit, Verst?ndlichkeit und Erweiterbarkeit eine wichtige Rolle. Mein Ziel ist es, Entwurfsm?ngel in Software-Systemen zu erkennen und somit „‘schlechte“’ – unverst?ndliche, schwer erweiter- und ?nderbare – Programmstrukturen zu vermeiden. Prominente Entwurfsm?ngel sind z. B. die von Fowler eingeführten Bad Smells in objektorientierten Programmen. Entwurfsm?ngel werden abh?ngig von der Sichtweise und dem Erfahrungsschatz des Suchenden unterschiedlich interpretiert. Ich kombiniere daher bekannte Verfahren zur Erkennung von Entwurfsm?ngeln auf der Basis von Metriken mit maschinellen Lernverfahren. Damit kann Entwurfsmangelerkennung individuellen Sichtweisen angepasst werden. Ich pr?sentiere ein Werkzeug für Java-Programme und zeige die Ergebnisse einer ersten Fallstudie.

---

Criteria for software quality measurement depend on the application area. In large software systems criteria like maintainability, comprehensibility and extensibility play an important role. My aim is to identify design flaws in software systems automatically and thus to avoid “bad” – incomprehensible, hardly expandable and changeable – program structures. Depending on the perception and experience of the searching engineer, design flaws are interpreted in a different way. I propose to combine known methods for finding design flaws on the basis of metrics with machine learning mechanisms, such that design flaw detection is adaptable to different views. This paper presents the underlying method, describes an analysis tool for Java programs and shows results of an initial case study.

CR Subject Classification D.1.5,D.2.2,D.2.5,D.2.7,D.2.8,I.2.6     

Aufbau und Realisierung des MeDoc-Volltextspeichers

Zusammenfassung.   Ziel von MeDoc war der Aufbau eines verteilten Informationsdienstes, über den eine kritische Masse an Informatik-Literatur im Volltext nachgewiesen, langfristig vorgehalten, recherchiert und beschafft werden kann. Bei einem Teil der angebotenen Literatur handelt es sich dabei um elektronische Versionen des regul?ren kommerziellen Angebots der beteiligten Verlage. Aus diesem Grund war ein wichtiges Teilziel von MeDoc die Entwicklung eines Volltextspeichers, der solche kostenpflichtigen Dokumente abrechenbar und sicher nachweist, liefert und zum Browsen zur Verfügung stellt. Der Artikel beschreibt die Konzeption und Architektur dieses Volltextspeichers. Eingegangen am 17. April 1998 / Angenommen am 21. Juli 1998     

Editorial

Ohne Zusammenfassung

---

Editorial

     

Innovation processes in large technical systems: Market liberalization as a driver for radical change?

Electricity supply is a large technical system, which exhibits strong path dependencies and high barriers for radical innovations. Recent market liberalization, however, has initiated a fundamental restructuring. The paper analyzes how liberalization has altered innovation processes in the field of electricity supply. We examine three radical innovations under monopoly conditions and contrast the findings with the results from a survey on the innovation behavior of electric utilities in liberalized markets. We argue that the selection environment for innovations has changed in various respects. In our sample, new options open to innovation activities have emerged at the level of the firm and utility strategies turn out to be more heterogeneous. This may be interpreted as an increase in the variety of search processes, which may lead to a mid-term decrease of path-dependencies.     

From apprentice to colleague: The metamorphosis of Early Career Researchers

While the studies of Early Career Researchers (ECRs) have contributed politically important insights into factors hindering ECRs, they have not yet achieved a theoretical understanding of the causal mechanisms that are at work in the transition from dependent to independent research. This paper positions the early career phase in a theoretical framework that combines approaches from the sociology of science and organisational sociology and emphasises the transitional process. In this framework, the early career phase is considered as containing a status passage from the apprentice to the colleague state of their career in their scientific communities. In order to capture the mechanisms underlying this transition, it is important to analyse the interactions of these careers as they unfold over time. The usefulness of this approach is demonstrated with a pilot study of Australian ECRs. We show (a) that misalignments of the three careers stretch the transition phase; (b) that the two major factors affecting the transition are a successful PhD and a research-intensive phase prior to normal academic employment; and (c) that the most important condition hindering the transition is the lack of time for research. It can be concluded that as a result of a ‘market failure’ of the university system, the transition from dependent to independent research is currently being relocated to a phase between the PhD and the first academic position.