| |
|
| |
July 2007
Joint SQRL and HSE West Workshop: "Software for Support of Early Childhood Intervention".
|
May 2007
“Improving the Quality of Software by Keeping Documentation and Implementation Consistent”
Gurcan Gulesir from the Software Engineering Group, University of Twente, The Netherlands.
|
Abstract
The size and complexity of modern software systems are increasing. As a result, more effort is spent for developing and maintaining such systems, and more errors are made during the development and maintenance.
To reduce the amount of effort and the number of errors, we propose a solution that is composed of (a) a visual, intuitive, and formal language for documenting the design of software behavior, (b) an easy-to-use tool that can check whether the design and the implementation are consistent, and (c) an efficient and effective way of working to use the language and the tool.
The proposed solution was applied in the context of a large-scale embedded system, and a controlled experiment was conducted with 23 professional developers and 21 M.Sc. students. Benefits were measured in terms of effort reduction (76%) and error prevention (one error per 140 lines of source code). Based on these results, the company that produces the large-scale embedded system decided to use the solution during software development and maintenance.
Enquiries: Michelle Lanigan ext 3572
Further Details: www.sqrl.ul.ie
|
May 2007
The Aspect-Oriented Language Compose. Its characteristics, aspect interference detection, language independency and "semantic pointcuts".
Presenter: Prof Mehmet Aksit
Chair of Software Engineering from the University of Twente, The Netherlands.
|
Abstract
The Composition-Filters (CFs) is one of the first aspect-oriented language concept; it was defined as early as end 80's. However, like all practical languages, CFs has gone through various evolutions. The Compose* is a language which implements the concept of CF filters in the .NET environment. Compose* has several advantages such as compile-time and run-time weaving, language independence and declarative high-level specifications.
In this talk, I will first present the unique features of the Compose* language and environment. I will then focus on how the declarative features of the language can
help in verifying semantic interface in aspect compositions. In particular, verification of filter and advice (user Defined Advice Types) compositions will be discussed. Second, I
will demonstrate the language independence feature of Compose* by composing aspects written in different languages. Finally, the "semantic pointcut" specifications will be illustrated. If time permits, a demo will be given.
Enquiries: Michelle Lanigan ext 3572
Further Details: www.sqrl.ul.ie
|
February 2007
Title: "Scenarios Read by People and Software"
Presenter: Dr. Thomas Alspaugh
Department of Informatics,Bren School of Information and Computer Sciences, University of California
|
Abstract: Scenarios are widespread in software requirements practice, where they written almost exclusively for human readers. As a result, tool support for scenarios remains weak, and software development does not receive the full benefit of the work put into them. Despite the informal prose form of scenarios, people interpret and use them in consistent patterns that follow relationships embodied in the text. ScenarioML is a markup language with which scenario authors can make these relationships explicit, so that software tools can give effective support for working with scenarios, and programs can read scenarios in order to use them for more purposes. ScenarioML's semantics are defined in terms of how scenarios describe the world, resulting in equivalences and specializations between structurally-related events that can be exploited for scenario refactoring, event recognition, and other software processing.
These well-defined semantics combined with tools for presenting scenarios effectively show promise for a representation of requirements that is clearer and more effective both for non technical stakeholders and for developers. We discuss three recent and current applications of ScenarioML for scenario tool support, automated multimedia presentations of scenarios, and requirements-based testing.
Date: Friday 23rd February 2007
Venue: CS2-037, 11h00, Tea/Coffee at 10h30 in the CSIS Atrium
Enquiries: Michelle Lanigan ext 3572 or Ita Page ext 4246 |
February 2007
Title: "Towards Context-based Autonomic Services"
Instructor: Professor Schahram Dustdar of the Distributed Systems Group, Information Systems Institute, Vienna University of Technology.
Friday 2nd February 2007
|
Abstract: Knowledge workers are increasingly involved in new kinds of organizational structures and work interaction patterns that require highly dynamic forms of collaboration, ranging from Nimble (short lived) to Virtual and Mobile/Nomadic teams. Teams morph from one kind of organisazational structure to another. In most cases, workers engage in many such teams simultaneously and require support from adequate software services. To meet the requirements of dynamic, multiform team working, current Internet-based Collaboration Working Environments must envolve towards large-scale, loosely-coupled, trusted service-oriented systems, with increased emphasis on P2P capabilities.
In this talk we discuss we discuss some scientific approaches focused on a new blend of human collaboration and service-oriented systems that explore two basic research strands:
- Efficient and effective support for human interactions and collaboration in various teams through dynamically aggregated software services;
- Use of human to human or human to service interactions in applying intelligent mining and learning algorithms that can detect interaction patterns for pro-active service aggregation.
In addressing these issues, we present our current findings for mining human activities and providing context-relevant services, at the right time and granularity, to human interaction partners in those various team forms. To this end, relevance-based context representation models and autonomic service adaptation methods for context-coupling and enrichment will be developed.
Date: 10.30 am, Friday 2nd February 2007
Venue: CSG 25, CSIS Building, University of Limerick
Enquiries: Michelle Lanigan 061 213572 or michelle.lanigan@ul.ie
|
November 2006
Course Two: The use of Precise Documentation in Software Development
Instructor: Prof. David Lorge Parnas, Marius Dragomiroiu Duration: 1 day, Thursday November 30th 2006 November
|
| Documentation that is well structured, complete and precise can speed up software development, while increasing both trustworthiness and maintainability. Poor documentation is simply a waste of time. Unless it is easier to get accurate information from the documentation than from the code it will not be used. If it is not accurate and complete it will cost more time than it saves.
Interface document for software components is the most important kind of documentation; if an interface document is unclear or wrong, both the component described, and the components that use it, will have bugs. This tutorial describes an approach to component interface documentation that results in highly structured documentation that the programmers find valuable. These documents assist in review, coding, inspection,testing, preparing user documents and maintenance.
Participants will learn how to read components interface documentation and how to prepare such documents.
Biographies of Instructors:
Professor David Lorge Parnas: (IEEE/ACM Fellow) has been studying and publishing ideas on software development for more than 30 years. He has worked in both industry and academia, using his time in industry to understand "the real problems" and his time in academia to find fundamental solutions
Marius Dragomiroiu: Is currently a Ph.D. student working under the supervision of Prof. Parnas. He is co-developer of the method that will be taught and is using them in studying methods of program family development.
Lunch, refreshments and course materials included.
Cost: 1,200 Euro/participant for combined course one and two (4 days)
Venue: Castletroy Park Hotel, Castletroy, Limerick
Enquiries: Ita Page at 061 234246 email Ita.Page@ul.ie |
November 2006
Title: "Software Tomography-Analysis and Visualisation of Large Software Systems "
Presenter: Professor Claus Lewerentz
Technical University of Cottbus, Germany.
|
Abstract: Industrial and business software systems have grown over the last years to considerable size and complexity. Typical banking systems or telematic applications comprise programs in the size of several million lines of code written in programming languages like Java or C++. Such large software systems belong to the most complex artefacts we have to deal with in construction and long running maintenance processes. Besides providing appropriate construction methods it is a major challenge in the S.E. field to support the analysis and understanding of such complex structures .Typical questions asked about program systems, especially during the evolution process are: What does the actual overall system structure look like?Does the structure of the program code conform to the design architecture? What is the quality of the software particularly with regard to further extension and evolution?What are the most relevant quality problems?What impact has a change of a particular substructure to other parts of the system?How did the system structure and quality change over time? Tools to support large scale program comprehension are based on static analysis techniques, software metrics, and program structure visualisation methods. The talk will give an overview on analysis and visualisation techniques used for so-called “software tomography”. The ideas have led to the development of a set of analysis tools, in particular “sotograph”, a comprehensive software analysis workbench. The analysis approach and the tools have been applied in many real-world projects in industry and have shown considerable impact on improving software quality. Date: Thursday 23rd November 2006
Venue: CS2-037, 10h00
Enquiries: Michelle Lanigan ext 3572 |
June 2006
Title: "Computable Dependability Bounds for Large Markovian Models"
Presenter: Professor Pierre-Jacques Courtois
Catholic University of Louvain-la-Neuve, Belgium.
|
Abstract: A method to bind the steady-state solution of large Markov chains is presented. The method integrates the concepts of eigenvector polyhedron and of aggregation. It is specially suited for Markov chains with high locality and very large state spaces.
A model of a repairable fault tolerant system with 16 millions states is used as an example. Bounds on its availability are obtained by considering a small part of its state space only. The method is potentially useful to bound other types of dependability requirements.
Date: Tuesday, 20th June 2006
Venue: FG042, 11h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246
|
June 2006
Title: "Towards a deductive approach for the safety justification of computer-based systems"
Presenter: Professor Pierre-Jacques Courtois
Catholic University of Louvain-la-Neuve, Belgium.
|
Abstract: This on-going research is based on experience gained in assessing nuclear reactor safety critical computer systems. The objective is to find ways to make the demonstration of the safety of a computer based system as deductive as possible; that is, to put in place concepts, principles and methods to structure and rationalize this demonstration.
The findings so far are that the demonstration should be:
- Based on two prerequisites: validated dependability requirements, and preliminary black box system specifications,
- Hierarchical: relying on four main levels of evidence,
- Recursive: with mechanisms of claim expansion and delegation
- Inductive: with layered arguments of backwards inferences,
- Model-Based: with four distinct but related layered families of models.
Date: Monday, 19th June 2006
Venue: FB028, 11h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246
|
May 2006
Title: "Meta-Tools: Advanced Tools for Java Meta-Programming"
Presenter: Mr. Baltasar Trancon y Widemann
Technische Universitat Berlin.
|
Abstract: Generative or meta-programming is an approach to software construction that focuses on algorithms and tools for the semi-automatic generation of software artifacts. It has the pleasant properties of being highly adaptive and neutral with respect to underlying paradigms, platforms, architectures and tools. In this talk, I present a particular meta- programming framework that has been developed as a synthesis of academic formal methods and the needs of industrial practice. I describe selected elements and connections of the implemented tool chain that provides automatic support for the construction of medium-size projects involving the technologies Java and XML. The tools are designed for maximal transparence to the responsible programmer and emphasize an open and inductive approach to code generation, thus easing validation and smoothing transitions between generated and hand-written code.
Date: Monday, 22nd May 2006
Venue: CSG025, 10h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246
|
March 2006
Title: "Defining and Validating Object-Oriented Design Metrics"
Dr. Yuming Zhou
The Hong Kong Polytechnic University.
|
Abstract: The object-oriented approach to software development moves toward mainstream acceptance among commercial organizations, which promises better management of system complexity and thus leads to quality software. However, research on metrics for object-oriented software development is limited and empirical evidence linking the OO methodology and software quality is scare. In this presentation, I will discuss two topics from my current research on object-oriented design metrics.
The first topic is on a new class cohesion measure. Class cohesion is regarded as a key attribute of design quality and can be used to predict many aspects of software product. For example, software with a set of highly cohesive classes is considered to be easier to maintain, develop and reuse, and also substantially less error prone, compared to software without this attribute. Although a number of cohesion measures have been proposed in the last decade, empirical results showed that our understanding of cohesion is still weak and more precise measures are needed. In this presentation, I will present a precise and operational cohesion measure, which is based on dependence analysis technique. The new cohesion measure not only excludes special methods in a class that do not contribute to class cohesion, but also takes into account four important dependencies within a class and their direction. Also, it has some good properties.
The second topic is on the validation of existing object-oriented design metrics. Existing empirical studies on object-oriented design metrics have shown some of them to be useful for predicting the fault-proneness of classes in object-oriented software systems. This research did not however distinguish between faults according to the severity of impact. It would be valuable to know how object-oriented design metrics and class fault-proneness are related when fault severity is taken into account. I will report the usefulness of object-oriented design metrics, specifically a subset of the Chidamber and Kemerer suite, in predicting fault-proneness when taking fault severity into account. The used techniques are logistic regression and machine learning methods, and the used data set is a public domain NASA data set. The results indicate that existing design metrics are better able to predict low severity faults in fault-prone classes than high severity faults in fault-prone classes.
The speaker Yuming Zhou received the B.Sc. degree in Computer Science from Zheng Zhou University of
Technolgy in 1996. He received the M.Sc and Ph.D. degrees in Computer Science from Southeast University in 1999 and 2003, respectively. His research interests include software metrics, program understanding and analysis, and soft computing. He is currently a research associate with the Department of Computing at Hong Kong Polytechnic University.
Date: Monday, 6th March 2006
Venue: CS2037, 15h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246
|
March 2006
Title: "Program Analysis and Test Data Generation"
Presenter: Professor Jian Zhang
Institute of Software at the Chinese Academy of Sciences, Beijing.
|
Abstract: Correctness of programs is an important issue in computer science. For conventional imperative programs, there are several general
approaches to this problem: formal verification, static analysis and testing. Full-scale verification methods based on Hoare logic and other formalisms are not easy to use. Most static analysis techniques are aimed at discovering specific types of errors in
programs. Testing methods are widely used, yet they often fail to offer the user enough confidence in the program's correctness. In this talk, I shall describe our approach to the problem, which analyzes program paths using symbolic execution and
constraint solving techniques. We are developing a toolkit for analyzing programs written in a subset of C. The tools may be used to generate test data for a given program, or find potential errors in the program (like out-of-bound array index expressions,
or violation of assertions). A distinguishing feature of our tools is that the generated test cases are executable, and the errors found in the analysis are not false alarms. I shall also briefly describe test data generation for database applications.
Date: Wednesday, 1st March 2006
Venue: CSG025, 14h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246
|
February 2006
Title: "Satisfiability Checking -- Beyond the Propositional Logic"
Presenter: Professor Jian Zhang
Institute of Software at the Chinese Academy of Sciences, Beijing.
|
Abstract: The satisfiability (SAT) problem in the propositional logic is a well-known problem. Many people have worked on this problem and there are quite a few efficient SAT solvers. However, in most applications, one cannot use these tools directly. In this talk, I shall discuss some improvements in SAT solving. The first part of this talk describes some tools that check the satisfiability of first-order formulas. They can demonstrate the consistency of formulas, or generate counter-examples to false conjectures. The second part of this talk describes a tool called BoNuS, which combines SAT with linear programming. It can be used to check certain properties (like disjointedness and completeness) of software specifications. I shall describe the basic algorithms of the above tools, and give some examples of using them. I shall also briefly discuss test case generation for combinatorial testing.
Date: Monday, 27th February 2006
Venue: CS2037, 15h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246
|
January 2006
Title: "Is Computer Science to Software Engineering as Physics is to Civil Engineering?"
Presenter: Professor Spencer Smith
McMaster University, Canada.
|
Abstract: This talk explores the relationship between computer science, software engineering and engineering. The questions that will be explored during the presentation include the following:
1) What are the distinctions between science and engineering?
2) How appropriate is the analogy between software engineering and the other engineering disciplines?
3) What is the role of requirements analysis and documentation in software engineering versus the role of requirements in other engineering fields?
4) What is the difference between the design of software and the design of other engineered products?
The structure of the presentation follows the presenter's personal experiences with a transition from civil engineering to chemical engineering and finally to software engineering. Although some of the examples are based on one person's experiences, an attempt is made to make the observations as general as possible.
Dr Spencer Smith, B. Eng. C. S., M. Eng., PhD, EIT, is Assistant Professor in the Dept of Computing and Software at McMaster University, Canada. Areas of research interest include the application of software engineering principles to scientific computation and improvements in the application of algorithms to practical engineering problems. For more details see http://www.cas.mcmaster.ca/~smiths/.
Date: Friday, 6th January 2006
Venue: CSG01, 16h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246
|
July 2005
Title: "A Proposed Initiative on Software Certification"
Presenter: Professor Alan Wassyng
McMaster University, Canada.
|
Abstract: Software is ubiquitous. Unfortunately, software reliability has proven to be remarkably elusive. Where most manufactured products carry a guarantee, software is sold typically with a disclaimer. This situation clearly cannot continue. There are many critical, even safety-critical applications in which software errors may result in loss of life, financial ruin, political chaos, or just major inconvenience. Software certification will serve at least two main functions. The first is obvious. The consequences of errors in some applications are so serious that the applications should be certified as safe or secure or of sufficient quality before they can be used. In other words, these applications need to be regulated. The second function concerns more “normal” software. Certification of these applications will probably, in most cases, be voluntary. This stamp of software quality can be used as a marketing edge, or even a check on a company’s internal quality assurance. This is a long-term initiative that will succeed only with the participation of academic, industrial, and government partners. This seminar will present an overview of the initiative, planned activities, example projects, and an invitation to participate.
Date: Tuesday, 26th July 2005
Venue: CSG01, 11h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246 |
July 2005
Title: "Timing Tolerances in Safety-Critical Software"
Presenter: Dr. Mark Lawford
McMaster University, Canada.
|
Abstract: Many safety-critical software applications are hard real-time systems. They have stringent timing requirements that have to be met. We present a description of timing behaviour that includes precise definitions as well as analysis of how functional timing requirements interact with performance timing requirements, and how these concepts can be used by software designers. The definitions and analysis presented explicitly deal with tolerances in all timing durations. Preliminary work indicates that some requirements may be met at significantly reduced CPU bandwidth through reduced variation in cycle time.
Date: Monday, 25th July 2005
Venue: CSG01, 11h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246
|
February 2005
Title: "An Explorative Talk on the Design of a Paperless Aircraft Maintenance Information System"
Presenters: Mr Brendan Hyland BEng, Mtech (C.I.M.) and Dr. Paul Ashworth, Programming Consultant
Hyland Precision Engineering.
|
Abstract: Aircraft maintenance is a highly regulated industry in which there is little or no room for error. Current research shows that 25-30% of labour costs is due to paperwork. This information system is designed to replace the paper.
Date: Wednesday, 2nd February 2005
Venue: CSG01, 14h30
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246
|
October 2004
Title:"Testing a Computer
Algebra System"
Presenter: Professor Jacques Carette
McMaster University, Canada.
|
Abstract: Testing a computer algebra system is in many ways similar to testing any piece of sequential, mostly state-free software. In some ways it is easier: as most functions are referentially transparent, unit testing is
frequently sufficient to catch problems which would only
be caught at integration testing time in state-based systems.
But in other ways it is much more difficult: the 'correct'
answer is often given only by an equivalence class. Worse,
the equivalence testing problem is often undecidable!
After a quick introduction to CASes (using Maple), an overview
of the testing infrastructure used to test Maple will be
given, with special emphasis on those areas where things
are easier (eg: fully automated cross-platform testing),
harder (eg: zero-recognition), and where some novel features
have been implemented (eg: automated reporting of likely
cause of failure).
Date: Tuesday, 5th October 2004
Venue: CSG01, 10h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext.
4246
|
September 2004
Title:"Computing System
Reliability Using Markov Chain Usage Models"
Presenter: Professor Jesse H. Poore
University of Tennessee, Knoxville,
Tennessee.
|
Abstract: A reliability model for software was developed
in 2003 that is based on usage models as represented by
Markov chains and data from testing and field use. The estimate
is based on reasonable assumptions, has good properties
in the short term and when testing shows no failures. The
model is flexible and supports combining information from
multiple sources. An overview of the process and tools available
for testing based on Markov chain usage models will be presented
with emphasis on the reliability estimate.
Date: Thursday, 16 September 2004
Venue: CSG01, 14h30
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext.
4246
|
July 2004
Title: "Interface Module Specifications
for Real-Time Systems"
Presenter: Professor Dennis K Peters
Memorial University of Newfoundland.
|
Abstract: Interface Modules (IM) are modules that encapsulate
input or output device hardware and the related software,
so that application software can be written without specific
knowledge of the particular devices used. In real-time and
embedded systems an IM will often relate real-valued external
quantities (e.g., time, positions in space) with discrete
valued software quantities. An IM specification must therefore
use a combination of notations and formalisms. Precise and
formal IM specifications are useful in determining if a
software design is acceptable with respect to a system requirements
document.
This talk presents a technique for IM specification that
is an extension of the System Requirements Documentation
technique, which is based on the Software Cost Reduction
(SCR) method. An IM is specified as a 'sub-system' that
interacts with both the external environment and other software
modules in the larger system. The interface quantities are
modeled as functions of time and the behaviour is described
in terms of conditions, events and mode classes. The technique
is illustrated for different styles of software interface,
including function calls, public variables, and a 'callback'
mechanism similar to that found in many graphical user interface
environments. The technique facilitates concise and formal
description of the module behaviour, including tolerances
and delays.
Date: Thursday, 1 July 2004
Venue: CSG25, 10h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext.
4246
|
May 2004
Title: "Ongoing work at Mechatronics Lab, KTH"
Presenter: D.J. Chen
Royal Institute of Technology (KTH), Sweden.
|
Abstract: In this talk, an overview will be given of ongoing work at Mechatronics Lab, KTH, in the areas of systems modeling and modularization for embedded computer control systems (ECS). It provides solution domain meta-models for the systems and the software subsystems, describing the key aspects, design levels, components, component properties and relationships with ECS specific semantics. By constituting the common basis for abstracting and relating different concerns, these models will
also help to provide better support for obtaining holistic system views and for incorporating useful technologies from other engineering and research communities such as to improve the process and to perform system optimization. Further, a modeling framework is derived, aiming to provide a perspective on the modeling aspect of ECS development and to codify important modeling concepts and patterns. In order to extend the scope of engineering analysis to cover flexibility related attributes and multi-attribute tradeoffs, this work also provides a metrics system for quantifying component dependencies that are inherent in the functional solutions. Such dependencies are considered as the key factors affecting complexity control, concurrent engineering, and
flexibility. The metrics system targets early system-level design and takes into account several domain specific features such as replication and timing accuracy.
Date: Monday 31st may 2004
Venue: C1062, 11h30 - 13h00
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext. 4246 |
March 2004
Title: "Open Market Software Development"
Presenter: Dr. David M. Weiss
Director, Software Technology Research, Avaya Labs
|
Two key issues in software development organisations
are how to assign work to developers and how to compensate
developers. In most organisations, work is assigned by a hierarchical
management structure, and compensation is determined by the
judgments of people within that structure. In this talk, He
will discuss different ways both of assigning work and of
compensating developers, to define a family of approaches
to organising software development. Based on the variabilities
of this family, He will suggest a new approach called open
market software development (OMSD).
OMSD assumes that an organisation is capable of producing
a modular architecture for a product line, as described in
recent work on software product-line engineering. OMSD is
designed to take advantage of the strengths and knowledge
of software developers in a way that keeps developers strongly
motivated and strongly rewarded for making decisions that
benefit both them and the organisation for which they work.
Similar to open source development, and to the cooperative
model used at Hewlett-Packard, it provides a framework within
which they can decide what software to develop and with whom
they work. Similar to agile methods, it gives them freedom
to work together in modes that are comfortable to them. Distinct
from these approaches, it rewards them according to the value
and quality of what they produce.
Date: Friday March 26th 2004
Venue: CSG25, 14.30 - 15.30pm
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext.
4246 |
|
March 2004
Title: "Trace Assertion Specifications
and Automata"
Presenter: Professor Janusz A. (John) Brzozowski.
University of Waterloo, Canada.
|
The Software Quality Research Laboratory
(SQRL) invites you to attend a talk entitled "Trace
Assertion Specifications and Automata" on March 30th,
2004
Speaker: Dr. Janusz A. (John) Brzozowski. Dr. Brzozowski
has published many papers in the areas of algebraic
theory of regular languages, finite automata, asynchronous
circuits, testing, and formal methods. He is co-author
of Digital Networks (Prentice-Hall, 1976), and of Asynchronous
Circuits (Springer-Verlag, 1995). His current research
interests include Asynchronous Circuits, Automata and
Languages, and Formal Methods in Hardware and Software.
Title: "Trace Assertion Specifications and
Automata" Trace assertions specifications are abstract
descriptions of software modules (such as stacks, queues,
etc.); they are "black-box'' models representing the
modules. A trace is a sequence of module operations
(such as PUSH, POP and TOP in a stack). In general,
two different traces can produce the same result (for
example, the same stack contents); such traces are considered
equivalent. Each class of equivalent traces is represented
by a canonical trace, and an equivalence relation on
the set of traces permits us to find the equivalent
canonical trace of any given trace. Using examples of
common modules, we show (informally) that there is a
very close relationship between specifications by trace
assertions and specifications by automata.
Date: Tuesday March 30th 2004
Venue: CSG01, 12.15 - 1.15pm
Enquiries: Prof David Parnas ext. 2731 or
Ita Page ext. 4246
|
December 2003
Title: "Using Simulated Execution In
Verifying Distributed Algorithms"
Presenter: Prof. Michael Ernst
|
The Software Quality Research Laboratory
(SQRL) invites you to attend a talk entitled "Using
simulated execution in verifying distributed algorithms"
on December 12th, 2003.
Speaker: Prof. Michael Ernst, MIT CSAIL (CS&AI Lab).
Prof. Ernst's chief research interest is programmer
productivity, which spans the spectrum from software
engineering, through compilation, to programming language
design. (For more information on his work, please visit
http://pag.lcs.mit.edu/~mernst/).
Title: Using simulated execution in verifying distributed
algorithms. Abstract: Prof. Ernst's talk presents a
methodology for using simulated execution to assist
a theorem prover in verifying safety properties of distributed
systems. Execution-based techniques such as testing
can increase confidence in an implementation, provide
intuition about behaviour, and detect simple errors
quickly. They cannot by themselves demonstrate correctness.
However, they can aid theorem provers by suggesting
necessary lemmas and providing tactics to structure
proofs. Prof Ernst will describe the use of these techniques
in a machine-checked proof of correctness of the Paxos
algorithm for distributed consensus.
Date: December 12th, 2003.
Venue: Charles Parsons Theater, 12.00 - 1.00pm.
Enquiries: Prof David Parnas ext. 2731 or Ita Page ext.
4246
|
June 2003 Title: "Software Engineering Education
Seminar"
Presenters include Dr. Johannes Siedersleben of
SD&M and the Fachhochscule Rosenheim, Dr Chris Exton
and Professor David Parnas.
|
SQRL / TTI present a one day seminar
on current practice and future planning in Software
Engineering education. Speakers include Dr. Johannes
Siedersleben of SD&M and the Fachhochscule Rosenheim,
DR Chris Exton and Professor David Parnas.
Further details are in this
document.
Registration forms are available for UL
and Non-UL
attendees. |
February 2003 Title: SQRL Lecture Series
|
Following his Inaugural lecture, DR
David Parnas will present a series of in-depth and technical
lectures for software developers and managers.
The lecture topics and dates are as follows:
- Precise specification of software requirements.
Monday, 3rd February
- Decomposition of software into components.
Monday, 10th February
- Design of component interfaces.
Monday, 17th February
- Precise documentation of design decisions.
Monday, 24th February
- Documentation-driven software inspection.
Monday, 3rd March
- Documentation-driven software testing.
Monday, 10th March
Each of these lectures will discuss both what we know
how to do now and future research.
Download the slides from
these lectures |
January 2003 Title: Professor David Parnas Inaugural
Lecture
|
Professor Vincent Cunnane, Vice-President
Research at the University of Limerick, invites members
of the campus and the wider industry community to the
Professor David Lorge Parnas Inaugural Lecture entitled:
Software Quality Research: Why,
What and How
on Monday, January 20th 2003 from 2.30-4.00pm at
CSG01 (Computer Science and Information Systems building)
at the University of Limerick.
Prof Parnas will discuss why software quality research
is important, what topics he will be studying and how
research at the Software Quality Research Laboratory
(SQRL) will be conducted. The lecture is intended for
the broader University and Industry community and assumes
no expertise in software development.
Download
the slides from this lecture (Adobe Acrobat PDF
format). |
|
|
| |
|
|
