Declarative Systems & Software Engineering Newsletter

Issue 2 - 31st October 1994

Editor - Hugh Glaser

CONTENTS:

EDITORIAL
TODAY'S SEMINAR
TRIP REPORT - Keith Phalp
FAST FINAL REPORT - Hugh Glaser
TRIP REPORT - Pieter Hartel
STUDENTSHIP APPLICATION - Hugh Glaser
GOSSIP - Stuart Brodie
CHINA EDUCATION AND RESEARCH NETWORK (CERNET) - Fu Xiao Ying
* DSL SOFTWARE LIBRARY - Danius Michaelides
SHORT COURSE - An Introduction to Message Passing

* For action please

EDITORIAL

Well now, that was a surprise! This first real issue seems to have grown consideraby more than I expected. So what do you think? Are the articles of interest? Should we cast the net wider or narrower? Make the articles shorter with pointers to ecs.research.dsl? Do no such thing? How could the format be improved to make it more readable (while still being VT100 compatible)? I would appreciate any such feedback.

In the meantime, I at least am finding the input interesting, so please keep sending me articles.

TODAY'S SEMINAR

The DSL Discussion on Monday, 31st October at 13:00 will be given by Adrian Smith.

Design Models for Distributed Systems

Monday's talk will be a discussion on the development of a distributed system for the "electronic supermarket". A model will be specified in a suitably "formal" language and its use in system simulation on a single PC will be described. Then an actual implementation consistent with the established model will be outlined based on conventional client/server machinery. In this way, the advantages of a modelling phase in developing distributed software should become clearer.

TRIP REPORT - Keith Phalp

Report on the Second International Symposium on Software Metrics October 24-26 1994, London.
Attended by Keith Phalp and Rachel Harrison.

This conference was intended to bring together the major players in software measurement research and practice (and the likes of me). However, one of the recurring themes of the two days was the substantial gulf between these camps,which was further exacerbated by a number of highly technical, inaccesible, and (OK Let's be honest here) poor research presentations.

Depsite this there were some fabulous talks, notably the keynote speeches by Bev Littlewood 'Learning to Live with Uncertainty in our Software', and Bill Curtis on 'Measurement Conundrums', and a useful measurement tutorial by Ross Jeffrey which gave a great deal of practical advice on implementing software measurement programs.

Like many of these events the real value is in talking to those who represent the state of the art, and finding out what everyone else is up to. From this point of view we were both very encouraged, because its clear that there is still plenty to do, and a place for us to make a contribution.

FAST FINAL REPORT - Hugh Glaser

I'm pleased to report that the final assesment of the FAST Project rated it Alpha 4 (Gamma, Beta, Alpha 1-5) for Scientific/technological merit, although only satisfactory for management. (I will be commenting on the latter!)

TRIP REPORT - Pieter Hartel

Eduard de Jong and I have just been to CARDIS, the first smart card research and advanced application conference in Lille, France (Oct 24-26). There were 65 participants, coming from academia as well as industry. Most were in a sense testing the water. Potential smart card users want to know fact from fiction in the sales talk of the smart card vendors.

I presented [Har94f], which went down rather well. We received many comments and an invitation to return to Lille University later for a seminar. Our work seems to be unique in the sense that we design for testability. This is particularly relevant in view of the emerging ITSEC standard. Products that satisfy this standard at a high level will be rare, and highly successful.

There is quite a lot of research required in this area, comprising elements of architecture and system design, formal methods and design for testability. Needless to say that functional programming is a key element in all this. 

@inproceedings{Har94f,
        author = {P. H. Hartel and E. K. {de Jong}},
        title = {Towards testability in smart card operating system design},
        booktitle = {1st Smart card research and advanced application conference ({CARDIS} 94)},
        editor = {V. Cordonnier and J-J. Quisquater},
        publisher = {Univ. de Lille, France},
        address = {Lille France},
        month = {Oct},
        year = {1994},
        pages = {73-88}}

STUDENTSHIP APPLICATION - Hugh Glaser

I have applied, with Celia Glass from Mathematics, for an Earmarked Studentship from the Maths Board of EPSRC to research into OR techniques and Distributed Computing, to go alongside the DAGSOD Project.

GOSSIP - Stuart Brodie

The new BSc in Info Eng was the subject of much hilarity on the News Quiz this afternoon (Saturday). Referred to dwb as the "crackpot professor" who invented this Internet based course.

CHINA EDUCATION AND RESEARCH NETWORK (CERNET) - Fu Xiao Ying

An Overview

China Education and Research Network (CERNET) is the first nation-wide education and research computer network in China, which is presently under construction. The CERNET project is funded by the Chinese government and directly managed by the Chinese State Education Commission. It will connect all the universities and institutes in China in the near future and will connect high schools, middle schools, primary schools and other education and research entities by the end of this century. Furthermore, it will link to the global Internet and will become a major part of the China Internet community. The CERNET project will greatly improve the education and research infrastructure in China and train network experts as well as experienced network end users. In a word, it will help to boost China's education, research and economic developments. Full article on ecs.research.dsl

DSL SOFTWARE LIBRARY - Danius Michaelides (dtm93r)

Declarative Systems Lab Account email address: dsc

As some of you may or may not know, there is a small collection of software for the lab's unix machines located in /usr/local/dsl. Now that the new year has started and people have settled down, perhaps they have a clearer idea of what software they may want.

Please EMAIL ME and indicate:

List of currently installed software

  • akcl - Austin Kyoto Common Lisp
    Version(1.605) Fri Nov 1 15:11:53 GMT 1991
    clpr - the CLP(R) constraint logic programming system
    gofer, goferc - An extended subset of Haskell.
    About a fast as Miranda(tm) version 2, and in many ways nicer.
    There is also a version for PCs.
    feel - Eulisp implementation (interpreter), version 7.05. Sun4 version only.
    latprint - A program for printing text files which uses LaTeX.
    The main advantage is that it draws a line between short files rather than starting a new page (like enscript). For help try it with no arguments. It has options for numbering lines, peculiar tab sizes and printing 2, 4 or 8 pages per piece of paper (-4 is quite readable).
    m3 - Modula 3
    pdss - PDSS-KL1 V2.52.17 (Thu Jun 4 17:42:46 BST 1992)
    scc, sci - The Scheme-to-C compiler and interpreter.
    scm - Scheme interpreter (SCM version 4a14) slib1b5
    smlp - A profiling version of Sml 0.93 of New Jersey.
    rc - A shell. As used in Plan 9.
    tclsh/wish - Tcl and Tk shells. X11 for thickies (including me!).

    • SHORT COURSE - An Introduction to Message Passing

    Topics covered:

    Attendees should have a basic grounding in UNIX and programming experience in FORTRAN-77 or C.
    Duration . . . . . .0900-1700
    Venue . . . . . . . Lanchester Building, Room 3003
    Date. . . . . . . . 24 November 1994
    Then on a regular monthly basis.
    To obtain further information on any of these courses and to register please contact on Email support@par.soton.ac.uk

    Copy deadline: 5pm Friday for Monday's newsletter, but send the articles any time.

    Hugh Glaser
    Declarative Systems & Software Engineering Group
    Department of Electronics and Computer Science
    University of Southampton