 |
Massey UniversityInstitute of Information Sciences and Technology |
 |
CS Seminars
I arrange seminars for the CS teaching group. If you'd like to present a
seminar about a Computer Science topic at Massey,
contact me, and I'll
find a place for you in the schedule. There are
plans to automate the schedule so that you can
add a proposal to the list and get email reminders when seminars that you've
indicated an interest in are about to occur. But that hasn't happened yet -
the schedule is static and maintained by hand.
Topics as yet unknown
158100 (Computer Applications and the Information Age)
Computer Architecture
159102 (Computing Fundamentals)
159233 (Computer Architecture: Course Controller)
HCI
158359 (Human-Computer Interaction: Course Controller)
159710 (User Interface Design: Course Controller)Research
I used to run 159.799 (Honours Project: Course Controller). I'm no longer in charge, but I wrote some documents that are still used for the course, and are relevant to writing about research in general:Restrictives (when to use the words "the", "a" and "an", amongst others)
Visual Programming Languages
159.776 (Visual Languages and Computer Graphics: Course Controller)
Visual Programming Languages, Data communications, Boolean algebra, Digital Logic, HCI, Processor Design, Microcode, Compiler Construction, Silicon Compilation, C, VHDL, Memory Management, Operating Systems, Delphi, Fortran, Cobol, Pl/1, Snobol, Lisp, Small, Basic, Pascal
Paul Lyons is a member of the Institute of Information Science and Technology's
Human Computer Interaction Research Group,
EMERG (Electronics and Microprocessor Engineering Research Group).and
Data Communications Research Groups.
Current HCI Research
Colour Harmony
A Ph.D. student (Computer Science Senior Lecturer Giovanni Moretti) is working on the Colour Harmoniser. This device is a software tool designed to enable colour-naive programmers to choose aestheticaly pleasing colour schemes for applications without having to learn colour theory. It automates a number of standard heuristics for achieving colour harmony, but the colour schemes it produces are not completely fixed; the user has a chance to alter the scheme in a restricted way, to apply her or his colour preferences, without upsetting the colour harmony or the usability of the resulting interface.
The thesis underlying the research is threefold;
- that a usable tool for choosing colour combinations can be based on a mathematical model of colour that incorporates practical constraints and aspects of the human colour aesthetic
- that the mathematical model should be founded on empirical measurements of human response to colour stimuli
- that, in use, the model should be complemented by human input, via an interface that facilitates the expoloration of a three-dimensional colour space.
We describe foundation of the work in this paper.
Net Worker
This research was initiated by Jaimee Alam, a Masterate student in Computer Science. She invented an idea for a value-added web browser that facilitates the capture of information from multiple sites, and makes it easy for the user to combine the information and manipulate it in a single text document.
In the first instance, I got involved in the project in a consultative capacity. Now I've taken over the implementation, and I'm working with Jaimee's research supervisors, my colleagues Chris Phillips and Elizabeth Kemp, on the further design of the system.
HyperPascal
HyperPascal is a Visual Programming Language which has been under development for several years. A pilot version was implemented to prove the concepts. It worked, but was impossibly slow. Current work is centred on developing a base version that can support later developments, and implementing Active Templates - a graphical representation of dynamic data structure (linked lists, binary trees, etc.) manipulation. The overall project is an ambitious one, and any number of Honours projects, Masters theses and Ph.D.s could be carved out of it. Talk to me if you're interested.
Flexible Music playback
Popular musicians often have difficulties in rehearsing and performing because of lack of availability of some members of the group. Recorded accompaniments can help in these circumstances, but they are very inflexible; a taped accompaniment plays through from the beginning to the end, without any variation. Human musicians are more accommodating. One member of the band can signal the others to repeat the chorus if the audinece really enjoys a particular song, or to cut it short if the audience looks like rioting. This project is developing a system that allows a MIDI accompaniment to be split into sections that are strung together in a default order that can be redefined while a performance is occurring. Thus a chorus or a verse could be repeated, or a section of the music omitted. This isn't quite as flexible as a group of musicians improvising Jazz, but it's a lot better than a totally predefined recorded accompaniment.
Autowiring
In a previous industrially-sponsored project, I designed and supervised the implementation of a specialised CAD package for capturing electronic circuits. One aspectof the design that was not implemented was a system for automatically laying out interconnecting wires in an aesthetically pleasing, and easily followed way. This is widely regarded as a difficult problem. Frezza has published an algorithm for wire layout that takes a more global view of the problem than standard algorithms such as the A* algorithm, and is consequently more likely to find an acceptable route in a reasonable number of steps. The algorithm does not use a grid, unlike most other layout algorithms. Consequently it is not afflicted by the tradeoff that afflicts grid-based algorithms; closely spaced wires can only be achieved with a fine grid, and a fine grid implies long search times.
In this project a student is implementing Frezza's algorithm, and incorporating it into the VISTA software as a real-time background task, to minimise the impact of long search times - which are expected to occur in spite of the reduced number of steps of the algorithm. The HCI aspect of the project is to determine whether this arrangement produces a usable system.
Current Data comms Research
The ATMSWITCH achitecture
Based on previous work on MasseyNet, the current ATMSWITCH (ATM Switch With Integrated Trigger and Chandelier Hardware) architecture has been designed for 1.2Gbps links, and incorporates two distinct hardware modules to provide ATM's Classes of Service, which divide into two types of operation which have mutually contradictory requirements; a single service algoriuthm is not able to handle both type of traffic effectively.
HyperPascal Compiler
IDE for HyperPascal
High-quality expressions in HyperPascal
Modern I/O in HyperPascal
Distortion-oriented Display in HyperPascal
Tree layout in HyperPascal
Parser from tramline diagrams
Colour Harmony for graphic displays
User action lookahead
Prototype ATMWSITCH
Fuller descriptions of these projects (and those of other Computer Science staff) can be found on the Massey Computer Science Research Projects page
Books - Patents - HCI and CSE papers - Networking papers - Miscellaneous - Chemistry
(the list below contains descriptions of some publications, they're also listed separately in a pdf file)
Books
| Doing
Philip Roy, Jason Smith and Paul Lyons Book: ISBN
0-473-06716-1, published by |
The previous edition of this book was troubled by production problems; the index file that was printed was incorrect, and index entries were out by two or three pages after about page 50. There were numerous typos, and formatting inconsistencies. These errors were largely eliminated in the second edition. |
| Doing Battle with the
Delphi IDE Philip Roy, Jason Smith and Paul Lyons Massey University, 2000 ISBN 0-473-06716-1 |
This book was written as a course text
for an HCI class. the students are stage III Computer Science and Information Systems
students who have varying amounts of programming experience. Some have had a brief
exposure to Visual Basic; others have significant experience with Visual C++. Delphi is a serious commercial programming system with specialst features to handle database programming, internet programming, ActiveX components, OLE programming, and visual interface design. Consequently there's a great deal of Delphi to learn. Faced with the problem of producing something of manageable size, authors of books about Delphi, generally find that they have to leave things out. Now this causes a problem. How can they claim to have written a book about Delphi when they've left some of it out? By leaving out the elementary stuff. They assume their readers are already experienced programmers, and can absorb information about Delphi by making analogies between Delphi and other programming languages - or even the current version of Delphi and a previous version. For example, one of the best books in the field, Marco Cantu's book, Mastering Delphi 4 (the version of Delphi that is current as I write this) addresses itself to a readership of Delphi programmers. And even though it actually omits to describe Object Pascal (the actual programming language), it still fills 1275 pages. I sprained my wrist picking it up. Doing Battle... takes a different approach. Because of the varying ability of the intended readership, it begins with a tutorial on the IDE (Interactive Development Environment), and it concentrates on aspects of Delphi and the Windows environment that are relevant to interface programmers. That is, it deals mainly with the development of Graphical User Interfaces. Two types of interfaces are covered: first, interfaces made up from standard components, and secondly, interfaces that incorporate special-purpose graphical components that have been designed specifically for use in a particular application, or class of applications. Even so, the current edition omits a discussion of Object Pascal, the textual programming language that the Delphi development environment supports. That's covered in the HCI course that the text accompanies (though I may put a chapter about Object Pascal into a future edition of Doing Battle...). The result is a book of less than 250 pages - a much more manageable (and cheaper) document than many of the other tomes that are currently available. |
| A Multipath Local Area Network P.J. Lyons and A.J. McGregor New Zealand Patent Number 221572 December 1990 |
|
| Patent, explaining the things that the MasseyNet packet switching architecture does that are new. | |
| A Multipath Local Area Network Paul J. Lyons and Anthony J McGregor United States Patent Number 4,953,162, August 28, 1990 |
Similar patent, issued in the US |
|
User evaluation and overview of a visual language for real-time image processing on FPGAs C.T.Johnston, P.Lyons, and D.G.Bailey Proc CHINZ 2009, (10th International Conference of the NZ Chapter of the ACM Special Interest Group on Human Computer Interaction (SIGCHI-NZ)), 6-7July 2009, 85-92 |
Abstract: FPGAs are often used for image processing, but existing FPGA design tools lack syntactic constructs for some specialized activities that are important in this field, such as timing, resource handling and scheduling. This forces the developer to work at too low a level and makes it difficult to produce a genuinely hierarchically decomposed design. This paper outlines these deficiencies, as the background for an overview of and justification for each of three views in VERTIPH, a visual programming language for defining image processing algorithms on FPGAs. This updates the overview presented in [1]. The paper then presents the results of two user evaluations of VERTIPH, a pre-implementation paperbased user evaluation which found no major changes were required and a post-(partial)-implementation user evaluation. The latter evaluated the novel parts of the language using participants experienced in the field. The key parts of VERTIPH were found to be useful visualisations for the developers, and the only major problem was the interaction required for defining type-connections between views. |
|
Approximate
Graphical Methods for Inverse Regression Geoffrey Jones and Paul Lyons Journal of Data Science, 7, 1, 2009
|
Abstract: Graphical procedures can be useful for illustrating and evaluating the process of inverse regression. We first review some simple and well-known graphical approaches for univariate linear and nonlinear models. We then propose a new graphical tool applicable to situations where the response is bivariate and repeated measures data are available. The proposed method is illustrated with an example of the age determination of tern chicks using measurements on body weight and wing length. |
|
A Visual Notation for Processor and Resource Scheduling Chris Johnston, Paul Lyons, and Donald Bailey Test & Applications DELTA 2008 on 23-25 January 2008 in Hong Kong |
Abstract: FPGAs offer an excellent platform for use in real-time video processing applications. There are several modes of operation that can be used to implement image processing algorithms; streamed, offline or a hybrid of the two. We discuss these and consider the suitability of the present types of languages for the implementation of image processing algorithms on FPGAs. Examples of Hardware Description Languages, Language Extensions, and Hardware compliers are discussed along with their strengths and weaknesses. We propose VERTIPH, a new multiple-view visual language that is designed for image processing on FPGAs avoid these weaknesses. Three views show different parts of the system; an overall architectural view, a computation view and a scheduling view.
|
|
A Visual Environment for Real-Time
Image Processing in Hardware (VERTIPH)
C. T. Johnston, D. G. Bailey, and P. Lyons EURASIP Journal on Embedded Systems
|
Abstract: FPGAs offer an excellent platform for use in real-time video processing applications. There are several modes of operation that can be used to implement image processing algorithms; streamed, offline or a hybrid of the two. We discuss these and consider the suitability of the present types of languages for the implementation of image processing algorithms on FPGAs. Examples of Hardware Description Languages, Language Extensions, and Hardware compliers are discussed along with their strengths and weaknesses. We propose VERTIPH, a new multiple-view visual language that is designed for image processing on FPGAs avoid these weaknesses. Three views show different parts of the system; an overall architectural view, a computation view and a scheduling view. |
|
Towards a Visual Notation for Pipelining in a Visual
Programming Language for Programming FPGAs C.T. Johnston, D.G. Bailey and P. Lyons CHINZ 2006; Design Centred HCI, 7th International Conference of the NZ Chapter of the ACM Special Interest Group on Human Computer Interaction (SIGCHI-NZ) 6-7 July 2006, 1-10 |
Abstract: VERTIPH is a visual language designed to aid in the
development of image processing algorithms on FPGAs (Field Programmable
Gate Arrays). We justify the use of a visual language for this purpose,
and describe the key parts of VERTIPH. One aspect of importance is
how to clearly and efficiently represent a pipeline of processors and in
particular, distinguish a pipeline from the simpler serial or parallel
structures. This paper develops a number of pipeline representations,
discussing the rationale behind and limitations associated with each
representation. The culmination of this development is the Sequential
Pipeline with Detailed Bars, visually an efficient and unambiguous
representation. PowerPoint Presentation (written and presented by Chris Johnston) |
|
Paul Lyons and Geoff Jones CHINZ 2006; Design Centred HCI, 7th International Conference of the NZ Chapter of the ACM Special Interest Group on Human Computer Interaction (SIGCHI-NZ) 6-7 July 2006, 11-16 |
Abstract: We present a computer interface that represents the age of chicks as dynamically-generated chromatic bands on a set of plots drawn on conventional Cartesian axes. The chromatic information allows us to increase the number of variables on the plot from two to three, so that the user can infer a chick’s age from two easily obtained body measurements. The overall image contains a number of realisations of a non-linear multivariate hierarchical model of chick growth; the model is parametric, and each plot is produced by running the model with a different combination of parameters. Even with chromatic banding, the plots can be difficult to interpret because, when many plots overlap, later plots obscure earlier ones. To reveal the information hidden thereby, the interface incorporates a component that allows the user to generate a pie-chart showing all the ages that fall within a 95% confidence ellipse around any point on the plot. |
|
Incorporating Groups into a Mathematical Model of Color
Harmony for Generating Color Schemes for Computer Interfaces
Paul Lyons and Giovanni Moretti VECIMS 2005 (Proceedings of the 2005 IEEE conference on Virtual Environments, Human-Computer Interfaces, and Measurement Systems,80-85, Giardini Naxos, Italy, 18-20 July 2005
|
When
applying Munsell's (or anyone else's) rules for colour harmony to a
computer interface, one is faced with the problem of characterising large
numbers of identical items (buttons, check boxes, and radio buttons, for
example). It is possible to group such items and colour them as a
unit. However, there are isuues to be solved; for example, should one aggregate the
area ofthe items (which would result in a darker, less saturated colour)
or treat them as though they had the area of a single item (which could
result in a garish appearance if the number of items was very large)?
Grouping may also be used to handle a set of items that should be coloured
differently, but need to be related by having similar colours. Here we
describe these issues, and one possible solution.
Abstract
– We
describe an approach to developing a mathematical model of color harmony.
This will be applied in the Color Harmonizer, an automated tool for
coloring computer interfaces and websites. The tool will incorporate a
color harmony engine that can incorporate a variety of theories for color
harmony, and in the first instance, will use the rules proposed by Munsell
and adapted to use in computer displays. We describe abstract and concrete
color schemes, the Chromotome (a tool developed to facilitate the
selection of colors) and techniques for grouping interface elements. |
|
Controlling the Complexity of Grouped Items in Colour Interfaces
Giovanni Moretti Proc. CHINZ 2005 (6th International Conference NZ chapter of ACM's Special Interest Group on Computer-Human Interaction), pps 19-23
|
Abstract: In this paper we describe a method of managing the complexity that arises when automatically colouring a realistic GUI interface. This complexity primarily comes from two sources, from the number of items to be coloured - which in interfaces of realistic complexity grows very quickly - and from the interactions between both the items’ colours themselves and their background colour. Attempting to satisfy the relationships between the colours of items in an interface, their backgrounds, whether items may be coloured identically or must be distinct, and all the while creating a colour scheme that is harmonious requires multiple constraints be satisfied simultaneously. The addition of visually related groups of screen elements complicates this optimization. The use of spring-based algorithms will allow groups to have their own local colour schemes (sub-schemes) while still satisfying the need for overall colour harmony. |
|
HyperPascal
with a persistent memory manager: why and further developments
Todd Cochrane, Paul Lyons and Giovanni Moretti Proc NACCQ 2005 |
|
|
Nine Tools for Generating Harmonious Colour Schemes
|
Surveys nine pieces of software that have been designed to generate harmonious colour schemes, most often by rotating wireframes superimposed over colour wheels, and choosing the colours where the wireframes and the colour wheels intersect. Abstract. We survey a number of tools that have been developed for generating sets of colours according to commonly accepted rules for colour harmony. Informal manual techniques for generating harmonious sets of colours have been known and used for at least a century. Although superficially simple, they have not been precise techniques, as pigment-based and dye-based colouring techniques are not susceptible to accurate measurement, in terms of area of colour used or mathematical relationships between colours, and mathematical analysis does not appeal strongly to the design community. Now the historical separation between engineering and design has narrowed. First, the development of colour television brought numerical precision into colour specification. Secondly, in computers, the medium for colour representation and the tool for calculating colour parameters have been integrated. Consequently, it has also become feasible to derive sets of harmonious colours automatically |
|
Chromotome: A 3D Interface for Exploring Colour Space
Moretti, G, Lyons, P and Wilson, M. |
Outlines the design of Chromotome, an interface component
for exploring a 3D colour space
Abstract. When continuous 3D shapes or enclosed structures, such as solid objects or skeletons, are mapped onto a 2D screen, simplifications such as hulls and wire frames are suitable visualization tools, because most or all of the information is concentrated along discontinuities that occupy only a small proportion of the space. Visualizing a colour space is more difficult. Colour spaces are three-dimensional solids with no discontinuities, and every point in such a space represents a unique colour. A colour space visualization tool must therefore facilitate the exploration of a solid, continuous, three-dimensional shape. Here we describe Chromotome, a software tool that has been developed for this purpose. Chromotome provides a cutaway view of a spherical colour space, and has controls for rotating the space (to alter the hues displayed), for altering the shape of the cutaway, and for visualizing sets of colours positioned according to simple geometrical relationships within the space.the interior of a 3D colour space. |
|
NetWorker: A Practical
Web-Based Tool to Support the Collect-Compare_Choose cycle
Lyons, P.J., Phillips, C, Kemp, E.A., and Alam, J 6th Asian Pacific Conference on Computer Human Interaction (APCHI 2004), Rotorua, New Zealand, June 29 - July 2, 2004, pps 231-240 |
Describes the design, evaluation and modification of a
Networker, a specialised browser tool. The tool allows a user to download
information from a number of sites (the Collect phase) to extract and view
relevant subsets of the information they contain (the Compare phase) and
to make choices based on this integrated presentation of the
information (the Choose Phase)
Abstract.
An
earlier paper has covered the development of a paper prototype of
NetWorker, a tool designed to facilitate a Web usage referred to as the
Collect-Compare-Choose cycle. Here we describe an initial implementation
of the tool, a small scale evaluation, and modifications that were
implemented subsequent to the evaluation. NetWorker is a PC application
with a single window containing multiple web browsers and a text pane. It
allows users to download and view pages from multiple web-sites in
parallel, to drag text from the web pages into a working area (the
WorkPad) and to edit and rearrange information within the WorkPad.
The browsers and the WorkPad can be rearranged in various ways to
make optimum use of the available screen space. |
|
Tools
for the Selection of Colour Palettes
Moretti, Giovanni & Lyons, Paul
CHINZ 2002 ACM
SIGCHI_NZ, 11-12 July 2002, Hamilton, NZ. |
|
|
Colour
in Computer Interfaces
Paul Lyons and Giovanni Moretti
IEEE Symposia on Human-Centric Computing Languages and Environments,
|
|
|
Colour
in Computer Interfaces
Paul Lyons and Giovanni Moretti Proc IHM-HCI 2001, (Vol II) (Treizièmes journées sur l’ingénierie de l’Interaction Homme-Machine de l’AFIHM, 15th Annual Conference of the Human-Computer Interaction Group of the British Computer Scieiety ), 227-228, September 2001 |
A tutorial presented in Lille, 2001. Available for presentation at other conferences. |
|
Colour
Group Selection for Computer Interfaces
Paul Lyons and Giovanni Moretti |
|
|
Some
Possibilities of Visual Programming Languages
Paul Lyons, Giovanni Moretti, and Chrissy Reeves Proc NZ HCI Symposium 2001 |
|
|
Industry
tutorial on colour use in Computer Interfaces
Giovanni Moretti and Paul Lyons
The Use of Colour in Interface Design, OZCHI, |
|
|
Chromatic
Interpolation for Interface Design
Giovanni Moretti, Paul Lyons and Mark Wilson
Proc. OZCHI 2000
“Interfacing Reality in the New Millenium” |
|
|
Colour
Group Selection for Computer
Interfaces
P.J. Lyons, G. Moretti, and M. Wilson
Proc
Conf Human Vision and Electronic Imaging (SPIE2000), |
|
|
Visualising Music for
Performance Flexibility Paul Lyons and Yiding Zhang Proc. SoftVis ‘99, (ed. A Quigley), December 3-4, 1999, 31-39 |
|
|
Programming in Several Dimensions Paul Lyons Proc. SoftVis ‘99, (ed. A Quigley), December 3-4, 1999, 31-39 |
A paper summarising the state of the HyperPascal language at December 1999 |
|
ACCOMPANIST
(Adaptable
Computer-Controlled Output of a MIDI Prerecorded Accompaniment, Not In Strict Time) and
TRAINEE (This Rendition of ACCOMPANIST Is Not Entirely Evolved) Paul Lyons and Yiding Zhang Proc. First Iteration, Monash University, December 1-3, 1999, eds. A. Dorin and J. McCormack, CEMA, Melbourne, 116-125 |
This paper contains a description of a system for presenting a musicaccompaniment to a performer in sucha way that she or he can control the structure of the performance as it occurs. It is based on the availability of a MIDI representation of the accompaniment which has been divided into sections that can be manipulated via a direct-manipulation graphical display on a touchscreen. The paper describes both ACCOMPANIST, a design for a complete version of the system, and TRAINEE, a prototype implementation. |
| Using Object Flows to specify Communication in a Visual HDL M.Pearson and P.J. Lyons Proceedings of the Fourth Asia-Pacific Conference on Hardware Description Languages (APCHDL '97), Hsinchu, Taiwan, 1997, 32-37 |
In this PICSIL-related paper, it is proposed to associate the intelligence required for controlling links between dataflows with the dataflows themselves, so that it does not have to be replicated in each of the communicating processes. This association of processing power with a link - which has a data type definition - is analgous to associating methods with data in an object-oriented programing langauge, and the name Object Flow is therefore used to describe the resulting diagrammatic components. |
| The Oval Menu - Evolution and Evaluation
of a Widget P.J. Lyons, M. Pitchforth, D. Page, T. Given, and M.D. Apperley OZCHI '96, 252 - 259 |
VISTA, a specialised circuit design application constructed for PDL Eectronics, contains a component-selection widget called the Oval Menu. It is intended to allow a circuit designer to keep her or his attention on the insertion point in the circuit diagram, and to allow the circuit construction activity to seem as much like dawing with a pencil as possible, but a with pencil which is capable of drawing complex structures with a single stroke. This paper considers the decisions which were made in designing the Oval Menu and describes a programme of interviews with users which was undertaken after the first version of the application had been implemented, and the improvements which were suggested as a result of these interviews. |
| High-level Graphical Abstraction in
Digital Design M.W. Pearson, P.J. Lyons, and M.D. Apperley VLSI Design , 5, 1, 101 - 110, 1996 |
This is the first major publication describing the PICSIL dataflow-based VLSI design system. PICSIL is designed to allow a VLSI designer to take advantage of the particular ability of two-dimensional diagrams to reveal relationships between circuit modules. Each module is represented by a process in a dataflow diagram. Communication between the processes is represented by dataflows. Abstracting these functions into graphical elements assists the designer to separate the design of a device's dataprocessing (which is defined textually) from the design of its modular architecture. |
| Surveying the VISTA - A Report to PDL
Electronics P.J. Lyons Computer SciencDepartment, Massey University, May 1996 |
This confidential report summarises an interview programme which was undertaken with number of users of the VISTA special-purpose circuit-design application, and describes a number of proposed improvements to the first release of the application, based on those interviews. |
| PICSIL: Integrating Graphic System Design
and Automatic Synthesis Pearson, M.W., Lyons, P.J., and Apperley, M.D Working Paper 95/3, Department of Computer Science, University of Waikato, Hamilton, New Zealand, 1995 |
Based on Murray Pearson's completed Ph.D. research, this paper presents a description of PICSIL, an IC design system which separates the design of an IC into two distinct categories; the first concerns dataprocessing activities, and is described in PPTL (PICSIL Process Transform Language), a textual language based on, and extending HardwareC; the second concerns interactions between dataprocessing modules - their communication, their control, and the data they store - and is described diagramatically in the PICSIL dataflow notation. A variety of designs have been synthesised (converted to CIF, the standard low-level IC design language, equivalent to conventional programming's assembly language level) and verified by simulation. One chip has been fabricated and verified in operation. |
| Designing with PREVAIL Lyons, P.J., in consultation with Apperley, M.D Massey University Computer Science Department, December, 1994 |
A confidential report containing a semi-formal definition of the interface to the PREVAIL (later called VISTA) software designed and later implemented for PDL Electronics Ltd. |
|
Active
Templates: Manipulating Pointers with Pictures P.J. Lyons, M.D. Apperley, A.G. Bishop OZCHI '94 Proceedings, November 1994, pps 93-98 |
A paper outlining a diagrammatic notation for defining dynamic data structure manipulations, such as linked list insertions, binary tree searches, AVL tree maintenance. The notation is based on the concept of defining data sructure manipulations using before- and after-pictures of the appearance of the affected part of the data structure. To define a data structure, the programmer draws a picture of a standard element. This is equivalent to, and replaces, a conventional type declaration; the picture can be used as a drawing template by the IDE whenever the programmer references an instance of the structure. The word active is applied to the notation because the IDE can use references to particular configurations of the data structure to prompt the programmer to specify what to do when the data structure exists in other configurations. This allows complete algorithms to be constructed pictorially with minimal input from the programmer, but without the need for the IDE to infer parts of the algorithm, as would be the case in systems based on the programming-by-demonstration paradigm. |
| Automating
Digital Design using PICSIL Murray Pearson, Paul Lyons, and Mark Apperley Proceedings of the First Electronics New Zealand Conference (ENZCon), August 1994, pps pp 87 - 92 |
One of two conference papers (see also Synthesis of Digital ICs from Data Flow Diagrams) describing the planned PICSIL system, an IC design environment based on the use of dataflow diagrams to describe the IC's funcional modules, and their relationships to each other. Essentially a progress report midway through Murray Pearson's Ph.D. research. |
|
Synthesis
of Digital ICs from Data Flow Diagrams Murray Pearson, Paul Lyons, and Mark Apperley Proceedings of the First Asian Pacific Conference on Hardware Description Languages, Standards and Applications; December 1993, 84-88 |
One of two conference papers (see also Automating Digital Design using PICSIL) describing the planned PICSIL system, an IC design environment based on the use of dataflow diagrams to describe the IC's funcional modules, and their relationships to each other. Essentially a progress report midway through Murray Pearson's Ph.D. research. |
|
HyperPascal: A Visual Language to Model Idea Space Paul Lyons, Craig Simmons, and Mark Apperley Proceedings of the 13th New Zealand Computer Society Conference, August 1993, 492-508 |
A paper describing the philosophical underpinnings of, and some of the simpler concepts involved in the design of a picture-based general-purpose programming language. It covers the use of structure diagrams to represent data manipulations and control structures, the "porthole" notation (since superceded by Active Templates) for dynamic data structures, hyperlinks within the program to connect instances of a variable with its declaration, subprogram invocation, the Scope Tree for representing declarations in context, and it introduces the State Tree, a concept which did not achieve a practical form until the Active Template notation was devised. |
|
PICSIL - A Data Flow Approach to Silicon Compilation M.W. Pearson, P.J. Lyons, and M.D. Apperley Proceedings of the 28th National Electronics Conference August 1991, 168 - 173 |
An early paper describing Murray Pearson's Ph.D. research project, an experiment in applying Visual Programming language ideas to the design of Integrated Circuits. Preliminary notions of defining circuits in terms of Algorithmic State Machine diagrams (essentially Finite State Automata) had by this stage been superceded by the idea of borrowing the dataflow notation to describe the major functional units of the circuit, and the communication paths between them. The synthesis tool based on capturing this modified dataflow notation was called PICSIL, for PICtorial SILicon compiler. The notation has the same name. |
| An Overview of GED, A Language-
independent Syntax- directed Editor G.S. Moretti and P.J. Lyons Australian Computer J., 18, 2, 61 - 66, May 1986 |
A paper describing Giovanni Moretti's M.Sc. research. GED is a system which accepts a BNF language definition, and generates a syntax-directed editor. GED worked, but it was unexpectedly difficult to use, so from an HCI perespective, it might seem at first blush to have been unnsuccessful. However, an understanding of why it was difficult to use has informed much of the authors' later HCI research work. |
|
Simulation Studies of two ATM Switch Architectures P.J. Lyons, S. Evans and D. Ferry Proc. NETWORKS 3 (the Third New Zealand ATM and Broadband Workshop), January 21-22, 1999, 30-38. |
Simple simulation studies were performed on two approaches to output scheduling in ATM , the ATMSWITCH regular triggering mechanism and the Queue Prediction algorithm. It was found that the triggering mechanism produces zero jitter for certain types of CBR circuit, bu appreciable jitter int he general case, whereas the Queue prediction algorithm produced less jitter in general, but had no "zero jitter mode." Both approaches performed well with ABR and UBR circuits. |
|
Estimating
Clock Speeds for the ATMSWITCH Architecture P.J. Lyons and Bo Yan Proc. NETWORKS 3 (the Third New Zealand ATM and Broadband Workshop), January 21-22, 1999, 39-53 |
The ATMSWITCH architecture relies on a single shared buffer memory, which has sufficient bandwidth to transfer information continuously via 14 1.2 Gbps ports. The centralised controller for this device must run at a high clock speed, in order to accomplish all the housekeeping associated with the ATMSWITCH's sophisticated output scheduling hardware. This paper reports on a behavioural VHDL simulation which was carried out on the architecture to determine the number of controller clock cycles required to perform each data-buffer read and write, and hence (since the buffer memory takes 12ns to perform each read and write operation), the speed at which the controller clock will need to run. It was found that the controller performs 10 clock cycles during the read or write operation, and another 2 beforehand, in preparation for the read or write. Consequently a 1.2 ns clock would be suitable. |
| Output
Triggering adds Classes of Service to the Associative Chandelier P.J. Lyons Proc. ATMWORKS 2, February 1997, 99-116. Eds A.J. McGregor and J. Cleary. |
Regularly triggered output events, keyed to a particular channel and its the bandwidh requirement, are used to schedule output from CBR and VBR channels. The chandelier, running at a lower priority, is used to schedule access to the remaining bendwidth, for ABR and UBR channels, thereby making efficient use of the bandwidth, and allocating ir vary to the ABR and UBR channels. |
| Bandwidth
Reservation in an ATM switch using the Associative Chandelier P.J. Lyons FIMS Report Series A: 96/1, May 1996 |
A modificaton to the behaviour of the associative chandelier, which would allow ABR and UBR channels to reserve bandwidth, is proposed. The modification affects the design of the chandelier's control algorithm, but not the chandelier; the control algorihm simply keeps track of the number of output slots which each channel should have used, according to its reserved bandwidth , and removes cells from its buffer queue until the difference between the number of cells which it ought to have output and the numer of cellswhich it has output, at that time, is zero. |
| The Associative Chandelier - Fair, Efficient,
Prioritised Buffering in ATM Switches P.J. Lyons, A.J. McGregor and G.S.Moretti Proceedings of the First New Zealand ATM and Broadband Workshop, |
A modificaton to the behaviour of the associative chandelier, which would allow ABR and UBR channels to reserve bandwidth, is proposed. The modification affects the design of the chandelier's control algorithm, but not the chandelier; the control algorihm simply keeps track of the number of output slots which each channel should have used, according to its reserved bandwidth , and removes cells from its buffer queue until the difference between the number of cells which it ought to have output and the numer of cellswhich it has output, at that time, is zero. |
| MasseyNet Administration Documentation - Release 2.1 P.J.Lyons Computer ScienceDepartment, February 1998 |
|
| Performance Measurements in the Masseynet Local Area
Packet Switching Network P.J. Lyons, A.J. McGregor, and E. van Oeveren Proceedings of the 1987 NZ Computer Society Conference, R126-R125 |
|
| Masseynet: A University-oriented Local Area Network P.J. Lyons and A.J. McGregor MICROS PLUS: Educational Peripherals; Proceedings of the IFIP Working Conference on the Educational Implications of Connecting Tools and Devices to Micro-computers, August 1986, 155 - 167 |
|
| Masseynet - Current Development P.J. Lyons, A.J. McGregor, B. Meads, E. van Oeveren, C. Eagle, and L. Gill Nelcon Proceedings, 23, 107 - 112, August 1986 |
|
| Masseynet - A Local Network Using Wider Area
Techniques M. Murphy, A.J. McGregor and P.J.Lyons Nelcon Proceedings, 23, 107 - 112, August 1986 |
|
| Networking Research Group 1986 Report P.J. Lyons Massey Computer Science Report 85/8, October 1985 |
|
| Preliminary Specifications for a Data Communications
Exchange P.J. Lyons and A.J. McGregor Massey Computer Science Report 85/7, October 1985 |
|
| Computing Equipment RS232 Interconnection Standard T.W. Livingstone, P.J. Lyons and A.J. McGregor Massey Computer Science Report 85/7, October 1985 |
| Report on Computing Equipment supplied by
NZ Police: Confidential Expert opinion P.J. Lyons April 1997 |
Short description. |
| Report on Computing Equipment supplied by
NZ Police: Confidential Expert opinion P.J. Lyons July 1996 |
Short description. |
| Some Aspects of Prime's PRIMOS Operating
System A.J. McGregor, P.J. Lyons and T.W.G. Docker Massey Computer Science Report 85/8, October 1985 |
Short description. |
| Strong Typing and Related Issues J. Ansell and P.J. Lyons Massey Computer Science Report 86/6, May 1986 |
Short description. |
| Writing Reports and Theses P.J. Lyons Computer Scence Department, Massey University |
There are lots of documents that offer advice to research students about
how to undertake research and how to write their theses. Nearly all of them have an innate
bias that reflects their author's personal area of specialisation. This gives rise to
advice which is often wildly inappropriate for Computer Scientists. This document was written primarily for Computer Science Honours students. It suggests appropriate structure for reports, length, citation syle, how to avoid common mistakes, and a variety of other similar things. Although I haven't published it outside the Computer Science group, I occasionally receive requests for copies by reseaerch students in other departments, and believe it has been used by at least one other department in the university (though, for all I know, it could be as an example of how not to write!) |
| The Conformation of Malformin-A D. Hall, P.J. Lyons, N. Pavitt, and J.A. Trezise J. Comput. Chem. 3, 1, 1982 |
Short description. |
| An Appraisal of the Applicability to
Molecular Packing Analysis of some Global Minimisation Techniques David Hall and Paul John Lyons Computers and Chemistry, 4, 69-73, 1980 |
Short description. |
| An Investigation of the Application of
some Optimisation Methods to Molecular Packing Analysis P.J. Lyons M.Sc. Thesis, Auckland University, November 1976. |
Molecules are characterised by their shape, and the way they pack
toegether in crystals. It's really quite difficult to "see into" a crystal to
find out how the atoms are dispersed. The standard technique, crystallograpy, is expensive
and time-consuming. There have been various attempts to integrate our understanding of how
atoms interact in order to predict how the atoms of a compound with a known chemical
formula will be arranged in space. One technique, molecular conformational analysis, predicts the arrangement of atoms within the molecule. Another technique, molecular packing analysis, predicts the arrangement of molecules in a crystal structure. I investigated several numerical optimisation techniques that had been proposed to make this prediction more rapid. The investigation showed that one of the techniques was as effective as the old war-horse of optimisation, least-squares minimisation. (It might have been possible to pursue the research further, but I moved to computing instead.) |
Actor/singer/director/set designer/set painter in more than 70 theatrical activities during the last 30 years- plays, operas, musicals, ballet, TV.
Formerly a bass singer in the close-harmony group Not Just Jazz and the jazz choir Fascinatin' Rhythm. Cathedral cantor. Currently getting a new group going that could possibly end up being called "May Include Nuts." In 2005, appeared in the dual roles of Alcindoro and Benoit in a production of La Boheme.
Watercolour paintings in a style that ranges from naturalistic to semi-abstract.
Have a look at my gallery website.
An occasional hobby activity. At the start of 2002, I designed and built some leadlight panels for the front door of our house. The picture was taken with a cheap camera and is somewhat distorted; the doors aren't actually warped. When we bought the house, the estate agent told us it had a name ("Okioki," a Maori word meaning "restful place"). I thought that was just a tall tale, till I found a tatty old name-plate in the attic with the name on it. So the name got incorporated into the leadlight.
