EASE 90, Engineering Applications; Is the Future Parallel?: Further details will be distributed shortly, but note the article in this Issue. Networking is always an emotive issue with the Community and I draw your attention to the article Products for Academic Community Networking of Suns.
Community Clubs in CFD and AI are now becoming a reality - this Issue gives up-to-date information for your participation. I would encourage the Community to give these initiatives your active support.
The Computing Facilities Committee (CFC) has accepted the Control Transputer Application Community Club (CTACC) as a Community Club in Real-time Control.
CFC has provided funding for two man years of research effort plus recurrent costs. On the advice of the Community Club Executive CFC has decided to invite proposals from the Real-time Control community for two six-month EMR contracts aimed at producing demonstrators to show the potential of transputers for Real-time Control Engineering applications.
Ideally a demonstration will cover a range of technical domains, exhibit clearly the operation of concurrency based on a transputer architecture, be self-contained, possibly portable and ideally visual. It is likely that a contract would be used to develop an existing piece of equipment to fulfil this role. An award would be conditional on the deliverable being made available for two years beyond the end of the contract.
An outline proposal should not exceed two side of A4 paper, and should be sent to me by Wednesday 28 February. Proposals will be considered by the Computing Facilities Technical Advisory Group at its meeting on 3 April 1990, with advice from the CTACC Executive.
The present membership is as follows:
Acting upon the recommendation from the Computing Facilities Technical Advisory Group (CFTAG) the following Extra Mural Research contracts have been awarded by Computing Facilities Committee.
This will be undertaken by the Napier Polytechnic of Edinburgh and the work will be directed by Mr Teo Goh of the Department of Mechanical, Industrial and Software Engineering. The results of this six-month survey will be presented at an EASE seminar in the Autumn.
CFTAG has made an initial allocation of SPARC server 390 systems to the School of Electronic Engineering Science at the University of Wales Bangor, and to the Department of Electrical Engineering at the University of Bradford. This equipment is expected to be ready for installation during February. CFTAG also recommended that a server be installed in the School of Mechanical Engineering, University of Bath, but this award has been delayed pending the allocation of sufficient funds to purchase the equipment.
This has been awarded to the Computations Department at UMIST. The contract will be for two people for a period of three years commencing 1 December 1990.
Further contract opportunities will be advertised in the Newsletter; all Higher Education Institutes (HEIs) are eligible to respond to these to carry out specified work or apply for allocation of CFC funded facilities.
The Computing Facilities Technical Advisory Group (CFTAG) decided that as part of the Education and Awareness Programme a comprehensive survey of the community having access to EASE should be undertaken. In addition to establishing personal contact between the Higher Education Institutes (HEIs) and members of the Informatics Department at Rutherford Appleton Laboratory, the main objectives of the visits are:
By the time you read this edition of the Newsletter well over half of the visits will have taken place, and it is anticipated that the exercise will be completed by the end of March or early April 1990.
SERC staff involved in this important activity have commented that it has proved to be well worthwhile and very rewarding, and we extend our grateful thanks to all those at the HEIs who have been involved in the organisation of the visits. The results of the survey will be presented to CFTAG by the early Summer and the recommendations accepted by the Committee published shortly afterwards.
The major event of the EASE Education and Awareness Programme during the year will be the 1st Annual Symposium and Exhibition, EASE 90, being held at UMIST Manchester on the 18-20 April 1990, with the theme, Engineering Applications: Is the Future Parallel? It is intended that the Exhibition will include hardware and software products by the major suppliers in the marketplace, but space can also be available for members of the Engineering Academic Community to display their current activities and achievements in this important area.
If you are interested in being involved in this activity please contact me as soon as possible. Some assistance may be possible in the provision of suitable hardware etc and the costs of transport to Manchester will be covered.
Free software for your favourite PC or micro! Available for the cost of a call across JANET.
There is, amongst the micro/PC community, a considerable circulation of software for which you are not expected to pay. This Public domain software (sometimes called Shareware) has in most cases been written by users who yearn for fame rather than hard cash. They release their software to the world at no charge (although they may ask for a small sum to cover media costs or printing charges for the manual) and bask in the warm glow that fame brings them.
Wonderfully useful utilities, life-enhancing suites of routines, and interesting blocks of code for exploring the limits of your PC can be yours for free.
Most of the public domain software for the popular PCs/micros originates in the Unites States, and hence it is usually written for machines that are popular over there: IBM PCs, generic MS-DOS machines, Commodores and Ataris seem the usual hardware platforms. A few UK software writers have gone public and their code for the non-mainstream machines (BBC Micro, Archimedes, Sinclair) but the diversity of code for these machines is narrow in comparison with the IBM PC and MSDOS world.
Since there is intense competition for being the best at writing public domain code, advances are rapid. Many of the code-writers are employed by famous research laboratories (MIT, JPL, Bell Labs) and write public domain software in their spare time. Do not be surprised, therefore, if you find some extremely advanced techniques being used in an apparently mundane application.
If you can not understand the reasoning, fear not the documentation may have been written by Californians, or (even worse) UNIX gurus, who make obscurity their trademark.
There are numerous sources of public domain software - user groups, bulletin boards, electronic mailboxes etc., but the best source in the UK is a special service run by Lancaster University - the UK Public-domain software Archive, known universally by its JANET name of LANCS.PDSOFT.
LANCS.PDSOFT is a machine whose function in life is to hold and distribute public domain software. The machine is a Sequent Symmetry, running the Balance operating system, which has much of the look and feel of the UNIX system.
It is accessible through JANET on the address 000010403000, (but try LANCS.PDSOFT first), supporting FTP and interactive terminal login. The machine supports KERMIT as well, so as a last-ditch path you can log in directly and KERMIT the file to your PC. This is not recommended because it ties up terminal login slots on the machine.
The first thing is to log into the machine and look in the index.
If you call the system, it will announce itself and invite you to log in. Use the userid pdsoft and password pdsoft (note that this must be in lowercase).
Once logged in, typing ? gives a list of commands available, the most useful ones being dir to obtain a directory listing, and cd directory to move to a selected directory.
The most useful place to start is the micros directory, where software is indexed by machine (micros/ibmpc, micros/mac and so on).
Within a machine-type directory, the subdirectories all have instantly forgettable names (e44, f78), but what is held in each subdirectory is revealed by typing the file micros/index. It is a good idea to take a copy of this file for reference as the structure of the filestore is explained at the beginning.
Once you identify the subject of interest, position yourself in the relevant directory (using cd command), and on doing a dir you will find a file called xxx.msg (where xxx is the directory you are in). This file contains a description of the software held in this particular directory, and how to acquire it.
To cope with the vagaries of various networks, communications systems etc., the executable files are often held in special forms - known as ARC or BOO. These are compacted versions of the executable code, together with a checksum so you can determine if the file has become corrupt whilst transferring it. You will need the appropriate programs to deBOO or unARC the files before you can run them. The deBOO and un ARC programs are held on LANCS.PDSOFT.
The transferring of the BOO or ARC files from Lancaster to your PC can be done either directly using KERMIT (but set the following pad options first,
TRANSPARENT; WIDTH = 255; CONTROLCHARS; NOGREENBOOK)
or by FTPing the file to your local machine, then using KERMIT. The latter method is preferred, since tying up one of the limited number of login slots on the Lancaster machine will not make you popular.
Note that if you are using the FTP/KERMIT method via one of the NERC IBMs then you should use BFETCH when initiating the FTP, and SET FILE BINARY when using KERMIT. This ensures that the file is shipped as a bit pattern and circumvents any of the grosser problems with ASCII-to-EBCDIC translation tables.
Once you have the .BOO or .ARC file on your PC, run deBOO or unARC against it to create the .EXE file you need. If using unARC, you will almost certainly find several files produced as a result (ARC format allows files to be concatenated, shipped as one logical file, then disassembled at the remote end into the original named files).
One of the files will usually be called READ.ME, FRED.DOC or something similar. This should be printed off (it's the documentation!) and read to familiarise yourself with the program. Then, off you go!
Public-domain software has received a bad reputation, as frequently conveying viruses, or doing un-advertised things to your data. For this reason, it is a good idea to try out any public domain software on an isolated machine, first taking backups of any vital files. If the public-domain software does anything untoward, then you can re-format the hard disk and restore all your files. In practice, most public domain software is clean, and usually picks up any illicit modifications via bulletin-boards or malicious tampering whilst being copied from disk to disk. Remember that what you put on your PC is up to you - if you accept data and files from uncertain sources, you may get caught.
The supporters of the LANCS.PDSOFT archive offer no warranty that the programs held there are free from viruses or mischievous code. They cannot, therefore, be held responsible for any damage or loss of data that may occur if running programs acquired from their machine.
Sun in the UK have recently made enhancements to the availability and support of networking products conforming to the Joint Network Team's (JNT's) recommendations for interworking within the Academic Community. An edited version of Sun UK's press release explaining what is now available is given below; the following is a short explanation of where the products fit into the overall networking picture.
Coloured Books Applications software provides file transfer, mail and job submission to/from a host that also supports the protocols involved. The JNT's current recommendation is that any host connected to JANET, either directly or indirectly, should support these. They will be supplanted by products conforming to ISO Open Systems protocols over the next few years, 3-'> implementations become available.
To connect a host system (Sun or otherwise) to a network, suitable low-level networking software is also required:
EASE policy is to supply Coloured Books software, together with the appropriate low-level protocol support, on the Central Servers it provides as part of its programme. This allows workstation users supported by such a Server (ie not just users of Sun workstations) to utilise the services provided. However, there may be a requirement to mount the software on individual workstations as well in some cases, particularly where a Pink Book Campus LAN is in use.
For prices of Sun's products contact the company direct. Sun emphasise that the software is fully supported; this means that bugs and inadequacies should be reported to them for fixing as soon as they are encountered!
This implements Red (full server), Grey, Blue, Green Yellow Books. It does not have any of the lower level software support (X.25 and below) as part of its own functionality. It is available at one price, non-discountable, for a site-wide Architecture-wide license. The current architectures supported are Sun 3 and SPARC.
For each server or workstation that is physically connected to the JANET or PSS network SunNet (aka SunLink) X.25 product is required. This is normally a single node in a department or University. All Sun architectures are supported, including the 386i, and can be mixed within the license taken out.
In addition, of course, a site license for Coloured Book Applications will be required.
SunNet X.25 currently implements the 1980 standard only.
To implement Pink Book communications on exclusively the Sun 3 and SPARC products (including servers) the Sun Pink Book Software is required.
If the site has a site-wide license for X.25, then the first copy license for Pink Book Software is considered site-wide too.
The next release of SunNet X.25 is expected to contain enhancements for the 1984 standard and will operate over ethernet, in addition to wide area networks. This is effectively the same functionality as the Pink Book Software (PBS) product which will be replaced by this new release of SunNet X.25. Customers under Sun maintenance will be able to move to this new release from PBS without charge.
Pricing implications: Those customers with a site license for SunNet X.25 and current maintenance for it, will be able to run Pink Book functionality free of any further charges (other than possibly minimal ones for media) on all architectures supported by this new release of SunNet X.25. Should the Coloured Book Applications suite become available for the 386i, those customers that already have a licence will be able to run the 386i version without obtaining a new license (though new software will need to be installed).
In all cases Coloured Book Applications software will be required for functionality above basic terminal support (X.29/X.3) and the current release will require modification - customers not on maintenance will need to re-purchase the product, others will be invited to upgrade free.
It is anticipated that the SunNet X.25 release referred to above will be ready for customer shipment during 1990.
As detailed in the January issue of the Newsletter, we are at present gathering users' requirements for X toolkits In order to survey and subsequently recommend suitable toolkits for the community. If you are writing, or intending to write, a program which will use a mouse-driven, highly interactive User Interface, we would like to know what your toolkit requirements are, and the constraints within which it will need to work.
In particular we are interested in:
You can inform us of your requirements by filling in a user requirements form which is available from your Department's EASE contact.
Until quite recently, the major role of computers In science and engineering research has been to find numerical solutions to problems that are too complicated to be solved using algebraic methods. Finite-element packages and subroutine libraries such as the NAG library play an important part In many areas of engineering. As a result, it is not widely appreciated that computers can be used to perform calculations using symbols as well as numbers, and hence to obtain algebraic solutions to certain types of problem. Computer algebra systems such as MACSYMA, Maple and REDUCE have been widely used In North America for many years, but they have not enjoyed widespread use in the UK.
Computer algebra systems can perform a wide range of algebraic calculations including:
There arc several computer algebra systems which can be used for engineering research. They include REDUCE, Maple, MACSYMA and Mathematica, and they can be run on a variety of computers from micros such as the Apple Macintosh, through popular UNIX workstations such as the SUN-3 and SUN-4, up to mainframes such as VAX/VMS and IBM VM/CMS machines.
An EASE Awareness Seminar on Computer Algebra and its Engineering Applications is being organised for Tuesday 3 April at Sheffield City Polytechnic (Pond Street site).
The AI Support for Engineers Community Club was launched at the one day seminar Why Should Engineers Use AI Techniques? on 6th December 1989 at RAL. The event was well attended with around 140 participants. The meeting was chaired by Miss Hilary Kahn, University of Manchester. The speakers for the seminar were Professor Ken MacCallum, Strathclyde University; Dr Roy Leitch, Heriot-Watt University; Mr Peter Jones, Rolls Royce and Associates; and Dr Eddie van de Kraats, Shell. All four talks were very well presented. Miss Kahn started the discussion of the Club by first presenting the Club's objectives which were:
She emphasised that the Club is about people meeting people, sharing their interests. There is no extra money to support new projects, but a small amount of money is available to support the organising and running of workshops.
Dr Paul Chung gave a fifteen minute summary of the background and current activities of the AI Support for Engineers project to set the scene that would help people make suggestions for what activities the Club should undertake. Once the meeting was open to discussion, Dr Allwood of the Department of Civil Engineering, Loughborough University of Technology, welcomed the idea of a Club. He said that the engineering community requires a channel and a voice to express its needs. For example, commercial IKBS tool developers are not providing appropriate tools for engineers. Engineers must say so loudly and clearly if things are going to be improved. He listed the tools that he had used in his projects but did not find them satisfactory.
Communication is an issue that the Club must address if it is going to be successful. It is encouraging to know that out of all the delegates only a small number of them do not have access to e-mail. This is a remarkable improvement compared to last year. There was some discussion about the advantages and disadvantages of direct email and electronic bulletin board. Delegates from industry complained that they have very restricted e-mail access, if any. It is not a good way of working for them. Another channel of communication is obviously this Newsletter. Some of the delegates were not on the mailing list so their names and addresses were requested.
One point that came across very strongly is that we need to publicise what AI can do but we also need to demonstrate what AI can do. Some people have already heard a large number of success stories but they have never seen a working system. Most of the 'successful' systems are industrial applications and access to them is restricted. Therefore, it is very important for the Club to develop convincing demonstrators and make them widely available. A number of people were prepared to give demonstrations and some were prepared to give talks.
Someone suggested RAL should do a search through the grant application forms to find out details about projects that apply AI techniques to engineering problems. This was thought to be a good idea but in practice it would be a very difficult thing to do. It was commented that the directory being compiled by AIAI, though not exhaustive, goes some way towards meeting this objective.
One delegate suggested the forming of special interest groups within the Club. The themes of design, analysis and real time application were suggested.
Dr Roy Leitch was approached to be the chairman of the steering group and he accepted the invitation. Eleven people put forward their names to be considered for the steering group.
The ex-officio members of the group are Dr Paul Chung and Mr Robert Rae of AIAI, and Dr Gordon Ringland of RAL.
The steering group was asked to draw up the terms of reference for the Club and publish it in this Newsletter. One thing that needs to be worked out is how people from industry fit into the Club. It was suggested that links to industry via DTI would be an appropriate way of working. Some of the current DTI activities under the Manufacturing Intelligence Programme are also relevant to the Club.
In Issues 8, 9 and 10 of this Newsletter we announced that an AI in engineering bibliography was available for copying over JANET from AIAI. The response has been good. Our log shows that there are around a hundred people who have copied it. Since then the bibliography has been extended and the second release is now ready. In this new release, the number of entries has increased from around three hundred to a thousand. The bibliography files, in Unix's REFER format, are available electronically. To receive a copy, send us your request by e-mail and we will send you the files using your return e-mail address.
The course schedule in the first six months of 1990 is:
These short courses on AI arc provided as part of the EASE programme and they are free of charge to suitable applicants who are working on projects funded by SERC grants.
The Computing Facilities Committee agreed last year to set up a Community Club in Computational Fluid Dynamics (CFD). The inaugural meeting of the Community Club will be held at RAL on Thursday 22 March 1990. It is intended that the programme for the meeting will contain presentations on current activities and future plans in CFD within each of the main Engineering Board areas which support CFD, viz Electromechanical Engineering, Environment, and Process Engineering Committees. There will also be reports on related activities such as ERCOFTAC (the European Research Community on Flow, Turbulence and Combustion).
An Advisory Group in CFD, chaired by Prof Hutchinson of Cranfield Institute of Technology, reported to the Electromechanical Engineering Committee in mid-1988. This was followed by a Workshop in CFD in early 1989. Both the report and the workshop recognized the multi-disciplinary nature of CFD, and subsequent discussions led to the recommendation that CFC should set up a Community Club in CFD.
The role of the Community Club is to help researchers in CFD by:
The Club will have regular meetings, probably twice yearly. It is intended that a small steering group will be set up to oversee the Club activities and to report to CFC.
This issue of the Newsletter contains, as a separate insert, a registration form for the CFD Community Club. Please complete and return this form if you wish to become a member of the Club. An application form for attendance at the meeting will be sent to all who indicate on the registration form that they wish to attend, and will also be available in the February issue of the Newsletter.
The 2nd International Conference and Exhibition: Transputer Applications 90 11-13 July 1990, University of Southampton The conference will be organised to include invited speaker presentations, which do not overlap with other conference activities, and parallel sessions for contributed papers.
Invited speakers include:
Applications given particular emphasis will include, but are not restricted to:
Multi-transputer applications will be given more emphasis by referees than single-transputer applications.
A large exhibition of transputer hardware and software will run concurrently with the conference.
SERC Seminar: IT in Geotechnical Engineering, 1 March 1990, University of Durham.
Sun New User Course : the next presentation of the course for new users of EASE Sun Workstations is scheduled to be held at RAL on 21 and 22 February 1990. The first day offers a basic introduction to using SunOS for the benefit of users with limited experience of the UNIX operating system. The principal aim of the second day is to introduce the concepts and skills involved in working with window based software. It offers attenders an opportunity to acquire and practise these skills through the use of some general purpose tools.
