[UKUUG Logo] Copyright © 1995-2004 UKUUG Ltd

UKUUG


The Newsletter of the UK UNIX User Group


Volume 5 Number 1 (February 1996)

 

UKUUG News

 

Editor's Column

(Susan Small)

I was delighted when Steve Kilbane offered to give me a report of the UKUUG Winter Conference held in York last December (see page 9). It is not often that a newsletter editor encounters volunteers and I am grateful to him for taking the time to do this. If you would like to report on local activities please feel free to do so; or if you want to review a book or product, let me know and I will do my best to get hold of the item. O'Reilly books feature heavily because they readily give out review copies, but let me know what books from other publishers you would like to see reviewed.

Check out the back page - it contains the Call for Papers for the UKUUG Winter Technical Meeting in December 1996.

It was like a Hollywood movie at home tonight - "Hold the front page" announced the answerphone. The stop press news which broke, just as the newsletter was going to bed, was the announcement that your Council have appointed Martin Houston to fill the vacant position, created as a result of Jim Reid's resignation. Martin will be well-known to most of you as the Linux SIG organiser and, in fact, had already contributed a piece to this issue. A warm welcome to Martin and we all look forward to his contributions to future Council meetings.


Report from the Chair

(Mick Farmer)

One of the drawbacks in being intimately involved in running a UNIX user group is that you have to read virtually every article that is written containing the word UNIX in case it contains something that may affect us. At the end of last year it was all doom and gloom, with the release of Windows 95 heralding the demise of UNIX, as Microsoft increasingly attempts to dominate our computing environments. A recent announcement that Oracle was no longer going to support UNIX as one its major platforms did not help either. On the other hand, the increase in Linux systems over the last year is extremely encouraging for those of us who want choice in what computing environments we use. I was especially lifted by Martin Houston, your Linux SIG organiser (see his article elsewhere in this issue), who said that a recent survey revealed that nearly 10% of the hosts connected to the Internet are running Linux. I suspect that most of these are around the edges rather than inside (which is still dominated by Sun), but it is a most encouraging sign that UNIX is far from dead.

Value for money

Your Council has, once again, been loooking at membership services. We have decided that existing services and our proposed new ones can be financed without increasing subscriptions for 1996 (see the article by our Treasurer, Ivan Gleeson, elsewhere in this issue) which must be good news. At this point it is probably a good idea to review the current categories of membership and what services you receive as a member.

Membership categories

Member services

If you have any suggestions for further member services that you would like us to offer, then please contact your secretariat.

Forthcoming events

Following the success of last year's meeting devoted to mobile computing (see review elsewhere in this issue), we intend to hold three major events during 1996.

Now it is your turn

Above, I have talked about a number of initiatives that involve you, our membership. So, now it is time for you to respond to these initiatives. I look forward to receiving a deluge of mail and messages from every one of you. If you want to comment on the services we provide for you, why not write a letter to your newsletter editor?


 

Report from the Treasurer

(Ivan Gleeson)

Hopefully by now you will have received your membership renewal invoice. I sure that you will notice that the subscription remains the same as last year. Membership has been hovering around 350 for some time. Consequently, the council have agreed to implement a strategy to achieve a steady growth in membership over the next couple of years. In part, this will be achieved by redefining existing services, introducing new services, and providing an effective events programme for the new year.

All this for the same price! Is it possible I hear you ask?

The costs for implementing such a strategy will be met by utilising existing funds and profits from future events, together with a reduction in administration costs. However, only your continued support, through membership and attendance of the events organised will ensure the survival of UKUUG. I hope you can join us next year.

If you would like further information or have any useful comments to make please e-mail, write or phone.

For the last three years I have been involved with UNIX and, as an IT Auditor with the Bank of England, tend to concentrate on security. I am a member of both the Institute of Internal Auditors (IIA) and the Information Systems Audit and Control Association (ISACA). I currently chair the ISACA UNIX Special Interest Group.


 

The UKUUG Archive

(Stuart McRobert)

We have been busy over the past few months gaining further donations.

Sun SITE is now an eight cpu SS1000 with a GByte of RAM, with two archive filestore areas of some 61 GB built as RAID 5s for resilience, with trans (logging) filesystems which are themselves three way mirrored, also for resilience. We plan early in 1996 to grow to around 200 GB and will be a central park for the Internet World Fair '96. Further, in order to improve global connectivity, we hope to directly link up a fast line for international traffic (all part of the World Fair) which will greatly assist our mirroring and allow many more users to reach us from all around the world.

Stuart McRobert is Head of Systems in the Department of Computing at Imperial College, London where he manages both the day to day system and network support teams and implements the future. Stuart's hobbies include building and planning the next generation of archives, since in their spare time both Lee McLoughlin and Stuart manage one of the world's most successful and ever expanding Archives on the Internet SunSITE Northern Europe.


 

Cambridge LUG

(Jane Shute)

The next meeting is being hosted at the APU in Chaucer Road. The date is to be Thursday 29 February 1996 and the start time is 1700, as usual. Mike Roberts, Sun's Customer Care and Quality Manager has offered to do a presentation on 'Sun's focus on Customer Satisfaction'. Mike says "I know it sounds dry, but I'll try to make it fun!" Full details are available from Jane.

Jane entered IT in 1985 after completing a degree in Physics. She was employed by Logica for four years as a systems manager and later as office systems manager before leaving to form Shute Associates. She has two kids, 18 and 13, one dog called Samson - a soppy Golden Retriver. She likes travelling and beer, and used to drive racing cars (but has now resorted to 4x4). She is also the Editor of the Sun UK User Group newsletter.


 

London LUG

(Andrew Findlay)

The usual meeting of the LUG will be held at the Rising Sun on Thursday 29 February 1996 from 1900hrs.

However, note this future date in your diary now: our June meeting (27 June 1996) will be preceded by the UKUUG AGM at the Institute of Education (University of London), Woburn Place, London, WC1. Expect a topical speaker in order to boost the numbers at the AGM! Sandwiches and light refreshments will be provided before the AGM.

Andrew Findlay organises LUUG events, barndances, and ox-roasts. Between these momentous events, he is in charge of the team that deals with the exponential computing requirements of Brunel University.


 

Linux Users Group Meeting

(Ted Harding)

There will be a Discussion Meeting of the Manchester Linux Group on Tuesday 20 February 1996 from 1900hrs - 2100hrs. It is intended for people to air their problems, solutions and views. All are welcome.

It will be held in Room G9, of the Manchester Computing Building, Oxford Road, Manchester. From Oxford Road, walk beside the Mathematics Tower (North side) and find the "side" entrance to the Manchester Computing building. Please sign-in at the Porters' Kiosk on entry.

With thanks to Dr D E ("Nobby") Clark for making the arrangements.

Hoping to see plenty of you there!


 

Linux SIG

(Martin Houston)

Sue asked me to write a short piece for news@UK on how the Linux SIG is getting along in the year since its inception.

Firstly, for anyone who does not yet know: Linux is a whole UNIX-like system made up from components that are Freely Redistributable under the Free Software Foundation General Public Licence. This means that a completely functional Workstation complete with X Windows, C and C++ compilers, text processing software and much more besides can be made out of a standard (and not even very powerful) PC, by using a collection of Linux software free via the Net or, for a few pounds, on a CD and last but not least a large dose of daring, expertise and a little luck!

I say a little luck because PCs being the variable beasts that they are a combination of software that works fine on one PC may fail miserably when tried on another.

One of the more frustrating things about running the Linux SIG is hearing tales of woe from people that can't get their machines to run X Windows properly or can't seem to be able to print. These are things that if you have a good quality Linux release and compatible hardware should work every time. What must be remembered is that Linux, with its virtually zero entry cost, is attracting many people who are fundamentally dissatisfied with the Microsoft dominated world of the PC but do not have the years of experience of coaxing a sometimes reluctant UNIX into behaving itself. Possibly it is the installers' ability that is even more variable than the PCs?

Moving your PC to Linux is a bit like trading in a simple but boring family saloon for a luxury sports car. Yes you can have more enjoyment from a high performance car but they tend to need more looking after to keep getting the best from them.

The SIG has had a fairly slow start in its first year and we only managed to produce three instead of the intended six newsletters. I wonder if all UKUUG members are aware that there is no extra charge to join the SIG? Just tell the secretariat and you will be put on the SIG membership list.

There has been some success with local Linux meetings in Manchester, Reading and East Sussex/Kent. This is something we want to encourage further in the new year by getting the LUGs to organise and publicise Linux and free software orientated meetings, in the same manner as the London LUG managed back in October.

The biggest success of 1995, which will bear increasing fruits in 1996, is the great efforts we have made to bring Linux to the attention of the general PC- using public. A year ago Linux was considered too esoteric to interest mainstream PC magazine readers, but 1996 will see Linux articles and Linux software appearing in several major magazines.

With the Windows 95 hype, Microsoft had the attention of the world pretty much to itself last year. However, if increasing numbers of people can be convinced of the prudence and choice of choosing open rather than proprietary computing technologies, then Linux, and the Internet on which it is carried, will trigger a re-birth of popular interest in UNIX.

Linux is on the whole remarkably stable, considering it is developed in a state of organised anarchy! Without Posix and FIPS certification and strict change control policies it may never threaten established UNIX systems in mission critical roles. But it is more than good enough for non-mission critical tasks, such as original software development, Internet access and general personal productivity tools.

An interesting scenario would be a single, commercial and supported central UNIX machine or machines providing a mission critical core, surrounded by Linux machines running the client side of client server applications. In other words, replacing the mixed and uneasy world of UNIX on the server and Windows on the desktop, with the tighter and more cost-effective fit of UNIX on the server and Linux on the desktop. A combination of consistent known hardware and a stable Linux system that is not updated at every whim, would be a very useful technology base that may have a greater learning curve to set up in the first place, but would be a lot less of a headache to look after than any network that has to bring the alien worlds of Windows and UNIX together!

The beauty of a Linux system is that even an old 386 based computer, costing a couple of hundred pounds, works in essentially the same way and is thus excellent training for massive UNIX data processing systems or super-computers.

Linux is being promoted hard in this way in many Universities - if a student has a PC they can be set up as a Linux system with all the software tools that they need to do really useful work.

In a few years time these students will be making major buying decisions and their natural preference will be for 'Linux-like' rather than 'Windows-like' systems. Linux, even though most of the software is given away, could well be the salvation of UNIX as a popular operating system, when faced with the opposition of the Microsoft juggernaut. Even if people start with free, but essentialy unsupported software, their needs will change. They will get busier and less interested in working things out for themselves. Letting people grow an increasingly sophisticated canon of free software running under Linux, does not kill a market for commercially tailored and supported packages, it enhances it!

Martin Houston is 33 years old, married with a daughter and four cats! A UNIX Systems programmer and system administrator for 13 years and organiser of the UKUUG Linux SIG since 1994.


 

News from Owles Hall

(Jane Morrison)

There are two events to report on since the last Newsletter. The Winter Conference in York, held at The University of York, was not very well attended, but the event was voted a definite success by those present. There was a lot of lively exchange, not only in the meeting, but also in the bar until quite late on Tuesday evening. The Conference dinner was a little intriguing - "A Taste of Yorkshire", but we are still trying to work out exactly which item of food related to Yorkshire! We did have roast beef, but not any Yorkshire Pudding!

The Rik Farrow Seminar - UNIX & Internet Security - was held again in London in early January. After last year's seminars, when we had to turn people away because we couldn't fit any more into the room, this year the numbers fell quite significantly. We can only put this down to the date, which possibly was too soon after the Christmas and New Year holidays.

The 1996 Annual Subscription invoices were sent out to you all early in January and we trust that prompt payment can be made. Please contact us if we have the wrong address or contact person, or if in fact you have any query relating to the payment.

Your Council, who met during the Rik Farrow event, have been planning events for 1996 and I'm sure that full details of these forthcoming meetings will be found in our Chair's report.

With plans for the AGM in June and three other events to organise, it looks like a full year ahead.

As reported in the last issue our Internet connection was about to go live, but as these things have a way of being delayed, I can report that this will now happen next week!

Jane has worked at the Owles Hall Secretariat for almost 8 years. She looks after the administration for the UKUUG, SUN UK User Group and EurOpen. When not working, her pastimes include gardening, swimming, painting and decorating, going on holiday and doing absolutely nothing and trying to keep her 18 year old son on the straight and narrow!


 

Reviews

 

UKUUG Winter Conference

Location Independent Computing

(Reviewed by Steve Kilbane)

There was a very disappointing turnout: less than thirty people, including delegates, speakers and user group folks. I know it was the week before Christmas, but where were you?

Professor Peter Honeyman of Honeydanber UUCP fame gave the keynote speech, describing his work on mobile filesystems. This was an interesting talk based on extending the concept of connected/disconnected states to include partially-connected and "fetch-only" intermediate states. Peter's premise is that if your underlying architecture, in this case the filesystem, handles the problems of occasional connectivity, then your applications don't need to. He's not entirely successful here, since he still has to decide whether to store or send e-mail, but it sounds good. His implementation is based on a modified AFS system, and seems to work; his cache is, if I recall correctly, about 20MB, and only half full.

Chris Cook from Vodafone described their new callout kit, the systems that field engineers take with them in the field. He made us all envious by pointing out that, not surprisingly, Vodafone find it cheaper to use their cellular network than standard public networks, such as BT. Their resulting kit is a small laptop connected via a PCMCIA modem to a GSM telephone. Security was provided using SecureID cards. These are credit card-sized devices which display a pseudo-random number. They can be used as a one-time access key, in addition to the normal username and password. Once successfully authenticated, a dialback system is used.

Andrew Macpherson of UKUUG and EurOpen described how EurOpen are planning to act as a PGP Certification Authority (CA). The point is that while a PGP signature allows you to trust that any two given messages originated from the same source, it doesn't let you trust that the source is who it says it is. PGP allows an intermediary to sign the source's key, in order to vouch for the source's identity. EurOpen are offering to act as this intermediary for their members (well, for their members' members). It should be noted that EurOpen are making it clear that they're only saying that:

EurOpen are also arranging things so that if you want to challenge their "good reason", you'll need a court warrant to get this reason out of them. In other words, they're taking this seriously.

There was a bit of discussion about how far you can take this level of trust: if EurOpen signs a University's key, and the University has signed each of its students' keys, can you trust a student of the University to be who they say they are? It was suggested that in order to trust the chain of signatures, you needed to know the criteria each link in the chain uses to determine validity of the claim. In EurOpen's case, they're using legal documents for identification a passport is the preferred one.

A couple of Birds of a Feather (BOF) sessions followed, for Linux and Plan 9. Martin Houston, the UKUUG Linux SIG coordinator, led the former, while I led the latter, with Charles Forsyth of the University of York giving a demonstration of their Plan 9 installation. In both cases, most of the allocated hour was given over to a description of the operating system in question, followed by a short amount of discussion. Linux is a freely-available UNIX-like operating system that runs mainly on 386s and up, while Plan 9 is Bell Labs' successor to UNIX. Not surprisingly, there were significantly more people who have tried or used Linux than there were who had heard of Plan 9.

The second day kicked off with Donal Daly from Oracle, Ireland, who described their "Oracle Mobile Agents" product. This is a commercial product, and is essentially a toolkit for building applications that use a client-agent-server model instead of client-server.

The client is assumed to be a mobile machine which is often disconnected from the network. The agent and server both live in a constant network environment, and the agent handles the client's occasional appearances, passing requests to the server, and returning responses. If the client has vanished since submitting the request, the agent holds responses until the client puts in another appearance. Donal made the point that this requires a slight shift in applications design, to reduce the number and size of exchanges between client and server: it's not ideal to fill in a form entry by asking for a full list of options, and then selecting one, because that's a significant amount of traffic when you add up all the entries.

Simon Kenyon of Koala Systems digressed from the conference theme to describe his own approach to web-based virtual reality. He rejected VRML, because of the processing power necessary to get reasonable performance from a VRML browser. Personally, I don't have a problem with VRML browsers. This is because I've yet to find a VRML demo that's worth the time required to download the enormous files, never mind getting around to running them on a browser...

Simon's system is based on imagemaps. A view of the user's current surroundings is generated on the fly, and updated each time the user "moves", by clicking on one of the controls on the screen. The images use flat shading instead of textures, and aim for a simplistic image that compresses well. Apparently, 4K is a typical size of a screen.

Simon also gave some interesting rules on web site design, and provoked some heated comments over what makes a site attractive. There was some contention over whether Simon's method would negate any benefits a proxy server could provide, making the site seem slower and less attractive as a result. The consensus seemed to be that this was probably slow, but that the cacheing in proxy servers was broken in this respect.

Dr Andrew Findlay of Brunel University gave an entertaining talk on Brunel's approach to solving provision of computing resources. Apparently, while getting hold of more computing equipment is hard, it's not as hard as getting hold of somewhere to put it. Brunel's possible answer is to provide each student with a PC, and let them worry about where to store it. Many students are in Halls, and some Halls are already wired up for network connections. For those living outside, a trial is underway with a local cable-TV company acting as a network provider, somewhat similar to Piete Brooks' experiment (described at the Oxford conference two years ago).

Provision of the PCs themselves is also a headache. The University does not want to simply give each student a machine, because the machines will be obsolete by the time the students are finished with them. Instead, some form of leasing is being considered.

Chris King spoke on the subject of security worries with the EuroISDN standard that our various telecoms providers have agreed to. This describes which of the normal ISDN standards we'll be getting, and when.

This was an unnerving talk, and the underlying question was "how much do you trust the person who is in control of the hardware you're talking to?". Mr King pointed out various ways that ISDN allows your connection to reach a rogue unit instead of the trusted unit you intended it for. As a simple example, ISDN normal drops the logical connection when you physically unplug the unit. However, there is a facility to physically disconnect briefly in case you want to, say, move your phone. It's therefore possible to have someone wait until you've connected to the trusted unit, then unplug that trusted box and replace it with a rogue one.

Unfortunately, while Phase One of EuroISDN supports the facility to determine the address of an incoming call, it does not support the facility to determine the address of the box you're actually now connected to; that isn't until Phase Two. Not that this will help much, however, because the various Caller ID services are also susceptible to certain forms of spoofing. The moral here is to provide solid security at the application layer too, and not to trust the hardware layer.

The final presentation was by Steve Platt, of Olivetti Research Labs. This was essentially the same talk as Steve gave at the Sun UK User Group's one-day meeting at Warwick University in September 1994. It concerned Olivetti's "teleporting" concept, where a proxy X server allows a user's actual X session to follow them around, appearing on whichever screen was convenient. Because X clients are talking to a proxy server instead of a real one, they're generally not usually aware of the change. Olivetti use "active badges" to make this easier: the badges are fastened to people and machines, and transmit their location every fifteen seconds. Thus, the teleporting software can locate the nearest screen to a person by default, rather than the user having to select it explictly. Unfortunately, technical problems and lack of time prevented Steve from giving a decent demonstration of this system.

Steve Kilbane is a software engineer and Plan 9 evangelist at Cegelec Projects Ltd, where he fiddles with the UNIX/X Plan 9 clones. Steve is a gradute of York University, so was one of the few attendees who knew where to find the place. He can be reached as Steve_Kilbane@cegelecproj.co.uk.


 

MH & xmh

E-mail for users and programmers

Third edition

Jerry Peek
O'Reilly & Associates Inc 738 pp
ISBN 0-56592-093-7

(Reviewed by Andrew Macpherson)

MH is a set of command line tools for handling mail. xmh, exmh and MH-E are a set of front ends to simplify the interface to the point where the non- expert can do most things; though one may still want the raw interface for "Wizard" tasks such as making up mailing-list digests. MH uses commands such as scan to list the contents of a mail folder, show, next and prev to view messages, comp to compose, etc. MH-E is a mode within emacs, and I'll leave it at that, and xmh/exmh are graphical front ends. xmh is a core demonstrator of X11, though fairly limited, while exmh is a front-end written in Tcl/Tk with a very full MIME capability (it can also read the pestilential sun-mailtool attachments).

The First Edition of the octopus book was huge this is enormous. When I reviewed the first edition (q.v.) I concluded that MH was a very large package, and getting the best out of it difficult, but that there was a great deal of satisfaction, and massively increased functionality to be had from doing so. I was not wholly convinced that the benefit was worth the effort.

Since that time I have been using various flavours of MH, and have found the book an enormous help in customising the environment. The "xmh getting started" section has been borrowed many times by users who have gone beyond the capabilities of mail, mailtool or elm. What then does the book's third edition bring to the party?

First off, the new edition is much larger. It's a book to read flat on a desk or reading stand, not lying on the sofa; despite this it does not have lay-flat binding. It has a new section on exmh, a much fuller description of MH-E, and thumb-tabs for the various sections. Far the biggest section is on customisation, and this is wholly appropriate given the flexibility of the MH system. This is at least a change in emphasis, if not a major reshuffle.

I was looking forward to the new section on exmh (contributed by Brent Welch, the author of exmh) for tips on programming extensions. I now use exmh for preference, and there are a few extras I would like... I admit to slight disappointment there is much in the book to help write code, and it does take one considerably beyond the notes distributed with the source, but I think I'll have to get the "Practical programming in Tcl and Tk" book by Brent for when I need to do that.

Who should buy this book? I think that the power user needs the hints and tips, the maintainer will find it useful, and should have two copies on their shelf, and the average user will be totally overwhelmed. MH remains one of the most fully featured mail systems, and I will not be surprised if exmh is one of the first mail agents to implement all the new privacy and notification extensions which are pending in the RFC drafts.

Last time I wondered whether one should bother with MH. I am now convinced that exmh is the graphical front end of choice for UNIX.

Andrew Macpherson works for an International Telecoms Manufacturer. His interests are computer and network security, distributed access and messaging systems. He is active in the user groups, both UKUUG and EurOpen. He never did submit his MSc thesis.


 

The Computer User's Survival Guide

Joan Stigliani
O'Reilly & Associates, Inc. 276 pp
ISBN 1-56592-030-9

(Reviewed by Mick Farmer)

This is one of those books that you can't put down. Sheer fascination keeps your interest, or at least mine. I read this book on my daily journey to and from work, and on more than one occasion I found myself leaping out of the train at my home station, book in one hand and bag in another, having got so engrossed that I nearly missed the stop!

I never realised how dangerous our computers, and especially our workstations, were to our health. We've all heard about the dangers of Repetitive Strain Injury (RSI), we know that our screen should be positioned so that there are no reflections, and that our posture is important. However, in this book, Joan Stigliani warns us about all the problems we're likely to encounter when sitting in front of a keyboard and screen. The main sections in the book are reassuring:

There are two appendices listing where to obtain equipment and resource guides (mainly in the USA) and a suggested reading list that looked reasonably extensive to me.

Once you start reading, the subject takes on another dimension. The author has clearly researched her material in great depth - the acknowledgements cover nearly two pages. What hits you is the detail in the text. A single chapter entitled "Get Comfortable" describes body posture, head posture, arm & wrist posture, and selecting a chair. This last section covers not only chair selection, but also chair adjustment and what to do if you don't have an adjustable chair. Naturally, there's a section on posture and laptop computers.

The section on RSI is extremely detailed. After some initial paragraphs outlining the demographics of RSI and how different countries react to this, often debilitating, disease, the author describes all the different forms that RSI takes, from tendon injuries such as tendinitis through to cervical radiculopathy (the trapping of a nerve root) injuries such as carpal tunnel syndrome. Each injury is carefully described, many with diagrams, together with its symptoms. The author is far from alarmist. She describes how to look out for the early warning signs of RSI, how to change the way we work, and how to deal with any injuries. I didn't realise that many cases of RSI can heal; it takes a long time, especially if it wasn't diagnosed early on. The author is also strongly in favour of prevention rather than cure. There are chapters on the tools we use, not just keyboards and mice but also ergonomically- designed software, and how to encourage good hand technique in order to reduce the risk of injury. The final two chapters in this section contain good advice on treating RSI and how to deal with the, often long, recovery process.

The section on eyestrain is no less detailed. It was here that I noticed the author's holistic approach coming through. Problems like eyestrain may be related to other problems that we haven't detected. Once again, posture and a visually comfortable environment are all important. If you need glasses, use the right ones for the task; use a screen with good visual qualities; use software that's easy on your eyes; balance the lighting levels. The level of detail is amazing - there's even a couple of paragraphs describing how to get the best from your contact lenses!

The section on stress is the shortest, just two chapters. In the first chapter, the author covers the usual physical, emotional, and behavioral signs of stress, but also describes the other factors that affect possible stress at work, such as certain types of jobs, the physical environment, and even the corporate environment. The second chapter is devoted to reducing stress. The author's tenet is that stress begins in the mind so, once again, she advocates whole-body solutions including posture and breathing, exercise to release tension, and how to balance computer time and social time.

The section on EMFs is probably the most controversial. Although there have been a number of studies, anecdotal accounts, and observations, there is still no general agreement on how EMFs affect us. The first chapter describes the physical aspects of radiation and what forms we are exposed to. The author asks the question: "How much exposure is considered safe?", and shows that because no level has been proven absolutely safe or absolutely dangerous, the politics of EMF research means that different countries have different regulations. For example, the political and economic implications of lowering emission levels of the entire electric power infrastructure are vast. The second chapter is concerned with the health risks which are sometimes associated with EMFs, such as miscarriage, cancer, cataracts, stress, skin rash, and chronic fatigue syndrome. For each risk, the author discusses the likely link with EMFs and the possible symptoms. The third chapter deals with the preventative measures we can take. The author's recommended guidelines are: stay as far away from EMFs as possible; don't expose yourself unnecessarily; and use equipment and devices that have low emissions. There is even a brief discussion of unconventional devices such as surge protectors and quartz pendants.

Much of what the author says is not new. However, she stresses that many factors contribute to our health and that we need to examine them all in order to create a healthy working environment. We should set up our work area to be comfortable and develop good habits of working, sitting, and using our eyes. Employers should provide ergonomic furniture, provide health- related training programs, and set reasonable working hours, breaks and vacations. Complain about that flickering light. Move your workstation to a better position. Ask for a document holder. Joan Stigliani certainly provides the necessary information for us to better survive our computers.


 

Website Directory

Compiled by Lindsay Marshall (and suggestions from the Editor)

http://altavista.digital.com
My current favourite search engine - you can seach the news as well as the web.
http://www.ukindex.co.uk/uksearch.html
A search engine for UK-based sites. Very much faster than anything in the USA!
http://www.banned.books.com
A well put together set of pages on the topic of banned books. Also features RealAudio (tm) clips if you have the software.
http://www.poopets.com/
The Amish Web site!! Set up by the company that makes PooPets - those wonderful little animals carved from dried cow dung...
http://www.nwu.edu/WNUR/jazz/
A great resource for jazz fans.
http://www.milk.com/barcode/
Generate barcodes for fun and profit...
http://www.rmplc.co.uk/orgs/firstand/index.html
The Schools Internet page supported by Research Machines. Lots of stuff for schools (as you would expect).
http://park.org
The Internet 96 World Exposition. Just like visiting the World's Fair, Olympia, Alexandra Palace all rolled into one, without leaving your desk. Beware, though, there are a large number of graphic images and downloading a page can take ages.
http://www.hillside.co.uk/tour/tour.html
Pete Collinson's photographic tour of the historic city of Canterbury. Virtual reality run slow!

 

Across the Pond

 

10th USENIX Systems Administration Conference (LISA '96)

30 September - 4 October 1996
Chicago, Illinois

Announcement & Call for Participation

LISA, the USENIX Systems Administration Conference, offers the most comprehensive programme for system administrators from sites of all sizes and at all levels of experience. The focus is bringing system administrators the latest tools, techniques, and information needed to keep apace with the rapid technology advancements, changes in public and legal policy, and changes in the ways that their employers do business.

The conference will offer up to 20 tutorials on two days. Tutorials are offered on all levels of system administration from novice to senior system administrator. If you are interested in presenting a tutorial at this or other USENIX conferences, please contact the tutorial coordinator: Daniel V. Klein.

The three days of technical sessions consist of two parallel tracks. The first track is dedicated to presentations of refereed technical papers. The second track is intended to accommodate invited talks, panels and Works-in-Progress (WIP) sessions. Conference proceedings, containing all refereed papers and materials from the invited talks, will be distributed to attendees and will also be available from USENIX following the conference.

Presentations addressing the following topics are particularly timely; presentations addressing other technical areas of general interest are equally welcome.

Refereed paper submissions dates:
Extended abstracts due: 7 May 1996
Notification to authors by: 11 June 1996
Final papers due: 15 August 1996

For more detailed author instructions and a sample extended abstract, send email to lisa10author@usenix.org.

Authors of an accepted paper must provide a final paper for publication in the conference proceedings. At least one author of each accepted paper presents the paper at the conference. Final papers are limited to 20 pages, including diagrams, figures and appendixes, and must be in troff, ASCII, or LaTeX format. We will supply you with instructions. Papers should include a brief description of the site, where appropriate.


 

USENIX Lifetime Achievement Award

to The Software Tools Users Group

Principal Recipients and Keepers of the Flame:
Dennis Hall
Deborah Scherrer
Joseph Sventek

Originators and Key Inspiration: Brian Kernighan and P J Plauger

The USENIX Lifetime Achievement Award recognizes and celebrates singular contributions to the UNIX community in both intellectual achievement and unparalleled service. Past recipients of the USENIX Award are the Computer Science Research Group at the University of California at Berkeley for producing the UNIX BSD releases, Van Jacobson and Mike Lesk for their contributions to networking technology, and Tom Truscott, Steve Bellovin and Jim Ellis for their work in creating USENET.

Before the general availability of UNIX, the Software Tools project popularized a new vision of operating system software, offering a bridge to portability and power for those limited by proprietary operating systems.

With its extraordinary focus on building clean, portable, reusable code shared amongst multiple applications and runnable on virtually any system, the Software Tools movement established the tradition of empowering users to define, develop, control, and freely distribute their computing environment. The contributions of STUG in retrospect can be seen to have been vital.

Simultaneously, The USENIX Association, acting on behalf of The Software Tools Users Group presents The Software Tools Users Group Award to Michael Tiemann.

Michael Tiemann's work in C++ led to fundamental contributions to the GCC, the GNU C Compiler, which has had an unparalleled influence upon the availability of efficient and standard code on a vast number of hardware platforms. GCC has provided a development base for thousands of projects.

The Software Tools User Group Award recognizes significant contributions to the general community which reflect the spirit and character demonstrated by those who came together in the Software Tools User Group (STUG). Therefore, recipients of the Software Tools Award exhibit one or both of these traits in a conspicuous manner: a contribution to the reusable code-base available to all, or the provision directly to users in a widely-available form of a significant, enabling technology.

Steve Johnson, President of the USENIX Association, will be announcing both 1996 awards on Wednesday 24 January at the Annual USENIX Technical Conference, in San Diego, California.


 

From the Net

 

Space Shuttle Computers

(David K. Cornutt)

There are a few points to be made about the Shuttle computers...

First of all, the Orbiter has five primary control computers (called GPCs, for General Purpose Computers). One of these normally runs an independently-developed emergency software that is capable of handling ascent and entry only; it stays in a deactivated state unless the commander or pilot activates it with a stick switch.

The other four computers run in the 4-redundant configuration only during ascent and entry. While the Shuttle is on orbit, they normally are split off to do different tasks. Shortly after the launch phase is complete and a good orbit is verified, one of the four is loaded with the re-entry software and then halted. (This way, if both the mass storage units fail, there is an independent source of re-entry software.) One unit normally runs what's called the GNC (guidance-navigation-control) software, which handles the on-orbit navigation and manoeuvres. One runs the SM (system management) software, which maintains watch over the various components of the orbiter and handles commands and configuring of systems (including the computers themselves), and the data communications functions. The fourth unit can be configured with GNC software for a 2-redundant configurations for critical mission phases (say, a rendezvous with another spacecraft), or used to control certain payloads (the remote manipulator arm, to name one). If it's not being used, it is generally loaded with re-entry software and then halted.

In additition to these five units, there is a sixth computer on board which is stored in a locker. If a computer fails, they get the spare out and replace it (assuming there is time to do so). Note that, on-orbit, it's not a major catastrophe to lose a computer, even the guidance computer. (If that happens, the spacecraft just drifts around for a while until they get the function restored.) None of the critical life-support functions are computer controlled.

However, the orbiter is a fly-by-wire vehicle. During launch and re-entry, it needs to have at least one computer functioning, or control is lost. Some of the voting in the redundant configuration is electronic, but a surprising amount of it is mechanical. For instance, each computer operates a redundant actuator on the elevons. If three of the computers are trying to push an elevon up, and one tries to push it down, hydraulic valves in the actuator circuits will cause the non-conforming one to be mechanically locked out. (The computer senses this and turns itself in to the authorities. :-)

Final point: In previous space programs, computers and electronics had to be designed to operate in a vacuum, which would of course have eliminated disk drives from consideration (had any reasonable ones existed back then). This was not a consideration for Shuttle, however. All of the avionics in the Orbiter cabin, including the computers, are air-cooled and won't survive long without air. Unlike earlier spacecraft, which used an atmosphere of pure oxygen at a very low pressure, Shuttle's atmosphere is near Earth normal in composition, and its nominal pressure is just a bit lower than Earth-surface average (13.5 PSI absolute vs. 14.5). [They do at times lower pressure to about 10.5 PSIA in preparation for an EVA, and this often requires that non-essential electronics be shut off to reduce the heat load.] There are no circumstances where the cabin would be evacuated intentionally, and a leak big enough to overcome the life-support systems would probably be fatal anyway. So, although it wasn't designed to use disk drives for program storage, there is no reason why suitably shock- and acceleration-resistant ones couldn't be used today.

And about those laptops: I think they come from GRID. They have plasma displays, but are otherwise pretty normal laptops. And, although they don't use them to control the Oribiter, they do often use them to control payloads. The most notable example of this is Spacelab, where they are now used as terminal devices to control the Spacelab computers, and in this service have proved to be infinitely superior to the original data display units (which burned out during Astro-1 and jeopardized the mission).


 

Netfuture

[Ed's Note:
Netfuture is a moderated forum concerning technology and human responsibility for the future. Postings normally occur once per week. Moderator of the list is Stephen Talbott, an editor at O'Reilly & Associates. The following extracts give a flavour and may be of interest. Details of how to subscribe are given at the end of the article.
]

What does it mean to be responsible for technology?

(Stephen Talbott)

Yes, you are a responsible citizen of cyberspace. You don't send email bombs, engage in sexual harassment, dig through other people's private files, or willfully disrupt and destroy discussion groups. On the positive side, perhaps you even take seriously the advice (not heard as often as a few years ago!) to treat newcomers kindly, share your ideas freely for the common good, and cultivate some sort of global consciousness, whatever that means.

But are there concerns and responsibilities, perhaps even more urgent than these, yet less present to mind?

How interested in such issues is the larger Net community? It is difficult to know. Send us your thoughts, suggestions, and references. We look forward to your participation in this wide-open exploration of the boundaries of human responsibility in the presence of the ubiquitous machinery of our lives.

The Phenomenology of Computing

(Dirk Brandts)

I recently spent two months in pretty rustic conditions in Ireland, this after having spent a solid three years doing computer graphics in a high-stress environment. It took me a good week to lose the craving for network access, which manifested itself as an actual physical impulse. My conscious mind was quite content to lay on the grass and read or draw, but my body (!) felt a pull towards the machine. But I was relieved to learn that it was relatively easy, after all, to wean myself from the device. In a short time I had, as I called it, "liquified" myself. I wondered whether this might have been due to my background, having grown up in rural East Africa and learned in youth to savor all the rich stimulation that unmediated nature has to offer. Would a person raised solely on TV and computers find themselves in an empty world without those "tools"? Perhaps it depends on the individual.

Upon my return from Ireland I sat down in front of my machine and was shocked at the sensation! It was as if something were reaching out and enveloping first my hand on the mouse, then my arm, then my shoulder. Soon my whole physique had been (for lack of a better word) subjugated to the machine form and process. The phenomenon extended to my mind, in fact, and I could feel my loose relaxed thinking start to square up and re-align with the programmatic avenues of choice that the machine presented. I gave in, with a little whimper to my wife about the strangeness of the feeling, and have been hard at work computing ever since.

So I would say that there's definitely a quality associated with computer use that's different from any other tool. Consequently I practice (to use Arthur Kroker's term) "ironic immersion." Despite being fully involved with the machine, I reserve a strong sense of doubt and questioning. We live on a small ranch, and after work I go to some lengths to complement my digital days with farm chores, paddling around on the lake, gardening and so forth. I see it as preventive medicine, although I realize that even these measures may not be sufficient in the long run. I don't ever get as "liquid" as I did in Ireland, for example.

How to Subscribe

To subscribe to Netfuture, send an email message like this:

To: listproc@online.ora.com
subscribe netfuture John Donne

No Subject: line is needed. Substitute your own name for "John Donne".

Send general inquiries to netfuture-owner@online.ora.com.

Send submissions intended for netfuture itself to netfuture@online.ora.com.


 

Calendar of Events

1996

TBA May, Surfing and Serving the WWW workshop (UKUUG)

13-17 May 5th UNIX System Administration, Networking, And Security Symposium (SANS V), Washington, DC, USA

17-21 Jun, 2ND Conference On Object-oriented Technologies And Systems (COOTS), Toronto, Canada

10-13 Jul, 4th TCL/TK Workshop (TCL/TK 96), Monterey, California, USA

22-25 Jul, 6th USENIX Security Symposium, San Jose, California, USA

TBA Sep, Crime and Punishment seminar (UKUUG)

30 Sep-4 Oct, 10th USENIX Systems Administration Conference (LISA 96), Chicago, Illinois, USA

29 Oct-1 Nov, 2nd USENIX Symposium On Operating Systems Design And Implementation (OSDI II), Seattle, Washington, USA

TBA Dec, Network Management Conference (UKUUG)


 

Acronyms for 1996

AFS
Andrew File System
COOTS
Conference On Object-oriented Technologies And Systems
GNC
Guidance-Navigation-Control
GPC
General Purpose Computers
LISA
Large Installation System Administration
OSDI
Operating Systems Design And Implementation
PCMCIA
Personal Computer Memory Card Industry Association
PGP
Pretty Good Privacy
RAID
Redundant Array of Inexpensive Disks
SANS
System Administration, Networking, And Security Symposium
SM
System management
Tcl/Tk
Tool Control Language/Toolkit for the X Window System
UUCP
UNIX to UNIX Copy Program
VRML
Virtual Reality Mark-up Language
WIP
Work in Progress
Tel: 01763 273 475
Fax: 01763 273 255
Web: Webmaster
Queries: Ask Here
Join UKUUG Today!

UKUUG Secretariat
PO BOX 37
Buntingford
Herts
SG9 9UQ