Feedback: your letters
Here at E&T towers we've been observing your enthusiasm to tackle the bigger engineering subjects of our time like: the social cost of outsourcing; is the law too ambiguous to be computerised? Ethical engineering and what happened to The Senster?
Tax the profits of outsourcing
In 'Counting The Cost' (Vol 4 #11) Malcolm Wheatley ignores one major component of manufacturing: the social cost. Outsourcing is a euphemism for destroying the manufacturing base of this country, and every engineer should quail when he hears the term in case his job disappears over the eastern horizon.
I live in a predominantly rural area and 20 years ago it was dotted with small factories employing between 50 and 100 people; some were subsidiaries of major companies and some were suppliers to national companies. All are now gone, their buildings empty and decaying, their yards empty of transport and their skilled staff driving taxis or serving behind the bar in the pub. Now he says "Whoops, we maybe got it wrong". But Humpty Dumpty, once destroyed by bean counters' twisted logic, cannot be wished back together again.
We must ask whose 'bottom line' bears this loss and the answer is, of course, UK plc. We as taxpayers cover the costs of unemployment, community destruction and the social ills arising from these.
I would expect any manager worth the name to consider that aspect of his business or his suppliers' businesses when he decides to 'outsource' and if he then confirms the choice he should make the payment of all tax on his increased profits a gift to the local district council who are picking up the pieces. Otherwise it is the rest of us who provide the extra profit provided by that choice of his.
Martin T Gilbert CEng FIET, Totnes, Devon
Why the law can't be computerised
Ian Pearson is mistaken in his belief that " [computerised] law-making would eliminate ambiguity" ('What the future holds', Vol 4 #11). Ambiguity is inherent in the law because it deals with common sense concepts such as intent, diminished responsibility, consent etc. As a consequence, there is not a one-to-one mapping of the law onto a real-world situation, but significant room for interpretation.
An analogous situation confronts engineers when they use the laws of physics to model the infinite complexity of the world: deciding what aspects are negligible and what are essential is an inescapable engineering skill.
Choice arises from the modelling process, not the physics. Engineers can test their modelling decisions by experiment. Lawyers, on the other hand, choose an interpretation of the law that favours their case. The acid test of their interpretation comes not from experiment, but the decision of the jury. The inherent ambiguity of legal interpretation is an another example of the inherent ambiguity in interpreting any natural language utterance.
Computer-based speech and language technologies do not and cannot eliminate this ambiguity, but they do use a variety of heuristic, and hence fallible, techniques to work around it. We cannot expect better than this from attempts to computerise legal reasoning.
Alan Bundy FREng FIET, School of Informatics, University of Edinburgh
Snags of tags
'Personalised pricing' might indeed bring some "consumer backlash" ('Machine for shopping, Vol 4 #11). This is obviously a plan to try and distort the free market by the introduction of an anti-competitive practice. Divide and conquer the customers. The solution to this is the application of existing competition laws and the consideration of introducing new specific legislation if it is needed. We elect a government to moderate the free market when the balance of power has become unacceptable.
Far from "in return for privacy" the law could require the disclosure of all the information at time of purchase in order to complete a valid contract. One can imagine the assistant or self-scan terminal requiring confirmation of the lowest price sale in the last 24 hours before the item can be bagged.
RFID tags may also please customers and vendors in another way. Imagine entering a clothing store knowing that the machine has read the RFID tags on all items of your clothing and banknotes. It then displays them on a screen in front of an assistant who waylays you in order to make a sale.
The solution is the same. Require a life-size screen giving all the size and price information for all to see. One can imagine a few people finding display of their vital statistics extremely convenient since remembering them is so difficult!
K Cadman, Marlow, Buckinghamshire
A question of ethics
I was disappointed to read Robert Winston's comment that "We don't teach physical scientists the ethics of science the way we do with medical scientists" in your report on the opening of the new Centre for Synthetic Biology and Innovation at Imperial College (Vol 4 #11), as it doesn't help move the question forward. Given the low percentage of evidence-based medicine that is practised, one could cynically throw it back at as "We don't teach medical scientists the evidence base of science the way we do with physical scientists".
This soundbite approach to ethics might make for good tabloid journalism, but it is unlikely to solve any problems. We do need to include better ethical training, but the problem is not just within the scientific community. We also need to look at the decisions of those who deploy science - politicians and advisors, who are often barely literate in technical terms.
One of the great advantages of being taught an evidential basis is the light it brings to risk management. The questions of what we do with science are rarely black and white, yet decisions tend to take the same soundbite approach, where the solution is either good or bad, for us or against us. I suspect that to prevent some of the abuses of the past it's not just a case of teaching physical scientists ethics, but making sure those who use their work are equally ethical and educated to make better decisions.
Nick Hunn, WiFore Consulting
Dennis McMahon (Letters, Vol 4 #11) is quite right in urging for more publicity for the issue of domestic hot pipe insulation. This applies as much if not more so to the hot water runs between the hot tank and basins, and the pipes between the boiler and hot tank. I found that I had to run off 2.5 litres of water before the water reached its normal hot temperature of 45°C, and my house is not particularly large.
Every time the hot water pipe run is filled by using a hot tap, heat is lost immediately afterwards from the heated and unlagged pipe run. Given an ambient cold water temperature of 13°C and the specific heat of water, I calculated that the resulting energy loss is 0.09kWh on each occasion, which could approach 1kWh per day in many households, at a cost of about 5p per day or £18 per year given the current prices of oil or gas.
Tim Denvir CEng MIET, Killin, Perthshire
J Weaver, in the reply to my letter published in E&T Vol 4 #12, claims that heat lost from unlagged pipes mostly goes into the house structure. I disagree. Heating a building efficiently entails directing heat in a controlled manner to where and when it is required. The space under the ground floor of a house is draughty. If J Weaver wants to convince me that heat lost from pipes here will "mostly" find its upwards through the insulation of boards and carpet, and not be blown outside through air bricks, I should like to see his analysis.
Heat lost from pipes between storeys will not benefit rooms below; it will simply heat unnecessarily bedrooms, mostly unoccupied by day, which in any case will receive background heat from rooms below. Water passing through unlagged pipes loses temperature, so radiators will be less hot and effective.
I have heard it argued that low-energy lamps are an unnecessary economy because GLS lamps contribute to heating the building. Likewise, why worry about unlagged hot tanks or overfilled kettles? Arguments like this do not bear rigorous analysis.
Control is key to efficiency. Indiscriminate leakage of heat from pipes is not control.
Denis McMahon, Basingstoke
Birds' eye view
National Geographic has recently reported the discovery by scientists of an interaction between cryptochrome and superoxides present in birds' eyes that could allow them to 'see' magnetic fields.
I am intrigued to know what effect man-made electromagnetic radiation may have on bird navigation and behaviour. While I haven't noticed seagulls following me while on my mobile, I wonder if the national decline in sparrows could be related? If birds can actually see EM waves, then perhaps I can sell the avian equivalent of neon signs - The 3G bird table (with an iPhone app, of course).
Sam Lanyon MIET, Falmouth, Cornwall
In 'Complex Communication' (Vol 4 #11), Peter Grant writes that in the early days of GSM digital mobiles, the data rates of 14.4kbit/s were offered. In actual fact, while the system was capable at launch of 14.4kbit/s most networks offered only 9.6kbit/s. Operators at the time were offered 14.4kbit/s and possibly higher (up to 57.6) as a software option called high-speed circuit-switched data (HSCSD) by the equipment manufacturers but many did not take it up. Initially, most handsets were only setup for 9.6 due to error correction methods, and so operators regarded the software option as a frivolous expense. Only later when manufacturers heavily discounted these features did it become more widely used.
I used to be on Orange with a handset that supported HSCSD. Orange marketed HSCSD quite poorly even though it did deliver 14.4 which at the time was a 50 per cent increase in the rates initially offered. It was a useful tool to use all the bundled minutes offered by Orange which I used to check my POP3 email. Web browsing was possible but difficult, and until the proliferation of spam, I continued to use the service.
The days of increasing mobility and bandwidth are a welcome change from those days, but we should all remember that we used to do a significant amount of business on a simple 9.6k connection!
Mark Paulton MIET, Perth, Australia
The robot mentioned in Arthur Moore's letter (Vol 4 #13) was called 'The Senster', and was created by cybernetic sculptor Edward Ihnatowicz. The site at Senster.com [new window] describes the fate of the sculpture - and a friend managed to find the location on Google Earth with a bit of sleuthing. There's a video of the Senster in action on youtube [new window].
Amazing stuff! Its a real shame it hasn't been preserved in working order. Maybe some bright entrepreneurial spirit will resurrect it - or help the students who are described on the Senster website.
Stephen Wolff, Dorchester, England
'Historic Building Monitored without Cables' (Vol 4 # 8) reports that the Society for the Protection of Ancient Buildings is using wireless sensors to monitor environmental conditions at its Grade II listed Georgian headquarters in London. So what's new? I worked briefly for Hanwell Instruments Ltd in Hertford during 1998-99. At that time the company had a well established system to do just this, so well established that it is the standard for the National Trust and is much used by museums. Versions are available for vehicles, for individual packages and for a range of other types of business. All British designed and made in a BS/ISO 9000 certified environment!