VSJ – May 2004

Symposium Report

As regular readers know, Council member Paul Lynham traditionally writes a report on the annual symposium to which we devote the whole of ‘IAP News’. Here’s this year’s.

We met at Trinity House on 18 March. Director General Mike Ryan welcomed delegates and summarised recent changes including the new office, administrator and Webmaster. He also mentioned the significant amount of outsourcing to Asia and suggested members focus on customer-facing skills to differentiate them from remote working competition. Mike introduced this year’s chairman, Jim Bates, IAP President. Jim noted the magnificent surroundings afforded by Trinity House and introduced the first speaker, Richard Allan MP.


Richard began a career in Archaeology but became interested in IT and completed a government sponsored MSc in IT. He spent 6 years in the NHS, helping roll out Internet and email facilities. In 1997 he was elected MP for Sheffield Hallam. Only 26 of 650 MPs have a background in science or technology – just 4%!


He listed the parliamentary groups with an IT focus, including PITCOM, which IAP representatives attend. He noted the select committees that look at IT spending. The National Audit Office (www.nao.org.uk) analyses projects that have gone astray and concludes that decision-makers are light on expertise. Outsourcing has made the situation worse, with weak client-side input. This is worrying, as IT is a critical plank in improving public services.


Large projects have poor track records, WERS 2, Tax Credit and GCHQ being examples. In the latter case, a project with a £20m budget actually cost £400m. This has led to a perception that ‘Government IT’ is synonymous with failure. So projects must now be ‘green flagged’ through the Office of Government Commerce. The department initiating the project and the industry partner each need a ‘champion’.


PFI’s promise of eliminating Government risk has proved illusory. A new model will use private third parties, but ‘in partnership’. Thus, ministers can no longer say, ‘ We contracted out and they failed’.


Disaster has attended political ignorance. For instance, a political decision was taken not to test a benefits system so that it could meet a deadline. The result was that, although the user interface worked, when connected to the production database, it failed miserably and benefits were not paid.


NHS Direct is a successful example but this project did not affect existing systems. Richard believes IT can deliver better public services but it needs commitment and understanding to succeed.


Peter Welch, Professor of Parallel Computing, University of Kent at Canterbury, gave an interesting presentation entitled ‘Concurrency for All’. Peter demonstrated that nature is dynamic, with entities interacting from the microbial to the astronomical level. Such natural systems are robust, efficient, long-lived and continually evolving. Concurrency is a core design mechanism in which the interconnection of components is achieved with simple composition rules and clean interfaces. Hardware systems are built like this but software often isn’t. Software concurrency is only used at need.


In the OO paradigm, an external thread must invoke all methods. Thus it is really caller, rather than object, oriented. Concurrent systems overcome resulting problems by locking. However, too high a level of locking leads to deadlock, too little results in corruption. But concurrency can make things simpler and safer.


To demonstrate this, Peter used a version of Pong with multiple balls, travelling at different speeds. Analysing it, he showed the many interactions between the game’s components, including the canvas, the keyboard controller, the balls and the paddles. Messages allow components to interact but sub-systems can be used in isolation, without unexpected interference. Changes only affect the required components. If, say, colliding balls need to be redirected, a service to do this wouldn’t affect other components. Thus we can handle each process on its own and reuse our serial programming skills to achieve concurrency.


Peter used a synchronised communication example of an object ‘A’, wanting to send data to object ‘B’ using channel ‘C’. It could be blocked if ‘A’ could not write to ‘C’. This synchronous I/O is concurrency with ‘1 to 1’, ‘1 to many’ and ‘many to many’ flavours. It only uses a 32-byte overhead per process and timing metrics show delays only of nanoseconds.


This can be applied to modelling bio-mechanisms (one of the UK Grand Challenges), so that biological experiments can be simulated within a computer. A nematode worm could be studied from fertilised cell right up to its adult stage, observing its reaction to stimuli. Occam-Л and JCSP can be used to achieve this. CSP has compositional semantics and JCSP enables direct Java implementations of CSP.


Peter observed that OO is about 30 years old. CSP is 10 years younger. It’s now, he says, CSP’s turn to be the leading development paradigm.


Mark Aldington, Managing Director of Addventure Partners, presented a topic entitled ‘Quantum Cryptography – Could it Work?’ He said the jury is still out on what quantum cryptography really is. However he summarised it in one context as a point-to-point securitisation of fibre optic cable using quantum effects to detect eavesdropping. Senders need no common code with the inherent risks of third parties compromising the system, as the keys are thrown away!


QIP Cientifica is a knowledge broker between scientific and academic communities, institutions and commercial partners. They want to showcase this technology and its potential applications and would like to demonstrate it outside the laboratory. A proposed trial would involve the connection of the Bank of England with the Treasury using fibre optic cable. Participants could be given a message to send between them under the same conditions. Hackers would be allowed to try to intercept the messages and the results monitored by a supervisory team and peer reviewed. This might convince interested parties that the technology is usable and reliable.


QIP wants to promote the UK as the location of choice for commercial development of quantum cryptography systems. The UK has a good track record in invention (although not in exploitation). It also has a history in cryptography, with the example of Bletchley Park’s exploits in the Second World War, leading to the development of the computer. Mark gave a rundown of the largest players involved in the field including NEC, IBM, MAGIC Corp, BBN Systems, ID Quantique, Quinetiq, Free Air and the Singapore government.


The limitations of quantum cryptography are distance (recently extended to 150 Km) and medium (fibre optic cable). There is also the cost of new hardware. A further problem is that commerce does not think that it has a problem with key distribution!


Mark identified organisations that would be interested in this technology. These included financial institutions, defence and intelligence agencies.


He said that this could be good news for IAP members. Any new technology provides new opportunities and new standards, protocols and hardware interfaces.


Mark concluded by quoting Richard Feynman – ‘Nobody understands Quantum Theory’.


After lunch, Ken Abraham, Chairman of PICT Innovation, gave a presentation on investment for innovation. He conducted a survey of the people in the room, which reflected the result of research that found the largest percentage of people work for micro companies. These companies (maximum of 10 employees) often don’t have the infrastructure and resources needed to attract investment.


Ken went on to consider the European Paradox –high R & D expenditure but low commercialisation spend. The reason is the lack of specific skills and entrepreneurial culture. This leads to a lack of venture capital investment. The skills sought are Transition Management Skills, Information Management Skills, Networking Skills, Knowledge Management Skills, Investor Persuasion Skills and IPR Protection Skills.


Ken quoted a Swedish investment group (third largest in Europe) which says, ‘Any company failing to implement these skills will disappear’.


Ken referred to the ‘Push and Pull’ of investment. As part of the technology ‘push’, he listed the sources of funding – soft money – which were the company’s own funds, FFF (family, friends and fools), bank loans and angels. The demands on the ‘pull’ were investors and venture capitalists. Often the latter base their investment on the management team’s qualifications, in particular their general skills, marketing skills and R & D. If a company has at least two of these skills, there is an 80% success rate. Otherwise, there is a comparable chance of failure.


Collaborations must maintain good communication between the public and private sectors and academia. This is where PICT can help, with other funding opportunities coming from venture capitalists and the EU.


With EU funding, there are 4 phases. R & D are followed by market preparation, with phase 3 being the commencement of business. The final phase should be profit. The EU funds the first 3 phases. For example a partnership of 4 equal participants put up half of each of their quarter share of costs (time, effort, money), while the EU funds the remainder. Thus each partner only has to find an eighth of the required funding.


Ken pointed to the success of the ARCTIC project that has 7 partners worth 6m Euro. Their product involved synchronous objects in Java and C++. One of the partners is a small Highlands company that now shares 4% of the IPR, which was only possible with the partnership and EU funding.


PICT gets involved when projects have stalled. It shares the IPR. Additional funding is provided and repayment made when viable. However PICT’s main aim is to get the company into a viable position and let them have the opportunity of buying back the IPR.


Another example is the THINK project worth 5m Euro with involvement from Portugal, Spain, Italy, Greece and Scotland. They are involved with training and work opportunities for people with disabilities.


Ken concluded that not all companies want to expand rapidly, but may want to take the lifestyle route.


Peter Ananicz and Don Tricker, founding directors of e421 Ltd, gave a presentation entitled ‘Tackling the Public Sector Market’. Don started by putting ICT in the UK in context. The UK’s spend on ICT in 2004 is expected to be £124.7 billion, with a growth rate of 3.2%. We will soon overtake Germany as the largest ICT market in Europe.


The three main drivers for this growth are IT upgrades, outsourcing and public sector spending. The latter comprises NHS IT system upgrades, broadband rollout and e-government. E-government is the delivery of local government services by electronic means, including phone, fax, email and the Web. The e-government target is to get all government services online by 2005. This was interesting, as Richard Allan had said earlier that providing services without carrying out a cost benefit analysis was not making good use of public funds!


Issues to be considered include business process re-engineering, project management, e-learning, security and the lack of e-skills. E-government applications include CRM, knowledge management, GIS and Web sites, portals, intranets and extranets.


The e-government budget is £220 million and has priorities for schools, transforming our local environment and promoting economic vitality. However, at present two thirds of the projects are behind schedule. Reasons for this include lack of funding, insufficient guidance and direction, poor infrastructure, skill shortages and cultural change.


Peter outlined some of the players involved. These include the Offices of the e-Envoy and Deputy Prime Minister, IDeA, LGA, SOLACE and SOCITM. Sources of funding include regional and national Government and the EU Structured Fund. The local public sector includes Business Links, Learning and Skills Councils and local authorities.


Suppliers must meet criteria including knowledge of public sector workings, understanding of SLP, VAT registration and professional indemnity cover.


Peter concluded with a 5 stage strategy that may help tap into these projects. These are: Contact the CEO; Offer free baseline consultancy; Target projects below £10K budget; Include weekly project management reports; Partner with other local suppliers.


Julie Howell, Digital Policy Development Officer for RNIB, began her talk by asking, ‘How does a blind person use a computer’? In fact there are few people who are totally blind. Visually impaired people can often see the screen, but may need to change the text size. Sometimes, smaller text is better, as for those with glaucoma. It’s important that the interface is flexible, so users can configure it.


Blind people use a ‘screen reader’ to translate html or text to audio. Relatively few (2%) read Braille, because most people lose their sight late in life. Braille displays do exist, giving tactile screen feedback. Partially sighted people can use a screen magnifier. There are also foot controllers and head pointers that help overcome disabilities.


There are WAI guidelines (www.w3.org/wai) for Web content accessibility. In law, there is the Disability Discrimination Act. The WAI guidelines are one component of a toolkit for alleviating disability discrimination. Others are user involvement and usability testing.


There are four groups whose impairments benefit from good Web design – those with sight loss, hearing loss, poor manual dexterity and cognitive impairments. Areas that need attention are image description and control of layouts, audio text transcripts, keyboard access and logical navigation respectively.


The government estimates that 8.6m UK residents have disabilities. But 9m are hard of hearing, 6m are dyslexic and 2m are blind or partially sighted. Another 4m fall into other categories such as stroke, epilepsy and industrial accidents. According to Forrester Research, 57% of work age population will benefit from ‘accessibility technology’ (see www.microsoft.com/enable/aging).


Julie concluded with an example of a case that benefited all parties. Tesco allows goods to be ordered via its Web site. In 1999 RNIB received 40 complaints from members who found the site unusable. Tesco visited RNIB and listened to suggestions. It spent £35,000 on R & D for a separate ‘Tesco Access’ site. In 2004 it forecasts £13m of new business. ‘Tesco Access’ has also been used in Korea and on the pocket PC, because it is easy to use. Julie says that many non-disabled people use it, purely because it is easier than the main site. Tesco now has plans to build its new main site to be easily accessible for all users.


Leave a Reply