The Economic, Social and Political Justification for Cybrarian

Date: 26/07/2002

0. Status.

0.1 This document is confidential to key staff in Cybrarian and to those to whom they distribute it.

0.2 It has been written by humanITy as an informal prospectus to assist detailed thinking on the way the Cybrarian Project should proceed but it is not an official document.

0.3 The statistical data and support for other statements has not been appended as this would take a considerable amount of time but humanITy is willing to answer any questions from accredited readers.

0.4 This document, written by humanITy is the intellectual property of HM Government but humanITy reserves the right to quote it subject to Government clearance.

1. Introduction

1.1 Cybrarian is the working title of an information system designed to attract to Internet use some 25 million people who, for a variety of reasons, are not current users.

1.2 The non-users fall into four broad classifications:-

The elderly *(of whom the disabled are largely but not exclusively a sub-set)
Ethnic minorities
Those classified by Government as "Socially Excluded" (see PAT15)
The 'untapped mainstream', not easy to define but encompassing people who are indifferent, alienated, lack incentive or do not know the potential of the system

1.3 Together the above constitute approximately 25 million people (* see Wild, Gabriel tba).

1.4 The Government figure is remarkably close to the demographic analysis of leading Internet charity humanITy which analyses non use of the Internet according to functional limitation based on four clusters of syndromes which, in descending order of magnitude are: cognitive/learning; physical; hearing; visual. Each of these clusters may most conveniently be viewed as a spectrum where the number of cases increases as the impairment becomes more mild so, for instance, there are only a small number of totally blind people but a very large number of people who find difficulty reading Times New Roman 10-point print on a PC screen (the default for many such screens). The four clusters account for approximately the following percentage of the total population:-

Cognitive/learning 20% - possibly low estimate given that 20% of the population are classified as being functionally illiterate in the analogue environment (OECD, 2000) on which basis the extra problems of the Internet such as navigation might raise the figure
Physical 14% - includes permanent serious conditions but also accidents and mild conditions which impair interface use such as arthritis, weak grip, shaking hands, clumsiness
Hearing 9% - usually divided between some 200,000 people with a severe hearing loss but eight million with a mild hearing loss; this may be a slight under-estimate
Visual 7% - approximately half a million with severe sight loss and two million with measurable sight loss but the figure is higher for Internet use because of problems with reading from screens.

These four clusters constitute approximately half of the population of which one fifth, or 10% of the global population, is generally regarded or actually Registered as "Disabled".

1.5 The humanITy sample would tend to a demographic of exclusion concentrated more heavily on the elderly than the Government figures. Nonetheless, in public policy terms the figures are so large and the discrepancies so small that a slight variance will make no real difference to the solutions proposed.

2. Government and the Citizen

2.1 The Government's target is to provide Internet access to all of those who wish to use it by 2005. This, however, can only be an intermediate objective because those who, for a variety of reasons, do not want to use it account for more than 80% of all public sector citizen transactions, largely made up of:-

Benefits claims
NHS use
Affirmative action programmes.

2.2 Government is also committed to reducing poverty and social exclusion, both of which will be severely increased by the 'digital divide'.

2.3 Government is committed to universal basic ICT skills. It is interesting that 20% of the school population is classified as having a Special Educational Need and that 20% of the population is classified as being functionally illiterate. Whether these are the same 20% at different stages of life is an interesting question for public policy. Whatever the answer, if Internet use excludes that very 20% of the population which most uses public sector services then proposed savings will not materialise because the eGovernment will be forced to operate dual systems.

2.4 The Government is committed to wider consultation and e-democracy, particularly where policy affects socially excluded people who need to speak for themselves. Margaret Moran M.P.'s consultation on domestic violence (House of Commons 1999) was a trailblazer in this respect.

2.5 For all these reasons, political, social and economic, the Government is committed to a policy of digital inclusion.

3. The Information Problem

3.1 Reference to Section 2. will show the bald figures for ICT exclusion but behind them there lies the complex question of information handling. Clearly, standard user interfaces (keyboard/mouse, screen) presents problems for people with certain physical disabilities and Cybrarian will make special provision for these; and there is also a clear problem for those for whom English is not their first language and, again, Cybrarian will take special measures; but a vast majority of those we wish to assist face a variety of information handling and processing problems.

3.2 20% of our population is classified by the OECD (“Literacy in the Information Age : Final Report of the International Adult Literacy Survey, OECD, 2000) as functionally illiterate in the analogue environment. Some of these will experience improved performance through ICT, such as dyslexic people who can take advantage of predictive word processing software and people with physical limitations can benefit from voice in systems. Cybrarian will provide the widest possible range of interface opportunities so that all such people benefit.

3.3 Many people, however, who can barely cope with standard literacy functions find digital systems bewildering or alienating; here are some of the major factors:-

3.4 Economic
3.5 Incentive
3.6 Alienating

3.4 Economic. Although the cost of all the components of Internet access - computers, software, telephony - access are constantly falling in real terms, it is still significant when put up against other priorities unless the incentive for use is strong. This area is not part of the Cybrarian remit but is included for completeness.

3.5 Incentive. It is remarkable that the Internet has flourished as it has when there has never been a strategic campaign to increase its use. It may well be of benefit to governments wishing to transfer information to citizens and it may be of use to retailers who want to sell goods but nobody has made a clear, general case why it is good for citizens. Clearly, in an unregulated environment where the abstruse and the pornographic can be presented alongside banal and useful information the incentive for use is somewhat muddled. Added to this, continued talk of convergence between the Internet and television reduces the incentive of many people to concern themselves with the first when they are familiar with the second. Cybrarian will need a clear marketing strategy.

3.6 Alienating. The PC accessed Internet is alienating for a wide variety of reasons but here are some of the more egregious; it is:

Unreliable to an extent which would not be acceptable for broadcasting
Technically complex to the extent that almost all people, even buying a PC/Windows 'bundle' cannot get started autonomously
Messy because academic papers, pornography, hysteria and official information can all be presented as items in a report of a search
Keyboard/mouse based which is non intuitive and often unusable
Unaccessible because the interface is difficult for many people (keyboard problems, as above) but also because navigating in disorganised 3-dimensional hyperspace is very difficult.

3.7 Conversely, people find email and its attachments, and SMS very useful and easy. The computer faces the classic problem of the VCR but in that case the incentive and end product were easy to understand.

3.8 There is a classic rule which says that people faced with choice optimally perform with 7 + or - 2 elements. The array of information on the Internet is massive. Even using an electronic programme guide (EPG) on the digital TV system can be daunting but at least it is genre, channel and time defined which means that people have three different ways of finding what they want. They even have some hope of finding something to entertain them when they do not know what they want but that does not apply to the Internet where the wider the search the longer the report and vice versa.

3.9 Academic research also increasingly demonstrates that people in general but adults in particular and female adults above all, are not prepared to learn 'just in case' but only wish to learn 'just in time' to achieve a measurable output. This is why people alienated from the formal learning system will not readily enrol on literacy classes in places called "Learning Centres". They will only learn osmotically (as they do from television), interactively (as they do from peers) and for a purpose (as they do to solve a problem at work or at home). This is an area where Cybrarian will have to be particularly adept.

3.10 The Government is committed, as noted above, to universal basic literacy, numeracy and ICT skills as the basis for economic self-determination, maximum self-determination and good citizenship. While this objective may achieve the last two of these three social goods it is unlikely to continue for much longer to achieve the first. The application of basic literacy and numeracy skills for economic purposes depends to a very large degree on the division of labour, the atomisation of production procedures and the performance of repetitive tasks. The problem for employment is that the more these three techniques are applied the easier it is for a computer to replace a human being. Language and metadata are already being created to allow much more powerful machine transaction processing, known as the 'Semantic Web' or Resource Description Framework (RDF) which are being built on to the increasing power of HTML specification and the development of metadata standards like the Dublin Core. In the medium term human beings must concentrate on doing those things which a machine does less well or not at all such as lateral thinking, content collation, content creation, advocacy, interpretation, assessment and matching with human sectoral or individual requirements. It is therefore important that Cybrarian pushes people towards these aspects of information handling and processing.

4. Cybrarian General Specification

On the basis of the above analysis, the Cybrarian project will have the following core characteristics:

4.1 Multiple hardware and software user interfaces
4.2 User profiling through configuration and system auto-adapt
4.3 Intelligent agency
4.4 Language engineering
4.5 Walled garden information base
4.6 Learning and creative strategy prompts and tools
4.7 Non-invasive automated individual performance metrics.

4.1 Multiple hardware and software user interfaces. This is the most obvious special requirement of the Cybrarian system which will include such software features as voice in/voice out, compliance with the Web Accessibility (WAI) Web Content Accessibility Guidelines (WCAG). Over time it should also provide a variety of hardware interfaces to run alongside the PC such as digital television and 3g phones.

4.2 User profiling through configuration and system auto-adapt. We are familiar with the customisation of information systems through, for example, altering the font and size of print, foreground and background colours etc. It is already possible for a system to perform simple tasks such as ranking in-coming email according to the user's behaviour. It is therefore important that Cybrarian allows people to maximise accessibility through initial configuration but it is also important that over time the system 'learns' about each user. A voice in programme, for example, should 'learn' from the user's pronunciation, a screen set-up should default to a user's standard characteristics but if the user changes these it should 'learn' to default to the changes.

4.3 Intelligent agency. As has been noted, intelligent agency in simple tasks is already with us. The system should, for example, know the kinds of information individuals are interested in, the reading age at which they want it and some 'understanding' of the kind of help they will need.

4.4 Language engineering. This is a difficult area because the tools for it are somewhat scattered: there are Spelling Checkers and Thesaurus functions on most standard machines; simplifying and parsing technologies are well developed; data sorting and interpreting systems like Autonomy can also be helpful. While this kind of language engineering within a language is somewhat messy, language engineering between languages is now widespread.

4.5 Walled garden information base. This is the central feature of the system. Cybrarian will collect existing content and re-purpose it for the special search engine *(*companion). This will allow all groups to work safely and confidently inside a protected and non-controversial environment. The project will conform to backwards traceability standards so that any re-purposed material can be traced to its initial source and format.

4.6 Learning and creative strategy prompts and tools. As noted in 3.10 above, it is important that the system provides a wealth of tools for data collection and processing, together with prompts for more creative or adventurous approaches, stimulating but fair tests and suggestions for next steps and for corrective/remedial action.

4.7 Non-invasive automated individual performance metrics. These should be invoked on a voluntary basis when an individual is being registered and her access customised. Tools might include, for example, spelling checker monitoring, numeracy monitoring, lexicographic range (reading age) assessment, breadth and depth of searches conducted.

The following are major areas which require further discussion:

4.9 Nature of content
4.10 Level of interactivity
4.11 Hybridity

4.9. Nature of Content. Although this has been broadly agreed but rather by the negative process of defining what is not wanted. We need, however, to be very clear that the whole atmosphere has to be non didactic and not 'heavy' in any way.

4.10 Level of interactivity. This may range from zero (library shelf model with digital assistant searching the stacks and carrying data to client), the provision of tools and prompts for choice (the equivalent of the digital TV choice of games and camera angles), email traffic bilaterally or through e-groups, chat rooms, human on-line mentoring.

4.11 Hybridity. This defines the place at which there is, if any, interface between the digital system and human agency.

5. Economics of the Cybrarian Case

The following are notes on economic headings:

5.1 Benefits
5.2 Health Care
5.3 Literacy/numeracy and Employability - basic skills
5.4 Information Handling and Processing
5.5 Shunting
5.6 Communities of Interest
5.7 Social Capital
5.8 Citizenship
5.9 Mobility and Transport
5.10 Self esteem

5.1 Benefits. Given the target groups defined in 1.2 and 1.4 above, whichever set of demographics are used these groups together constitute almost all people collecting benefits (as opposed to receiving tax credits). If Government is to pay benefits on-line (either through the digital television or, much more likely, through a personal mobile phone), benefit recipients need a basic level of ICT literacy above that required for using a remote controller or telephone dial. If major savings are to be achieved it also means that the degree of human agency will have to be decreased to be replaced by automated systems. Much of this is not going to be achieved via didactic numeracy and literacy instruction and so Cybrarian will have to develop innovative ways of transferring these skills to meet a major public policy objective.

5.2 Health Care. The elderly account for the overwhelming use of the NHS. A good deal of what people need such as diagnosis and treatment of minor ailments, providing comfort and counselling, information and mutual support, can be provided on-line. Cybrarian will provide a safe environment for exploring alternative strategies such as the use of image maps (displaying a body with named components for 'clicking'), the use of keywords for searching, general searches for descriptions that vary in length and technical complexity. NHS on-line services should not be clogged up with learners. These should be referred to Cybrarian for practice sessions. This will make on line services more efficient but, over time, it will increase the use of on line services by the general population, particularly the disabled and the elderly, and it might reduce the number of human operators providing basic advice, diagnosis and prescription.

5.3 Literacy/numeracy and Employability - basic skills. Basic literacy, numeracy and ICT skills are necessary for effective learning, realising life potential and participative citizenship. They have traditionally been associated with employability but it has already been noted that many jobs requiring basic numeracy and literacy only will soon disappear. Nonetheless, jobs where such skills are not intrinsic to the basic task still have health and safety, compliance and payments requirements which involve literacy and numeracy and increasingly people will be required to have a modicum of ICT skill. It is not easy to see how Cybrarian can demonstrate a direct link between its aspirations and increased economic effectiveness but, conversely, it makes people better informed, more likely to read and heed warnings, more likely to do simple things effectively, it will at least decrease inefficiency, accidents and waste.

5.4 Information Handling and Processing. Information handling and processing is at the core of the Cybrarian ethic. The aspiration is to improve everybody's effectiveness at finding, understanding, processing and building on reliable information. These skills will be central in this new century as their utilisation by humans will continue for some decades to outstrip machine capacity. They are also essential skills for happiness and citizenship. The economic advantage of an information literate society is obvious and the economic disadvantage of a society not comfortable with information handling and processing is equally obvious. This is a somewhat heterodox analysis but it is not meant to advocate an alternative to basic literacy skills but, rather, to alter the emphasis. Our culture predominantly relies upon the autonomous achievement of a massive variety of tasks (exclusive literacy) but the information age relies upon the collaborative performance of a variety of tasks in a multimedia environment where the network is real and not a metaphor. Almost all employment will involve gathering, processing building on and distributing information. Cybrarian can provide the first rung on the ladder for millions of people which will therefore increase their employability in the information age.

5.5 Shunting. A note is required here about the idea of 'shunting', the process by which each level of skills attainment is built on incrementally so that people are gradually 'shunted' up the learning scale. This is a vital iterative process in the information age and Cybrarian should be seen as the engine which gets people off the ground to the first level. The direct economic return on moving from ground to level one may be small but unless people can reach level one they will not be able to be shunted up to level two and beyond. In an era of ICT skills shortage this strategy is vital and is of central economic performance.

5.6 Communities of Interest. Whereas people are often forced by economic and social circumstances to live in geographical communities which they do not like or which discourage initiative, the Internet in general and Cybrarian in particular offer the opportunity for 'virtual' communities, primarily communities of interest. It is easy to dismiss such communities based on soccer teams and other sports but even these have interesting potential for improving numeracy, statistical literacy and information handling. However, the clear advantage of these communities is that they allow people to operate productively in spite of their physical environment. In many cases this will be liberating, in some it will protect people from intrusive and negative peer pressure and in turn it will reduce social instability, substance abuse and crime. On the other hand, communities of interest raise standards, productivity, self-confidence, ownership and awareness. Cybrarian should fit neatly at right angles to geographically based initiatives.

5.7 Social Capital. The two forces of social and physical mobility for the majority and immobility for the minority has seriously eroded social capital where it is most needed, in geographically deprived areas. Building on the idea of communities of interest, Cybrarian can, through its extensive stakeholder network, facilitate the development of social capital across geographical boundaries. Instead of the least well off and least gifted being asked to help each other, Cybrarian will bridge the gap between those who need support and those who can learn and provide support simultaneously. This aspiration depends upon defining a level of interactivity.

5.8 Citizenship. Citizenship and e-democracy are too often used to mean on-line voting only. The Margaret Moran consultation on domestic violence shows what can be done in the area of democracy through the use of the Internet. Very often the people in our target groups are described as the problem instead of being described as people with problems. Cybrarian can help citizenship through putting stakeholders in touch with those who work with them and represent their special interests. This, in turn, will lead to a more efficient formulation of and expenditure on special programmes to deal with segments of our user base. Citizenship also involves mutuality and the Cybrarian experience will provide people with another means of giving as well as taking.

5.9 Mobility and Transport. The Internet can be a virtual route to learning and employment which spares housebound people pain, time and money and which might provide them with a means of entertainment, education and employment. It might also allow the public and NGO sectors to provide more on line help more regularly replacing expensive and patchy face-to-face services. As long as this cost saving does not leave people prisoners in their own homes it should be mutually beneficial, to service providers and consumers.

5.10 Self Esteem. The lack of self-esteem is at the root of most of our social and economic problems. The ability to achieve more than one expects of oneself is central to progress. Cybrarian will allow people to fail privately but succeed publicly, to struggle alone but achieve in front of peers. The project will institute some automated metrics but these might be included on a random sample basis:

Do you know more/less people than at the beginning of Cybrarian?
Have you made more friends or lost friends
Do you trust people more or less
Have you helped more people
Do you like yourself more or less?

Return to Economics