The WBU ‘PURE VIP Book Project’ Interim Report

Return to Blindness and Visual Impairment

Platform neutral (high level) User Requirements for Electronic access by Visually Impaired People to (carrier free multi-media) Books Report commissioned by the DAISY Board

Date: 19/10/2007

Executive Summary and Consultation Framework

The WBU in association with the DAISY Board are anxious to ascertain high level user requirements for the 'net generation' of talking books. A Preliminary Report was submitted to the WBU's Technical Committee on 2^nd October 2007 and submitted to the DAISY Board on 9^th October 2007 and both bodies agreed that the main findings of the Preliminary Report should be carried forward into this Interim Report and submitted to all interested parties for consultation.

The following is a summary of Recommendations with a Consultation Framework but we welcome any additional comments not covered by this Report:

The sector as a whole finds it difficult to assemble representative samples of users to inform design decisions (Recommendations 1 and 2):
1. Do you wish the WBU/DAISY to define a standard for representative samples? Or
2. Do you require assistance in defining your own representative sample?
3. Do you wish to establish a 'pooled' sample with neighbouring countries?
4. Is your organisation prepared to commit itself to establishing and utilising a representative sample of users by the end of 2010?
Often the demographic of accessibility is sub divided according to age but it would be more effective to sub divide it according to task (Recommendation 3):
1. Do you agree with the above statement?
There is a variety of views on whether talking books should be accessed on commercial equipment, special players, linear broadcasting or on any/all of these (Recommendations 4, 6, 8, 11, 12, 15):
1. Which of these options do you consider suitable for your country:
  - Special players
  - Commercial players (PDA/MP3 etc)
  - Linear broadcasting
  - Subscription television?
2. Do you have any data on the use by blind and visually impaired people, particularly the elderly, on the autonomous use of consumer electronics products such as televisions, radios and CD players?
3. What is the maximum retail price you would pay for a special DAISY player?
4. What additional commercial features would you want on a special DAISY Player:
  - CD player
  - PDA/MPC
  - Radio
Many people say they either want simple controls on a special player or the option, effectively, to 'programme' the player so that more complex controls can be 'switched off' (Recommendations 5, 6, 11); do you favour:
1. a) A simple player
2. b) A defaulted player with the option to switch on more complex controls?
Many organisations expressed doubts over the issue of copyright in respect both of using commercial players for talking books and producing 'parallel'; audio and XML files (Recommendation 7, 12):
1. Is the obstacle to using PDA/MP3 commercial technology for DAISY books the result of a user requirement, copyright concerns or both?
2. Is the obstacle to producing 'parallel'[ audio and XML files on the same disc or in the same delivery package the result of technical or connected with copyright or both?

We would appreciate it if your replies to these questions could reach us by the end of January 2008. If you would like further clarification or if you would like to arrange a teleconference or a visit to your country, please let us know as soon as possible at: humanity@atlas.co.uk

1. Introduction

This Interim Report was commissioned by the DAISY Board following its consideration of humanITy's Preliminary Report at its meeting on 8-9th October 2007. This Interim Report is substantially the same as the Preliminary Report but it contains a consultative framework which interested parties are asked to complete by the end of January 2008.

For the Preliminary Report all DAISY Members were invited to hold seminars and submit data. Data was collected from Africa, Australia, Canada, the DAISY For All Project, India, Japan, The Netherlands, New Zealand, South Korea, Spain, the United Kingdom (Appendices 1-11) and the accessibility products sector (Appendix 12). humanITy included its own unranked preferences broadly based on all the other data (Appendix 13). The Spanish data broadly supports the findings of the other Appendices but could not be coded into a tabular format.

2. The Context

In 1996 the European Blind Union published Reaching for the 21st Century: User Requirements for the next generation of Talking Books. This was part of a development which ultimately led to the formation of the DAISY Consortium in 1997 whose aim was to develop standards and technologies for talking book services. 11 years on from the foundation of the Daisy Consortium, many countries have implemented DAISY talking book and text based services based on distribution by CD-Rom. A number of organisations are now beginning to contemplate delivering services online, via internet, mobile phone or radio networks. The Consortium is thus challenged to develop standards and technologies in this area.

Given that it is over 11 years since the original user requirements for digital talking books were developed and the digital world has changed the knowledge and expectations of what is feasible, the DAISY Consortium asked the WBU to conduct a review of user requirements which might inform future developments.

One issue with the original EBU Report was a lack of consideration of the needs of developing countries. The WBU was asked specifically to consider this in its report.

This Report is based on the results of a series of meetings with experts and seminars with users in a selected range of countries. It does not attempt to be a thorough piece of quantitative research but is rather a first review to collect issues which may require further research, and to promote debate.

Now that the Daisy Board has considered the recommendations in the Preliminary Report it requests Daisy member organisations to stage consultations in their own countries, submit comments and suggest improvements. By this method we hope to identify issues that need consideration as standards and technologies are developed and organisations start to plan implementation.

The Report is not intended to provide a blueprint for development of services, it is more a list of issues for consideration.

The issues are separated into:

Section 3 - Those pertaining to the Consortium and the way it works
Section 4 - Those which pertain to organisations delivering services
Section 5 - Those which pertain to the reading systems of the future.

3. DAISY Consortium Issues

Perhaps the major finding of this Report is that the sector is poorly equipped to elicit and interpret the needs of users both structurally and professionally. From the structural point of view there are few instances of attempts to assemble representative samples of users to provide brainstorming, participation in design, feedback on prototypes and assistance with pre-production modification. The theories of user-centred design and design for all are well known and frequently cited but their practical realisation and implementation is poor. At the professional level, few of those in R&D functions have Human Computer Interaction (HCI) and related experience and there are two serious tendencies in the sector which arise from this deficit:

First, people readily develop requirements which they think users ought to want or need
Secondly, there is a tendency to argue in adversarial either/or dichotomies rather than viewing a range of options in the context of both/and. Digital information, being plastic, is particularly suited to a wide variety of production methods, transmission, reception and processing.

On a number of occasions humanITy was told that a certain group of users had a particularly strong requirement; but the evidence was scant and sometimes ran contrary both to my limited findings and my professional experience.

Recommendation 1. The sector needs to pay much more attention to assembling representative samples of blind and visually impaired users to be involved in all stages of technological development from brainstorming to regular pre and post production evaluation.

This Report suffers from the shortcomings of the sector it seeks to analyse. Although there was a widespread and repeated request to organisations to assemble representative samples of users to provide feedback on their previous experiences and advice on the future, only four countries out of the ten provided user groups of any kind and only one user group was a representative sample of its peers. In mixed groups of users and policy-makers and/or technical experts, the latter were usually dominant and often needed to be checked. In one instance users gave completely different answers when left to themselves.

The data in this Report can therefore not be said to be scientific on a quantitative basis and the ranking in the data tables is almost always approximate but the credibility of what has been gathered has been saved by broad agreement on the main issues.

Recommendation 2. In view of the difficulty in assembling national samples the DAISY Consortium may need to assemble a representative sample of users across a range of tasks.

The second major issue is building up a rational case for access and/or mainstream technology. Not unnaturally, the access technology commercial sector largely thinks that there is a continuing need for specialist hardware and software, particularly for primary and secondary school children; but even the industry questions that need for students and adults.

There is overwhelming agreement that audio books should be so produced that they can be read on standard consumer electronics products. There is an important debate to be had about the special needs of users who require more than basic functionality on a commercial player. This has traditionally been discussed in terms of groups of users such as students, employees and the elderly and this Report uses these categories in discussion but in general humanITy believes that this is the wrong way of addressing the problem and therefore:

Recommendation 3. Any user requirements for the sector should be rationally sub-divided according to the task not the user demographic; and that where these can be met by commercial user interfaces without or with minor adjustments to the content and its metadata, this should be implemented.

Closely allied with this:

Recommendation 4. Where specialist user interfaces are shown to be required to complete specific tasks, such players should be capable of receiving, processing and playing commercial products such as CDs and DVDs, MP3 files and linear broadcasting (particularly radio).

Following on from Recommendation 3, it is widely held that students and people in work need much more sophisticated navigation systems to assist in their reading of books. This may well be true for a number of students and employees who are working at higher levels of education and skill but in the context of school children, students and people at work the peer group task framework is much more important than the demographic. In many countries the percentage of blind and visually impaired children with additional disabilities is rising. Conversely, policy making in most organisations of and for blind people relies heavily on highly gifted, usually congenitally blind people whose technical facility is remarkable even by peer comparison; therefore:

Recommendation 5. There should be detailed research into the particular navigation needs of those in formal education.

One natural conclusion from these recommendations is that different people need different functionality and, therefore:

Recommendation 6. Where needs cannot be met through the strategy in Recommendation 3, any specialist player should be capable of variable functional complexity so that, in effect, the functional default should be simple with additional functionality masked and turned on.

There was near unanimity among users, policy makers and administrators that the ideal delivery format for text is one which provides the capacity for producing synthetic speech, modified print and braille alongside the audio file. This would be particularly valued by students and those using screen magnification devices that could be linked to the book delivery system. There are also many people with learning difficulties (many of whom are classified as print disabled) who say they would benefit from the simultaneous delivery of audio and visual text. This requirement also meets the needs of developing countries, some of whom say they would prefer synthetic speech to nothing at all and others which say they would prefer synthetic speech to real audio if this meant reducing costs and increasing the quantity of text and the speed of delivery. The developing country proposal for synthetic speech was primarily driven by users and not policy makers.

Recommendation 7. The standard format for any DAISY text product should be an audio file, together with files for speech output, adjustable print and braille.

This requirement for simultaneous delivery still described by many policy makers and users as a ‘Holy Grail', implying that this is still beyond reach. The key question is why it is not current standard practice (see Section 4, Recommendation 14).

A number of these recommendations are somewhat unanchored as long as a basic issue continues to be unresolved:

Is the requirement for a special audio book player the result of user need or publisher pressure for a 'closed system'?

In putting this question humanITy never received a satisfactory answer, i.e. one where the two elements were separated and ranked. This issue is critical because massive ethical and financial issues arise if the need of the publishers is actually ranked higher than the needs of the users. humanITy was frequently told was that there was that a majority of blind and visually impaired people, at the upper end of the demographic age range, who need a special player with simple controls because they cannot manage commercial equipment. No evidence was offered for this assertion and almost 100% of the old people interviewed said they can autonomously operate their televisions and CD players. The sample was not big enough to be scientific but:

Recommendation 8. Research should be conducted into the autonomous use by blind and visually impaired people of consumer electronics and lessons drawn from this in respect of special player design and/or the need for a special player.

Recommendation 9. The issues of user requirement and copyright protection must be radically separated.

Recommendation 10. The campaign to resolve outstanding issues in alternative format production copyright permission should simultaneously: pursue the current diplomatic effort; underpin this with deadlines; and, in the long-term, aim for a generic right for information in the public domain.

Recommendation 11. Any specialist player investment must be supported by evidence of user need such that the opportunity cost of the investment is justified compared with alternative strategies or priorities.

These recommendations lead to a consideration of the third issue, the sector's response to changing circumstances.

The time lag between generic technological development and accessibility response is unacceptably long. In the late 1990' and early 2000s the response to mobile phones, MP3 players and digital television has been slow. Ironically, reaction to Satellite navigation systems has been swift but the evolution of the technology to provide instructions for pedestrians has largely overtaken the accessibility sector. In the area of audio books there was strong support for standard MP3 players but very little understanding of the potential of subscription television, owned by many older, visually impaired people and equipped with PIN protection.

The key question is not which technology should be used to deliver talking books but how many technologies can be harnessed to suit the particular needs of service providers and users.

Recommendation 12. Viable commercial technologies should be considered before investment in specialist access devices.

The problem that arises with this strategy is that it will inevitably reduce the numbers using specialist players and increase the cost per unit. That cost could partially be off-set if such specialist units were multi-functional. While most people agreed that any player should be multi functional to some degree, the actual bundles of services it might handle are a matter of contention. Most people agree that players should carry mainstream products like DVD, CD and radio but are divided about newspapers and telephony. There was some support for special players to contain a recording facility. It is difficult to know where the trade-off is in this area because many of those who might need a specialist player for its simplicity would, consequently, not want a complex, multi-functional player.

4. Issues for organisations delivering services

The structure of most organisations of and for the blind means that the text reading part of their organisations are naturally dominated by librarians and educators. This means that there is a default focus on books. Only when prompted did users think about other content, particularly newspapers which require much more refined searching than DAISY books. The organisational infrastructure of services for blind and visually impaired people should not over-ride user content requirements:

Recommendation 13. Governmental and non governmental organisations providing audio/text delivery services for blind and visually impaired people should ensure that services reflect user content requirements rather than organisational remits.

As noted in Section 3, Recommendation 7, there is overwhelming support for the delivery of audio files and text files which can deliver synthetic speech, modified print and braille. A number of organisations have been working on this problem since the last years of the 20^th century but there appear to be a number of outstanding problems:

Recommendation 14. National audio and text delivering organisations should collaborate to resolve outstanding difficulties over the production of a single bundled format for delivering parallel audio and text files.

At root, of course, one of the critical areas in delivering material in alternative format on a special player is the price. This is particularly true in developing countries but even in some richer countries it is identified as an issue; but in general the richer the country and the greater the subsidy to users from organisations of and for blind people, the lower price features in priority lists. The less price sensitive the users the less pressure there is for DAISY books to be accessible on cheap commercial players.

Policy makers (who might have to pay) and users agree a top limit of $400 for any specialist player but agree that this would have to be very specialist; they see no justification for paying $400 for an elderly light fiction reader.

There are now viable options to the use of a player with revolving data and moving parts. We have already referred to solid state players but there is a strong preference, on the basis of price, for talking books to be delivered in developing countries through linear broadcasting. This might also be an option for many users in developed countries who prefer a radio to a library model. Many people want a book for company but are not that concerned about which book.

Recommendation 15. Linear broadcasting of books should be considered in developing countries and research should be undertaken in developed countries into user requirements for a library or subscription, as opposed to a linear broadcasting, model.

5. Issues pertaining to designers

There was surprising agreement on basic functionality:

Off/on volume start/stop and book mark

were unanimous, this last eliciting by far the most comments from users of all ages. People simply wanted to re-start reading where they had stopped, whether the stop was voluntary or involuntary.

There was more division over navigation:

Most older people, given a book mark, were content
Some older people and most students wanted navigation by chapter and page
One technophile wanted to move forward and back by using time intervals.

Interestingly, one device worked at right-angles to the controls of standard mobile phones, using a book shelf metaphor which meant that top categories went from left to right. Most people want their player to look and feel like a standard device.

Initial discussions on navigation and searching revolved around books but the newspaper raise navigation questions which led to a speech recognition requirement or at least SMS word searching to complement qwerty word searching:

Recommendation 16. Alternative search mechanisms should be available including voice in, SMS and qwerty.

There is an apparent conflict between the requirement for portability on the one hand and large buttons on the other.

One solution which neither users nor policy makers considered is combining a standard processor system with cable free communication to a special user interface. The widespread use of remote controllers means this is no longer a strange idea. A special remote controller could be produced for a fraction of the price of a special player.
A second alternative is to develop voice in capacity, rendering most controls unnecessary. Most users did not spontaneously propose this solution when being asked about general features but the need for word searching on machines without qwerty or numeric keypads brought this need into focus.

Recommendation 17. The issue of the user interface, particularly for older people, should be considered separately from the production of a complete player system.

Other issues:

Portability. All but the very old were anxious to be able to read books anywhere, any time. This frequently led to discussions by policy makers and IT experts of streaming and downloading, of server and client side options but the general user perspective is that they do not mind how the technology works as long as the operations are simple and the results reliable. Clearly, however, there are some important technical factors to be considered: streaming might be viable in some domestic environments but for some time to come it will hardly be viable on the move.

Recommendation 18. Special players should be capable of battery operation and be portable.

Battery Life

Recommendation 19. The battery life of a special player should be at least one working day (8 hours).

Appearance

Most users want their equipment to look like ordinary commercial products:

Recommendation 20. Special players should, as far as possible, look like commercial products.