Adults and students who are blind or have other print disabilities depend on their national libraries serving the blind for leisure reading materials, school books, newspapers, journals, magazines, and employment related information. Currently analogue cassettes, braille, large print, and computerised books on floppy disk (E-Text) constitute the types of accessible information provided to the clients of libraries for the blind. `Talking Books’, the common name for an Analogue audio recording, provide by far the most people with essential reading materials for intellectual stimulation than any other format.

Most talking books are still recorded on cassettes, which presently provide the most accessible reading medium for visually impaired people. However, it is difficult given today's implementation to look up information in a talking book or even to simply turn pages. Thus the books most commonly enjoyed by sighted people throughout the world – cookery books, gardening books and religious texts – are difficult for visually impaired people to use in ways considered essential to their sighted colleagues. Books used in school or on the job require contents pages, indices and other `structure’ for fast and efficient reading. The next generation of digital talking books will provide this functionality!

When audio books referred just to those recorded onto magnetic tape, the very nature of the medium prevented anything more than a very simple structuring based on tone indexing. The DAISY concept does not have that degree of limitation and we now have to recognise the need to support a greater complexity of forms.

The need to digitise audio collections around the world is clear. Currently, each country has its own system and format for serving its clients. Some countries use the common two-track cassette. Other libraries use a four-track system and still others use a six-track format. The lack of standards severely limits inter-library co-operation. All of the libraries produce books for their blind or print disabled clients and many see the end of the analogue cassette on the 10-year horizon. With the rise of the compact disc and now the mini disc, the traditional analogue cassette will become less attractive as a playback medium. Already the music industry has discontinued music on compact cassette.

The digitisation of books intended for persons with disabilities provides opportunities to increase the quality and availability of information to print disabled persons. Visually impaired people have the same needs as everybody else: access to all types of information. Information should not be limited to the language of origin or the language group in which visually impaired people live. The visually impaired community is becoming increasingly multi-ethnic and multi-cultural. As culture, education and information become increasingly global in nature, libraries for the blind have been cooperating internationally to develop the next generation of `Digital Talking Books’ (DTB's) and then to go on and exchange the books, which are so desperately needed by the people they serve. Major libraries serving the blind or print disabled have agreed to work together to establish the International Standard for the next generation of information which includes DTB and braille.

The comprehensive system for the next generation of information to be delivered to blind or print disabled persons is called `Digital Audio-based Information SYstem’ (DAISY), and the organisations working together are called the DAISY Consortium.

The background vision of DAISY is to provide people with the ability to navigate and use audio information to a similar level to that enjoyed by sighted readers of a printed book. The consortium has recognised that this can be best achieved by using a combination of international standards to describe the text (often in a summarised form), the audio and the synchronisation between the two. The standards themselves are essential in that, copyright permitting, they will enable information to be inter-changed – an electronic inter-library loan.

At the heart of the whole DAISY project is the book, that most versatile of information sources which has been developed over many centuries. As far as our libraries are concerned, there is one great drawback to the book it is designed for a sighted world. As we consider how to translate the printed book into the audio domain, it is essential to reconsider what the book contains. The bulk of the information is in the actual textual content which can be interpreted either by a human narrator or by a speech synthesiser. However, the ability of the user to move around and through the book depends on the level and style of structuring. An added complication is the inclusion of non-textual information elements such as diagrams and pictures.

Most books have at least a minimum level of structure, even if it is simply a division into chapters. Where further division is required, this is often into sections and sub-sections. Such a book will also tend to have two other structural elements, a table of contents (TOC) and an index, from which cross references can be made into the main contents.

The users of audio books need to be able to move readily between sections without having to go through the intervening content. In a hierarchical arrangement, the user must be able to move within a defined level of the hierarchy or between such levels. Where there is a large amount of content between intervening headers then there is also a need to have additional anchor points, preferably at boundaries such as start of paragraphs.

To use the TOC or the index, the user must be able to carry out a search, possibly item by item, and make a selection by following a cross reference into the main content.

Although the hierarchical structure is the prime arrangement for many classes of book, including, in its simplest form, the novel, many books also include a number of other more complex structures. Some classes of book, e.g. an encyclopaedia, a recipe book or a directory, are fundamentally designed around more complex structures in keeping with the type of information which they contain. The main types of structure are:

1. Those in which the user needs to switch regularly between two related parts of parallel content, such as an ingredients list and a method within a recipe.

2. Those in which there is some ‘floating’ information which is related to main content but does not fit into the sequential flow and may not even be referred to from elsewhere. Typical examples are margin notes or diagram descriptions.

3. Cross references, either to another point in the main content or to a footnote.

For the audio user these structures pose two significant challenges; firstly knowing that the structure exists and secondly navigating within it.

Take almost any book and it becomes clear that it can be divided into a number of specific parts, each with their own defined structures. Even in the simple book described above there is a TOC, the main content and perhaps an index, each of which have their individual structures. Even the bibliographic information, preface, book jacket blurb etc. can be defined with as unique elements within a book.

Other books, such as gardening books, may have their content divided into entirely different sections. The initial part is often narrative, possibly looking at different types and designs of gardens. The next section may be a detailed description of growing, harvesting and propagating a wide range of plants. The last section could be a plant finder, basically a detailed glossary or cross reference of plant names. The book may contain two or more indexes, arranged against different criteria. Whereas narrated audio would probably be preferred for the first two sections, it may be better to use full text and synthesised audio for the plant finder and indexes due to the large number of botanical names.

Accepting these difficulties, the DAISY Consortium has planned and prioritised a list of strategic activities and identified some financial resources to conduct these activities. The cost of moving to the new technology is substantial and the organisations recognize the importance of sharing the costs of developing the technology. These activities, plans, and their financial implications are described in outline in this document.

To meet the vital needs, a number of major talking book, E-Text, and braille producers, throughout the world formed the DAISY Consortium in May 1996 to establish a world standard for compatibility among the next generation of information systems. The Consortium, which consists of non-profit organisations, intends to develop an open standard based on existing international standards, which can be used by any manufacturer. With an open standard, manufacturers of supporting hardware and software will find one integrated, large market, which will benefit both end users and service providers of talking books and braille. The DAISY Consortium was established as a legal entity to receive and distribute funds managed by the Consortium. The Board of Management (BOM) of the Consortium are elected by the Consortium full member organisations.

The DAISY Consortium will use existing International Standards wherever possible. The textual and structural information will conform to the standards defined by the `World Wide Web Consortium (W3C) (http://www.w3.org). The proposed file specification is an application of the SGML standard, ISO 8879. The DAISY Consortium will participate with working groups of the W3C to develop the appropriate standards, and to promote accessible information on the Internet and through libraries for the blind or print disabled. The storage of sound files will support existing industry standards. A global perspective for language differences will be adopted, ensuring that all languages will be supported in the standard.

The DAISY Consortium will contract with and encourage for-profit and non-profit companies to develop software needed to support the DTB and braille industry. Some of the software needed will include DTB authoring tools, analogue to digital conversion utilities, playback software, compatible web browsers, software synthesizers for many languages, encryption utilities, production management systems, etc. Some specific examples are:

- Recording workstations

- Analogue-to-Digital conversion tools

- Internet and DTB student browsers and playback systems

The DAISY Consortium will encourage the development of hardware devices that are needed in the DTB and braille industry. In general some of these devices are specialized keypads for production, hand held playback devices, reproduction systems, web servers for whole collections, dedicated portable recording units, etc. Some specific examples are:

- Simple hand held devices to meet the leisure reading client's needs

- A sophisticated hand held device for student and professional use

- A hand held device that includes synthetic speech in multiple languages

- A hand held device for connecting to the Internet and for reading DTB materials locally.

- A recording keypad suitable for DTB production.

We also expect that standard multimedia computers will be employed by students and professionals for reading materials using more sophisticated techniques.

Readers need to be able to use books in essentially the same way no matter what country or library produced the book. To ensure this level of consistency, production guidelines need to be standardised. This differs from standards for file specifications and other technically related issues. Instead, the Production

Guidelines are a comprehensive collection of recommendations and examples of how to produce various classes of DTB. These guidelines govern the application of the file specification standards in real book production environments.

The first draft of these guidelines is under review.

In the global environment, books, journals, and magazines produced in one country will often be needed in other countries. The DAISY Consortium will establish mechanisms for the exchange of materials between libraries throughout the Consortium. This will involve the free inter-library exchange and the sale of these materials. The protocols for these exchanges and the accompanying financial issues will be developed.

The DAISY Consortium recognizes the intellectual property rights of authors and publishers. The current international laws do not make it easy for libraries for the blind or print disabled to produce and exchange books, journals, and magazines on an international basis. The DAISY Consortium will seek to collaborate with other organisations interested in copyright issues to bring about improvement in International Copyright Law to benefit persons who are blind or print disabled.

The DAISY Consortium recognizes that DTB materials, like most digital materials, can be copied and could be used illegally. To protect DAISY materials from illegal use we will work with the companies developing secure technologies for business and commerce. We will implement accepted security measures where needed.

The DAISY consortium will promote the use of the DAISY Standard throughout the world to libraries and schools serving persons who are blind or print disabled.

The DAISY Consortium will recruit full and associate member organisations. There is no expected limit to the number of member organisations that may join. Participation from companies developing hardware and software or companies interested in supporting the DAISY mission will be encouraged.

Two major sources of funding are identified for DAISY Consortium activities. The first are the fees paid by the full and associate member organisations and the second funding source is grants.

Full member organisations pay an initial entrance fee and agree to pay a substantial annual fee per year for at least three years. Full members should expect to lead the development of the DAISY Standard. It is expected that full member organisations will `drive’ the DAISY Consortium and participate in the strategic activities described above. The DAISY Consortium will recruit leading libraries throughout the world to take a leadership role. All major decisions regarding the direction of the DAISY Consortium will be made by the full member organisations.

Associate members will pay a smaller fee per year and can participate fully in meetings and on sub-committees and work teams. It is expected that organisations with something to contribute to the DAISY Standard, but perhaps without the resources needed to drive the development, will join as associate members. The associate members will be kept current with all developments and decisions. The DAISY Consortium values the input from its Associate members and appreciates the support and assistance of associate member organisations in all activities.

The DAISY Consortium will apply for grants to develop, promote, and advance the DAISY mission. In addition, full and associate member organisations are encouraged to apply for grants that contribute to the DAISY Mission. The DAISY Consortium will present a unified granting system-member organisations and the Consortium as a whole should not compete for grants.

It is considered that the distribution of technology to developing countries, coupled with sympathetic training programmes will enable libraries serving visually impaired people in these countries to take advantage of the interchangeability of product. This course of action should enable their clients to become better informed, better educated and will instill a concept of usefulness.

From 1993 to January 1998, Consortium member organisations spent $2 million in support of the DAISY Project, (including the world field testing). In addition, the Consortium will spend an additional $1.3 million during 1998-2000 for software development, maintenance, training and support. From January 1998 to January 2000, the Consortium will collect membership fees of $1 million. In addition, each member organisation allocates significant funding to implement the DAISY Standard within their organisation.

The Consortium understands that developing countries will require assistance from the industrialized world to move the DAISY Standard into these regions. The Consortium will seek world-wide funding for expansion to developing countries.

The DAISY Consortium is conducting activities on a worldwide basis. To be effective, Consortium members will use modern technology (primarily E-mail) to communicate. The strategic decisions are made by the full member organisations, but work team structure is employed to effectively complete the work.

I hope that this short outline gives you an idea of the achievements of the Consortium so far. We are currently beta testing the recording software. There are two dedicated DAISY playback devices on the market. Libraries have commenced re-archiving. We all feel confident about the way ahead.