The Role of Digital Technology in the Lives of Deafblind People
Return to Blindness and Visual Impairment
Presentation given at The Royal Institution of Great Britain - Talking point series of Events in association with Sense
Date: 28/06/2005
Venue: The Royal Institution, London, UK
Article
Abstract: In considering the role of digital technology in the lives of deafblind people, we need to come to terms with the simultaneous phenomena of absolute advantage but comparative disadvantage. Public policy has tried to overcome ‘digital exclusion’ but it has severe limitations. The key developments which will benefit deafblind people during the next decade are: the use of programmable, cable free user interfaces to drive ubiquitous processors and a wide choice of feedback devices such as braille displays and large screens; the improvement in speech to text and text to speech automated translation; the employment of search algorithms which alter according to user preference and behaviour; the rendition of solid objects through 3D printing; and the representation of movement through haptics. All these developments are either in use or in advanced development in laboratories and the danger will not be that their prospects are hyped but that traditional approaches in the public and charitable sectors will persist.
Once upon a time, the rich man rode on a horse and the poor man walked; then the rich man rode in a carriage and the poor man rode on a donkey; then the rich man travelled in the First Class carriage while the poor man travelled Third Class; now the rich man flies in a plane and the poor man takes the bus. In these snapshots of transport inequality you can see that the differences between the rich man and the poor man are not simple and uniform. Over time the poor man graduates from walking to bus riding while the rich man graduates from the horse to the aeroplane; so both are better off absolutely but the gap between their speed and range of travel has widened, so the poor man is comparatively disadvantaged. At the same time, when they both travel in the train one has a higher quality experience than the other even though they are both going at the same speed; and by percentage a far larger number of people nowadays use aeroplanes than used horses a thousand years ago.
This was not precisely the case with information technology in the 20th Century; but it shows how technological development affects people unevenly. It is fashionable to talk about the "digital divide" even though the uptake of satellite television worked its way up from the poor to the rich; and although mobile phone penetration has taken longer to reach poor people there is not so much a divide as an element of differential speed. The problem is that however you look at the phenomenon of change, political, social, technological, the more rapid the change the wider the gap between the innovators and the cautious, the rich and the poor, the strong and the weak.
Of course this rapid change has massive public policy implications which we have seen recently at the macro level, for example, the UK's assertion in the EU context, that you cannot have an information society (The Lisbon Agenda) if you spend huge amounts of money on agricultural subsidies. At the EU level, then, there is a certain feeling of living in a parallel universe when discussing access to digital information systems. Still, the EU's great advantage is that it works within a broad rights framework, so it does tentatively acknowledge the right of people to information, even if that right has not been promulgated clearly enough and proved through legal test cases. The greatest failure here has been that of lobbyists to secure automatic copyright for information in accessible formats. The latest Copyright Directive derogates this to nation states but the problem here is that without a generic right of access, you have to lobby every time a new medium is developed. It took more than 100 years for the UK Parliament to pass a law granting automatic transcription rights for braille, so you can see how difficult it will be to get similar legislation on electronic text unless there is a generic right of access to information in the public domain.
In the context of the United Kingdom where we do not have a rights-based constitution, in spite of the rhetoric of rights, the approach is necessarily much more pragmatic; the Government has to deliver an education system that is fit for the information age and it has to provide skills training and an employment framework to match.
I want to say at the outset that the Government's willingness to move forward on this agenda has been exemplary since the PAT 15 initiative of 1997. There has been an almost endless consultation process and a steady flow of policy. There have been pilots in digital information in every educational, employment and social context but you still get the impression that although the Government knows that information technology is absolutely vital, it doesn't quite know why.
Here are a few characteristics of public policy that we should think about carefully from our standpoint as interested in the access by disabled people in general and deafblind people in particular to information through digital systems:
- First, there is always a gap between people and systems which can be narrowed from either side; you can improve both human and system performance. The Government has tended to put all the emphasis on human performance, through the development of the basic skills agenda. Its philosophy, put crudely, is that if only these people would sit down nicely and learn to word process, the world would be a better place; but they won't and it won't. In fact more than half the problem arises because of poor design. Bad design is a cost shift from the producer to the consumer and instead of paying for training to handle these badly designed systems, Government should use its procurement power, as should major industries, to guarantee a basic performance standard.
- Secondly, the basic skills development agenda has two serious problems: the first is that machine processing is moving faster than the agenda so IT jobs are becoming automated faster than new skills are being developed; in other words, we are giving people the kind of skills that will keep them unemployed. Linked to this, the agenda is still largely based on solo achievement when the core skill for working in IT is good collaboration. This reflects back into the National Curriculum which is still too much based on solo achievement and which therefore does not reflect real life; this presents special problems for disabled people who are much more economically viable in teams than alone.
- Thirdly, and we will come back to this in a minute, the public sector is obsessed with PC-driven systems and seems almost completely oblivious of developments in broadcasting and telecommunications
- Fourthly, when the Government thinks of access, it largely thinks of physical access to equipment rather than the accessibility of the equipment. There are many people in Government who think that the access agenda is complete even though only about 20% of public sector web sites are accessible at basic Level A to blind people. The reason why accessibility is dropping down the Agenda is not because the Government does not care but because it genuinely believes there is no longer a problem; and part of the reason for that is that disability charities have been woefully supine.
- Fifthly, and critically, there has been a debate about connectivity for more than two decades now which is currently exemplified in the vote tonight in the House of Commons on Identity Cards. In the late 1980s I went to visit Howard Rheingold, that great early guru of the Internet; on a visit to his seminal WELL Network I asked him what the future held and he said that the Internet was a wonderful recipe for anarchic free expression. I countered that if there was any money to be made out of it then global business would ensure the anarchy never happened, unless, as is unlikely, you were willing to pay for your anarchy. What has now happened is that this debate has been transmogrified into a paradox: we don't want Big Brother but we do want 'Joined up' Government. This debate is currently somewhat theoretical because neither of these is deliverable because of problems with scalability; but what this means for all of us - but which has particularly irksome consequences for deafblind people, is that you have to do a lot of learning every time you visit a new public sector web site.
- Finally in this section, I was recently asked for my attitude to Linux; my unlikely reply was that I am all in favour of the survival of Microsoft, at least for the time being. This is because by 2010 the world’s Internet will be overwhelmed by the entertainment industry which will want to set standards; and I would rather have Microsoft helping me to fight Disney and Murdoch rather than trying to do it myself.
Now let me turn specifically to technological issues which affect deafblind people; I want to deal with four in particular:
- Programmable user interfaces
- Text-to-speech and speech-to-text
- Reactive search systems
- 3d printing & Haptics.
So let us begin in thinking about the role of information technology in the lives of deafblind people by going back to the idea implied at the beginning of my presentation, the dual phenomenon of absolute gain enjoyed simultaneously with comparative disadvantage.
In a nutshell, it is easy to understand that the deafblind person is now able to access more written information than at any time in history but is simultaneously less and less knitted into a social fabric which increasingly absorbs audio visual information every minute of every waking hour.
It is easy enough to talk about the problem, of those aspects of the audio visual which can never truly come alive for a deafblind person no matter how faithful and passionate the textual rendition; there is something about the smile of the Mona Lisa which no art critic can ultimately describe, something about Bruno Walter's 1938 Mahler 9 which no music critic can ultimately convey, and something about the sound and gesture of the Commendatore in Mozart's Don Giovanni which makes realisation in words geometrically problematic; but, having said that, there are exciting areas of opportunity for weaving the lives of deafblind people into the social fabric, deepening their own personal satisfaction and broadening their life chances and I want to look at four of these in a moment after I have said a very brief word about the basis of what we are trying to do.
For better or worse, depending on your point of view, we are leaving the age of legislation and moving towards de-regulation and voluntary standard setting. There are exceptions to this and one of these, as I have said, is the growth in the idea of a right to information; but that is going to be highly contestable, particularly in the UK where we do not have a rights-based tradition. When the juggernaut of market forces makes contact with the bicycle of civil rights it is not difficult to guess the outcome. One symptom of this trend is the deregulation, except in the area of health and safety, of consumer electronics and communications hardware. There is no point spending any more time, energy or money on this. Conversely, we need to spend much more effort adjusting standard systems to special needs. A good example of this is the need to develop a robust SMS and email system for emergency notification rather than pressing 25 EU Members to agree to a parallel, special system.
Which leads nicely into my first major point. The key hardware developments over the next decade will be the use of programmable user interfaces and their physical separation and cable free communication with processors and monitors. This means that instead of having whole rooms full of hardware such as televisions, radios, CD players, MP3 players, games boxes, computers and mobile telephones, we will be able to run all these processes from one hand held device which can be customised so that we are comfortable with it. It may be of some marginal utility to try to persuade manufacturers to standardise features on television remote controllers, but pushing the development of programmable user interfaces is a much better option. These will not only work with your own processors at home and at the office but also with public information systems such as train departures screen; and, linked to GPS your system will vibrate when you are at the door of a shop you have previously specified. This cable free separation also means that braille displays will be more portable and lightweight, large screens will be carried and used everywhere. This will increase flexibility and customisation and will also cut costs. Of course, niche markets like ours will always be behind the early adopters of new technologies but in text access at least, this will not be critical.
Which leads me neatly to my second point. Although development has been painfully slow in the area of automated speech to text, systems are steadily improving. In the last twenty years this area has been particularly frustrating because there has been such good progress in automated text handling and even the conversion of text from graphics environments into manipulable text files, such as optical character recognition and PDF conversion. At the same time, text to speech quality is also improving so that its clarity, range and adjustability will extend its usability.
Many people would argue that there is quite enough text based information available without the need for automated speech to text conversion. That argument is one for another day but it does lead me to my third point. As information systems are increasingly deregulated, not only in a formal sense but also because publishing becomes an individual universal rather than a professional, corporate phenomenon, it will become more difficult to find what you want without some kind of brokering. I have no doubt that there will be brokering of the sort provided by algorithmic search protocols such as Google which broker on the basis of a set of rules. For deafblind people the key breakthrough is the use of search systems that react to individual requirements and behaviour. I like to tell the story of the email ranking system that puts the lovely Samantha's emails at the top of my in-box, even above those of my boss, until I see her hand in hand with another man at which point I ignore her emails which steadily slip down the in-box ranking to be superseded by those of the delectable Jackie. Behaviour based ranking, the elimination of repetitious instruction, the presentation of intelligence-based propositions as defaults and the cross referencing of, for example, correspondence and appointments books, will all take a great deal of the pain out of human/computer interaction.
After all that text it is time to think about harder things, about objects. In this area there are two major developments on the output side on which we need to focus, 3D printing and haptics but before that we need to take a brief look at the input side, at scanning. Normally we think about scanning in two quite distinct and separate ways: first, we think about optical character recognition scanning which turns a page of text, such as that found in a book, into an electronic file that can be manipulated and printed; secondly, we think of scanning in hospitals where a doctor and patient want to see a section of a brain or the status of a foetus. We usually see these outputs on a screen or a photographic plate but the scanning technology is actually recording a 3-dimensional artefact.
Having got that far, let us turn to the output side. There is something stunningly obvious about 3D printing. When you prepare a text file its output is notionally two dimensional, generating marks on a piece of paper; but those marks have depth; and so you can define a solid object by describing it in thousands of layers; just as threading thousands of still frames make a moving picture, so thousands of stacked frames make a solid object. Blind people have always been deprived because of the massive cost of analogue model making. 3D printing will cut that cost enormously so that individuals will be able to print the models they want. In a Gibsonian moment I once imagined 3D printers printing the components of 3D printers.
That is exciting enough but I think that in the long run haptics may be even more wonderful. In haptics you again have a computer file containing the specifications for a 3-dimensional object but instead of sending this to a 3D printer, you send it to a device which creates force fields. Confronting a solid object made through a force field feels like creating space by bringing two identical poles of two magnets together; you press hard to push them together but the mutual repulsion force creates something quite distinct from ordinary thin air; this is decided solid air. At the moment Haptics are generally quite good at simple solid objects but we do not yet know enough about how well individuals will be able to read highly refined contours; but what interests me about haptics is their potential ability, through combining stacked frames and sequenced frames, to convey the feeling of movement and change. The problem for blind people has always been the use of solid models to depict moving entities. How many blind people can learn from models how a horse gallops? Haptics is at an early stage of development and, as I have said, the refinement question needs to be looked at but as haptics files grow in size, containing ever longer sequences of ever higher stacks of frames, facilitated through compression technology, we may be able to use gloves to feel the beating of a heart instead of, as at present, using a thimble to poke a cube.
I always have to conclude talks such as this with a counter intuitive warning. The danger is not that you will walk away and exaggerate the possibilities that I have outlined but that you simply won't believe me. The reaction of most people to this kind of presentation is that it will no doubt happen at some time in the future but that day is far too far off for us to care about and so we must wearily continue to plough our familiar furrow. Or, to go back to my initial examples from transport, the danger is that we will continue to travel into the future with our backs to the engine so that we see the present in terms of the past instead of in terms of the future. All the things that I have talked about are already being used or are in an advanced stage of laboratory development; and the surest prediction I can make is that technologies such as those I have described will develop and come to fruition more rapidly than policy initiatives in the public and charitable sectors. In every organisation that works with disabled people there ought to be a champion whose job it is to look into the future and separate the hype from the reality and then push hard for change. In the area of deafblind need, the braille display and email have been immensely valuable advances but there is so much more we should be thinking about.
As a blind person myself, I have experienced far too much hype followed by disappointment and so I have been particularly careful not to subject you to a similar sequence of events. It is hard enough to be deafblind without having your hopes raised and then dashed by the febrile flow and ebb of technological hope and failure; so, as I have said, I have done my best to be absolutely realistic in what I have said. It is never going to be wonderful being deafblind; but it could be so, so much better.