Requirements for the Interconnection of Assistive Technology Devices and Information and Communication Technology Systems
John Gill
July 2001
Contents
Appendix 2 Relevant Assistive Devices in ISO 9999
Appendix 3 The Wireless Interconnection Technologies
Appendix 5 Responses to the Questionnaire
Appendix 6 Alternate User Interface Standard
Responses to the Questionnaire
The responses are listed in the chronological order in which they were received. They have been reproduced without corrections to the grammar or spelling.
1. In your opinion, would it be useful for there to be standard interfaces
between assistive devices and ICT systems? Yes. It will enlarge the
Digital Divide if there is an abundance of interfaces, each of which
would have to be learned independently.
2. What features should be covered by such a standard?
a. Type of modality (vision, speech/sound, touch/haptics)
b. Interface equipment (e.g. keyboard vs. standard mouse vs. vibrotactile mouse vs. speech, etc.)
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)? Yes; otherwise one has to learn a different interface for each transmission type.
4. What other aspects should be considered when preparing this standard?
a. User-friendliness
b. Cost
c. universality of access across software platforms
Golledge, Reginald G.
University of California, Santa Barbara
Santa Barbara, CA 93106
John
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems? How can there be an answer other than yes? The provisos are whether the introduction of the standards stiffle innovation and potentially increases the price of products.
2. What features should be covered by such a standard? I have given very little thought to this and will reflect further. Initial thoughts are
* Input, e.g. keypad on microwave oven. Enabling peope with disabilities to control the device using a specific assitive technology solution.
* Output, e.g. display on microwave oven. Enabling people unable to access the display (print impaired) to "read" information on the device. This will include status information, e.g. if your microave oven is downloading a recipe from the web.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)? I think so. However, I don't know if the transmission systems have their own standards for protocols. If they do, it may be expedient to follow them and bring the industry with us rather than pushing against it.
4. What other aspects should be considered when preparing this standard? Further reflection required.
Paul Blenkhorn
Senior Lecturer
P.L.Blenkhorn@UMIST.ac.uk
Department of Computation
UMIST
PO Box 88
Manchester
M60 1QD
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems? Yes. But I believe that in many cases an existing standard can be used, and this is always preferable to creating a new standard. Where possible the interface should text based, since text can be easily translated into all modalities (print, speech, sign, Braille). The text should be marked up with HTML and XML to show structure and semantics.
This argument leads to use of web interfaces. The assistive device is the client, and ICT system the server, or occasionally vice versa. A major advantage of such an interface is that it is distance independent. A disadvantage is the limited interactive possibilities of HTML - essentially there are only forms.
2. What features should be covered by such a standard? Ideally all the interfaces of an ICT product should be made available to the assistive device: but in practice some high bandwidth interfaces may not. Interfaces cover both input and output - generally the output has the greater bandwidth, especially if input is merely for control.
For example for a TV, with an assistive remote control, the volume control and channel hopping is available to the remote, but also subtitles should be available if there is bandwidth. However one would not expect the full audio to be available, and certainly not the full video.
With a mobile phone, as another example, the keypad input should be extended so the user of the remote device can dial numbers, e.g. by speech recognition. Text messages should be carried in either direction, so they can be sent and received on the assistive device. If there is bandwidth (as on Bluetooth), the speech should also be available, both outgoing and incoming.
The features should be sufficient to cover an assistive device withdrawing electronic cash from an ATM. So there needs to be privacy and security considerations, especially since other people in the vicinity may also have assistive devices.
The features should allow one to many links and many to one links, e.g. an ICT product with many assistive devices able to control it, or a single assistive device able to control a number of ICT products.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)? Yes. But note that the higher the bandwidth, the more modalities need to be covered.
4. What other aspects should be considered when preparing this standard? The AIAP protocol has been developed to allow a negotiation between the product and an assistive device, to customise the interface appropriate to the capabilities of the user. I'm not sure this is necessary. I think that the protocol should allow the extension of all input and output interfaces to the assistive device, and the device and/or transmission system will reject those interfaces it cannot handle.
A smart assistive device may present the user with interfaces according to the user's situation: e.g. vibrating alarm in the concert hall, but acoustic alarm in the bedroom (to wake up).
I also take a user-centric view of control. The user decides on the product to be controlled and the user initiates the protocol via the assistive device. However there is a case for allowing a product to announce its presence, e.g. when the user walks into a room with a TV, or as the user walks past an ATM.
John Nissen, Cloudworld Ltd., Chiswick, London, UK
Tel: +44 (0) 845 458 3944 (local rate in the UK)
Fax: +44 (0) 20 8742 8715
Email: jn@cloudworld.co.uk
Web: www.cloudworld.co.uk
and www.wordaloud.co.uk
1. Yes, definitely.
2. messaging, ??
3. Yes. The items communicating don't need to know what the transmission system is for this to work.
4. Existing and future planned standards should be considered. There is current work for protocols for household devices/appliances to communicate, among other things, which could possibly just be expanded upon.
Jason Fritz
Raytheon
Dear Mr. John Gill:
I am answering your inquiry on behalf of Mr. Kipke. In our opinion, it ist useful to have standard interfaces for different appliances. Part of our work and the work of other participants in the project embassi is exactly the problem of developping such interfaces. If you wish to receive more information on this subject, don't hesitate to contact Mr. Peter Albert (peter.albert@Teletech.de) who is our representative for this project.
Sincerely
Barbara Wahl
Sekretariat
Handy Tech Elektronik GmbH
barbaraw@handytech.de
I feel it is necessary to have standardisation of the interface. The standard address protocol may become the best choice. We should not worry about hard wiring. Push fit or press fit type connectors are being widely used and it may not be inconvenient for the handicap to use these.
Ashok Kumar
Thank you for your e-mail concerning assistive devices and ICT systems.
I apologise for the delay in responding but I was away last week - in
London in fact.
My initial responses to your questions are given below.
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems?
Better interfaces between assistive devices and ICT systems would be very welcome but care has to be taken to define what is meant by 'standard'. There are potential problems. First, if a standard is enshrined as a legal directive or statute, it has to be policed. I know of companies that prefer to disregard standards and face the risk of fines if they are caught out, on the grounds that standards may be expensive to implement and difficult to include in product development. Second, a standard may have the effect of 'freezing time' and stifling product development. The military field is notorious for this. For instance, the standard 'MIL-HDBK-217' is widely used by industry to placate civil servants but is regarded by many as a costly way for achieving little. Nevertheless, 217 is often seen as a law. Third, standards may operate in competition with one another. There are directives to use more recycled products such as glass but glass is heavier than plastic and more difficult to lift for people with impaired hand function.
I would like to see more evidence of 'fit-for-use' in product design and that, in turn, means there is a need for better definitions for prescribing the routes for CE marking of products. To site an example, a Zimmer walking frame is a class 1 medical device for the alleviation of a relief of handicap. However, an electronic voice in a passenger lift is accepted as a convenience for the general public even though it is also very important for a blind person to navigate around a building. Is the electronic voice in a lift an aid to 'alleviate handicap'? Is it a class 1 medical device? Do we really want assistive devices to be classed as medical devices?
Many companies still take the view that they will do not develop products for the 'medical market' and people with disability and handicap are regarded as a niche market. I believe we should encourage the development of products as main stream consumer goods.
My background is in physical disability, particularly upper limb disorders, and I believe there is a need to develop better interfaces for operating devices by hand. For instance, the typical consumer wants small mobile phones but small size inevitably means that the push-buttons must also be small. In principle, it should not be a difficult task to provide a mobile phone with an optional enlarged keypad - there is plenty of scope for mainstream products with optional extras.
2. What features should be covered by such a standard?
Whenever there is a possibility of developing a standard to promote both the interests of the consumer and the manufacturer, then the standard could cover fitness-for-use, as well as all the usual safety issues.
To provide a very loose parallel example, the medical profession typically seeks to reach a consensus that a particular surgical procedure is the preferred way of treating a particular surgical need. They don't say that the identified surgical procedure is the best or only one for an individual but a balanced consensus exists based on published material etc. They certainly do not prescribe standards. If a method of identifying 'the best features' were used in product design, then a consensus might be sought and reached for 'best practice' similar to 'best surgical practice'. Kite marks could be awarded to companies in recognition of their good product design. Just as there is an 'Investors in People Award' for companies, there could also be 'Investors in Inclusive Product Design Award'. Norms for good practice might be set. For instance, there is plenty of data on hand strength so it is possible to foresee the possibility of a product award, 'Can be opened by a person with fifty percent of normal hand strength'.
3. Should the standard protocols be independent of the transmission
system (eg infra-red, Bluetooth, hard wire connection)?
We should take care not to try to spot the 'winners' when technology is rapidly developing - we might chose the wrong one. The worst case scenario is a standard that backed the wrong technology/protocol.
In the area I am working in at the moment (barcodes for supermarket products), there are no standards yet each day, millions of products are moved from warehouses to supermarket shelves and then to homes. In the absence of standards, there are published guidelines which a company ignores at its peril.
Somehow (and it will be a difficult process), I would like to see best practices guidelines used by product manufacturers so that a company can say, 'We are working at the recognised highest level of product development for inclusive design'. The company will want to work at that level for both increased product sales and external recognition.
4. What other aspects should be considered when preparing this standard?
As you can tell, I am a little wary about standards. However, just as television adverts sell a soap because it is 'gentle on your skin' (whatever that means), could we aim for a product kite mark that says, 'developed, manufactured and sold to meet inclusive design requirements'. It would be a challenging task to define what is meant by ' ... to meet inclusive design requirements', but no more difficult than to define a soap that is 'gentle on your skin'.
I hope this helps. Incidentally, I hope to be in London again in the next few weeks so if there is a suitable opportunity, could I call in and visit you?
Dr David A Carus
Department of Mechanical Engineering
James Weir Building
75 Montrose Street
Glasgow
G1 1XJ
Tel: 0141 548 4851
Mobile: 07957 480298
Fax: 0141 552 5105
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems? Definitely yes! But equal emphasis should also be put in to researching and providing usability and inclusive design!
2. What features should be covered by such a standard? Voice recognition features as well as virtual vision/assistive reading - for blind users/poor vision/technophobes (elders?). Web acessibility guidelines and requirements for all public services websites, public information, libraries and online databases, distance learning courses, computers and wireless devices.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)? Definitely yes!
4. What other aspects should be considered when preparing this standard?
Sistematic consulting and simulation with mixed users regarding age, sexe, litteracy, type and intensity of disability AS WELL AS with:
a) researchers from various domains (not only ICT or disabilities' areas) such as neuro-physiology, psychology, sociology, education, applied social sciences;
b) previous existing services, systems and equipments which are already provided and fit for disabled users.
Please let me know more about your survey... Lao, if possible, please be so kind to indicate weblinks, references and databases regarding web acessibility / web inclusive design. Thanks a lot!
Best regards,
Ana Isabel B. B. Paraguay
Dept. of Environmental Health
School of Public Health / USP
Av. Dr. Arnaldo, 715 - HSA
01246-904 Sao Paulo - SP
BRASIL
Dear Mr. Gill,
the obvious answer to the three questions 1 and 3 is YES. Questions 2 and 4 need more time to complete.
Regards
Rafael Romero
1. In your opinion, would it be useful for there to be standard interfaces
between assistive devices and ICT systems? Yes, as an overall approach,
but it can be a quite complicated subject.
2. What features should be covered by such a standard? Extended keyboards, single-switch operation, magnified screens, voice synthesis, voice recognition
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)? Yes, it should be an independent communications layer, adaptable to any current or future transmision system.
4. What other aspects should be considered when preparing this standard? Probably a unified model of a basic human interface to extract a basic layer of user interaction.
Best regards,
Javier Romañach
1. In your opinion, would it be useful for there to be standard interfaces
between assistive devices and ICT systems? Yes
2. What features should be covered by such a standard? For all, location and configuration of control keys, keys should give tactile click when pressed, NO touch screen, universal accessibility for sensory impairment = customizable displays (e.g., with user defined setup parameters on a smart card). If low vision, optimum print size for a given working distance, contrast polarity, color contrast enhancement, screen luminance, field locator features (e.g., should active field be central on screen, control keys for jumping from field to field), font type, letter spacing. If low vision and or blind, speaking rate, speaking loudness, pitch, control keys.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)? Yes, but they need to be adjusted for expected working distance and screen size
4. What other aspects should be considered when preparing this standard?
Steve Whittaker
1. Generally I think that a standard interface always is better. It
makes it easier for the end user if you don't have to rely on just one
manufactory to give you access. On the other hand, sometimes you have
to go for a non-standard solution just to provide the end-users with
something until the standard is settled.
2. I am not updated enough to answer that question.
3. Yes if possible.
4. I am not updated enough to answer that question.
Diana Hiorth
Software Development Manager
Labyrinten
Data AB
Dear John Gill:
I will try to answer your questions below:
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems? Yes of course
2. What features should be covered by such a standard? I think that this depends very much on the type of assistive device and application. It would be difficult to put everything here on the fly. I guess the most important is the communication protocol.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)? Ideally yes. If the protocols are independent of the hardware (level below) everything will work finer. I would be a very good success if it is possible to create a protocol independent of the hardware. The only problem is the cost of the final system.
4. What other aspects should be considered when preparing this standard? I think that the aspects that should be covered are the requirements of different applications and is important to try to include all possible applications under the same standard.
J.M.Pardo Ph- +34-91- 3367311, +34-91-5437597
Director Dpto y Director G.Tecnología Habla Fax: +34-91-3367323
Dpto. Ingenieria Electronica
ETSI telecomunicacion- U. Politecnica Madrid
Ciudad Universitaria
28040 Madrid
Spain
e-mail: pardo@die.upm.es
WWW : www-gth.die.upm.es
I will attempt to answer your questions below.
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems?
I believe there should be a very simple interface between ICT and AD devices, so that they can co-operate and work together as one. The seamless integration of this is very important and more importantly one should not need a PhD to set-up and maintain. In summary extremely simple standard interface.
2. What features should be covered by such a standard?
I think in generic terms the standard should cover the ability to get and set attributes for objects. Where objects could be a button, screen or anything else.
Without specific examples I can't ascertain if the simple set and get will solve the problem. My only worry is that if there needs to be an event mechanism, for example the phone ringing, but this should not be difficult to incorporate.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)?
This is good protocol design and will happen. However what is important are the media factors of wireless versus hardwire. Extra intelligence is required to make or re-establish wireless connections securely where as the hardwire it is simpler.
4. What other aspects should be considered when preparing this standard?
I don't think we should prepare a standard but instead search for a standard or adapt an existing standard into achieving the final goals.
Also we should search for a standard that is widely adopted and has been tested and used extensively in industry and consumer products. Finally, if at all possible if the standard is maintained by Open Source Foundation which will mean that it is free and secondly contributed to by the best brains in the world.
Examples of such standards are :
SNMP - Simple Network Management Protocol, which views everything as a file structure and sets and gets attributes.
OBEX - Object Exchange, which views the world as a set of objects and gets and sets attributes. Obex is supported by most mobile phones and is used by the Bluetooth SIG, but is only used for data exchanges, while AT-commands are used for Telephony.
Although not in the above category the Bluetooth SIG are doing Human Interface Device (HID) profile. Where Phones can be controlled remotely across a Bluetooth link. However, this is a closed group so I do not have access to what it involves.
I hope the above answers your questions and I will do my best to get more information on the HID definition.
On a final note would you be interested in borrowing some of the Bluetooth Gadgets I have, so you can imagine different ways of using them. I have the following Bluetooth products:
* Headset
* Phone
* PC Cards
* Dummy Pen
Bill Munday
On behalf of blueAid founders
Email:: founders@blueAid.com
Voice-Email:: +44 (0)7092332348
Fax:-Email:: +44 (0)7092332358
Website :: www.blueAid.com
Dear John
Whilst I note your comments and agree in the main, I feel you would need to be more specific in order to initiate any discussion.
Assistive listening devises abound, some audio and others tactile. In my experience the convention that they follow is one of convenience and is market driven. Manufacturers in the ALD's market use conventional 'audio' plugs and sockets, however our latest range of products for use with TV / AV or even PC's are now supplied with a JACK lead (3.5 & 6mm), SCART lead and self adhesive microphone. The latter is only supplied so that the products are retro-gradable to work with products that don't have available audio sockets.
The personal alerting products that we now supply can be used by the visually impaired and have different vibrating patterns to indicate FIRE, doorbell, baby cry, Fax: machine and telephone etc. These could be standardised to provide uniformity but at an increased cost for everybody.
We could standardize general public provision under the DDA to insure compatibility and uniformity of product and service at venues such as theatres and cinemas etc. but that would mean nominating one system as the correct approach and expecting others to modify their products in order to conform.
Manufactures are often restricted by component or home-market / mass-market demands which become a case of pleasing all of the people, all of the time.
I appreciate that this is not answering your questions directly and that I am playing devils advocate, for which I apologise, however, I would be interested in discussing the topic further and would be happy to meet at some point if you prefer.
Very best regards
Owen Lang
Managing Director
BioAcoustics Limited.
To: John Gil
Response to your Accessibility questions from Myrna Votta, Lighthouse International. 5/01
1. "In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems?" Definitely. If assistive devices need to communicate with other systems and/or devices, there must be standards which must be followed.
2. "What features should be covered by such a standard?" One or two interfaces should be chosen and those interfaces need to be standardized. The BrailleNote is a device which has an infra-red interface. However, I believe that a standard for an infra-red interface does not exist. A USB interface would be another good choice because the consumer would only need to carry one cable which would work with all systems.
3. "Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)?" No opinion.
4. What other aspects should be considered when preparing this standard? Ideally, all communication between ICT devices/systems and assistive devices would be accomplished through one standardized interface and that interface would either be blue tooth or infra-red because there would be no cable that the person who is using the assistive device would need to carry or connect. If there were such an interface, someone using an assistive device would then be assured that he/she could successfully communicate with all ICT devices/systems.
John, my apologies for not responding earlier. Hope all is well with you...and you're obviously still active in these very important areas.
My answers (which are "personal" and not attributable to any organization) are as follows.
1. I don't think so. I think that assistive devices should conform to
existing standards of various types, which seem adequate to the purpose.
This will also permit best performance and best opportunity, since taking
advantage of the "mass market" is really the only way that disabled
people can really get what they need, in the long run (and I'd be happy
to elaborate on that and explain it more, with examples, if you wish).
However, that presumes that existing interfaces are not designed in such a way that there are intrinsic limitations to the bidirectional exchange of information. For instance, the transmission of interface information in solely graphical form is useless for blind people, and hence interface standards in such a case should be changed to mandate multi-media transmission. Probably, straight digital transmission (at least being "available") is adequate to meet the need.
2. My second paragraph above is the only general example that comes to mind.
3. Not exclusively for disabled people. Only if impelled by mass-market considerations. I also find the question somewhat unclear (but put this comment after my answer! ah, well...).
4. Since I think a "separate standard" is neither useful nor desirable, I have no aspects to recommend for consideration. However, for others who may differ, provision of a categorical list of such aspects (by you), which recipients may consider and respond to, might result in more useful response.
Cordially,
Larry Israel
4454 Deer Ridge Road
Blackhawk, CA 94506
Phone: 925-648-7582
Fax: 925-648-7584
Dear Dr. Gill,
Thank you for including us in your survey for information. I have inserted answers below each of your questions, below.
Many information and communication technology (ICT) industries are working on making their products usable by disabled and elderly users. When inclusive design does not meet all the users' needs, it will be necessary to use assistive devices. Currently there is little standardisation of the interface between assistive devices and ICT systems.
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems? From an assistive device manufacturer's point of view, standard interfaces would clarify our product requirements and ensure our customers maintain some consistency in their equipment.
2. What features should be covered by such a standard? From your recent publications, you have highlighted the significant features. e.g. button shapes, positions and typefaces. I would also suggest the standardisation of audio signal feedback. For example, a short quick beep is used to indicate success or a no-error situation whereas a long beep would indicate an error. There may also be the opportunity to define the frequency of a beep or the relative frequency. For example, an increasing frequency indicates a higher number is selected, and the audio feedback decrease in frequency as lower numbers are selected.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)? Ideally the standard should be independent of the application, for two reasons. Firstly, the user should not need to be aware of the underlying technology and secondly, the standard will endure changes in the technology.
4. What other aspects should be considered when preparing this standard? It appears you are already addressing the main issue - international co-operation. If these are to become standards, who is the regulatory authority?
John Davey
BRYTECH Inc.
www.brytech.com
johndavey@brytech.com
Tel: (613) 731-5800
Fax: (613) 731-5812
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems?
Absolutely yes. In our Walker, we are building a device that is 'modular', in the belief that we can continue to develop the device and add functionality and features over time. We not only want to build autonomous devices, but ultimately complete smart homes where devices interact with each other and with the person. This will happen faster and better if all electronic devices have the capability to interact with each other at a basic level.
2. What features should be covered by such a standard?
On the one hand it should be at a hardware level, but it would also be useful to see harmony at a user interface level, with switches/symbols essentially meaning the same thing, etc.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)?
Yes, I would have thought so.
I am sorry to answer you if late and also for the quality of my English.
It is very important to be able to work with a standard hardware interface which one can find on any market device likely to contain information as computer, telephone, washing machine, computer, téléviseur..
It would be interesting to be able to control information of input and output of an unspecified electronic device since outside by a hardware wire connection or infra-red or Bluetooth following the context. The data exchanges would be serial type. If such were the case the realization of a technical assistance for anybody handicapped would be more effective, simpler, more reliable, faster and less expensive. One could even consider a general device of interrogation of these systems adapted for each deficiency. The protocol of exchange should be independent of the system of transmission, it is still better, it is not too difficult to realize, I believe.
I think that a hardware wire connection at least with two wire (serial line input and ouput) on each apparatus would be a solution at low cost for the manufacturer. The Institutions could propose to the people handicapped general devices of transmission based on will infra red or Bluetooth or both at the same time to be able to communicate with any technical assistance. The control protocol of the market device must be public (available on Internet). This is interesting for all, handicaped and no-handicaped
I do not know if, all that, is easily realizable but this way was very interesting to create new devices to disable person or no-disable
Sincerely yours
Claude Liard
Dear Dr Gill,
Thank you for your enquiry regarding standards for interfaces between assistive and mainstream devices. At first glance, this seems to be such a huge area, with vast numbers of new systems and devices coming into being, that it is difficult to see where one begins to develop interface standards. However, the very size of the issue means that it should be examined; it is certainly a challenge.
Most new devices, such as PDAs for instance, are small and have small displays thus presenting obvious problems for visually impaired users. These devices incorporate sophisticated embedded systems and software, with either unused processing potential and in some cases the ability to expand their functionality by live software updates for instance. For that reason, I believe, that there is the potential to incorporate new interfacing protocols into many of these systems, to facilitate their operation with assistive devices.
So, here are a few thoughts given as answers to your specific questions.
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems?
It would be useful for mainstream designers to have some standards to follow, as there seems to be none. If it is going to happen anyway it might be a good idea if assistive technology developers are at least aware.
2. What features should be covered by such a standard?
Any standards should pertain to higher-level interface and functional protocols. The needs of different end-users should be considered, and related to the scope of assistive device functionality they require. It should allow a broad scope of functionality to be supported and not limited by the requirement of being "standard". We'd have to ask questions such as: is it to be a standard for blind and partially sighted assistive devices taken together? Will it include deaf, and deafblind or interface devices for physically disabled users and so on? Should we have a family of standards in fact?
As an example, blind and partially sighted end users of a hand-held device including a display showing graphical information might be required by such standards to output the display data alongside textual representations of: selected items, the whole display, the user's context in a control hierarchy (where-am-I?), and help information as requested to an attached assistive device. Depending on its own functionality, an attached device may then be able to speak or Braille the textual information yet also magnify the display.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)?
Ideally, yes. The control and data transmission systems could be assigned to lower interface levels in a hierarchical interface model. However, the requirements of the overlaying assistive interface protocol must be serviced. For instance, such things as the rates of transmission and type of data passed could limit the functionality supported; screen enlargement over a slow serial link to a special monitor may not work very well. They need then not be defined, but their requirements need be specified.
4. What other aspects should be considered when preparing this standard?
Cost of the solution should obviously be weighed against future savings once such interface standards have matured and are widely used.
I have touched on this one in 3, but responsiveness is another factor, which I think from our own experience is often not given due weight. It is possible for an interface to introduce system latency, which is frustrating not only for the end user but also for assistive technology designers dealing with a latency problem. If it arises from an interface they have to use, it is difficult to do anything about it. So, a standard can help circumvent the problem by requiring fast interface responses.
I hope this helps, and I look forward to hearing how the matter develops.
Kind Regards
John Rye
Technical Director
Dolphin Oceanic Ltd.
Dear Dr Gill
Apologies for the late reply.
You asked some questions about the standardisation of the interface between assistive devices and ICT systems.
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems?
If you are refering to hardware interfaces. I feel that the breadth and depth of the ICT applications that a disabled person may want to access, would make the task of defining one standard interface that suits, very difficult to solve. There is also the question of the standard interface being adaptable to the range and severity of the users disability or aptitude. Also, should the nature of the input requirement of the ICT application change this would necessitate a redefinition of the hardware standard and expensive upgrades. If you are refering to data/command exchange interfaces then this may be more adaptable for the disability and the ICT application input/output evolution.
2. What features should be covered by such a standard?
The standard would need to be generic and cover all forseeable ICT applications. We would need to understand the range of interactive objects that are exchanged between applications and user interfaces - eg command objects, data request objects, data objects. If we define a method/protocol for this exchange then it becomes hardware independent. Obviously some data objects would be useless to some disabilities so part of the standard must be the potential for the application to request what are the capabilities of the User Interface (UI) - or the UI must ignore those data objects and/or request the appropriate data object. This concept is an advancement to 'personal accessor' proposed by Dr Neil Scott on the archemedes project at Stanford (?) - although without the reliance on any specific hardware protocol convertors. I imagine that initially it will be a SW 'class' library type structure that may call upon any of the available comms services to 'talk' to the user's interface.
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)?
Most definitely. For maximum flexibility the standard must only define a logical structure for the control/interogation and data transfer aspects of the interface. The implementation for this may be in software, firmware, hardware or combination. The market place will determine which transmission system will evolve to be most useful for this purpose.
Dr Robin Williams
Development Manager
Pulse Data International Limited
V: +64 3 353 2870 or +64 3 384 4555
F: +64 3 384 4933
Email: robinw@pulsedata.com
Web page:www.pulsedata.co.nz
www.Braillenote.com
1. In your opinion, would it be useful for there to be standard interfaces between assistive devices and ICT systems? Most definitely YES!
2. What features should be covered by such a standard? Data and voice
3. Should the standard protocols be independent of the transmission system (eg infra-red, Bluetooth, hard wire connection)? YES if possible
4. What other aspects should be considered when preparing this standard? Current standards and practices used in industry and with Assistive Technology.
Toan Nguyen