Computer Services for the Blind

The ARTS system (ARTS is an acronym standing for Audio Response Time Sharing) was conceived by Dr. Kenneth Ingham, and he has brought the system to its present state of development. The ARTS system is a configuration of computing machinery that is programmed to provide a variety of services that should be useful to blind persons. These services are delivered by telephone, and since it is a time sharing system, 16 users can be accommodated at one time. The input to the system is provided by a typewriter-like keyboard, connected to the user's telephone. The output of the system is computer-compiled speech, heard over a loudspeaker, also connected to the user's telephone. Services planned for the system include dictionary consultation, composition of letters and other manuscripts, bookkeeping, filing, mathematical computation, transcription from print to braille, computer-aided instruction, computer programming, and so forth.

My report for the year before last included an account of the events I initiated that led to the establishment, by an action of the Kentucky State Legislature, of Computer Services for the Blind (CSB), a non-profit public corporation charged with implementation of the ARTS system and with the development of other computer-generated services for the blind citizens of Kentucky. The State provided $173,000 to meet CSB's costs during the first two years of its operation.

The transactions of CSB are supervised by a Board that includes the Director of the Division for the Blind, Bureau of Rehabilitation Services; the Superintendent of the Kentucky School for the Blind; the Director of Industries for the Blind; and six members appointed by the Governor of the State. I was one of those appointed by the Governor, and I am currently serving as Chairman of the Board. The Board appointed Glenn Smith as the first Director of CSB, and he served in that capacity xantil recently. The University of Louisville has made available to CSB, without charge, the space it needs for computing machinery for the administration of its business.

Initially, we believed that the development of the ARTS system was substantially complete. We expected to purchase a system which, after correcting the difficulties routinely encountered during the installation of a complex system, could be put immediately into service. However, as our attempt at implementation proceeded, it became apparent that considerable development of both hardware and software was still needed. As a result, we were unable to adhere to the schedule set forth in our proposal to the State of Kentucky.

By the beginning of the current fiscal year, the system was supposed to be delivering services and earning income to defray operating expenses. Instead, it is not yet operational. The principal hardware components of the system have been delivered and interconnected. However, the system, as it stands, does not have enough capacity to meet the anticipated needs for the storage of user files. The terminals provided with the system are only minimally adequate, and a source for better terminals has not yet been identified. The quality of the speech produced by the system is not quite satisfactory, and the system's vocabulary has not been selected with adequate care. Although some software has been created, a major developmental effort is still needed to provide the software that will be required for full implementation of the services that were expected when the system was purchased. The system was delivered without adequate documentation. CSB refused to accept the system without such documentation, and many months have been spent in acquiring it. A final acceptance procedure has been negotiated, and will be carried out shortly. If the vendor of the system can meet the tests specified by this procedure, responsibility for system maintenance will be transferred to CSB.

It is now clear to us that what we originally regarded as a service project will have to be regarded henceforth as a research and development project. The State of Kentucky funded our efforts in the expectation that services would be delivered. We have failed to do so, and the State of Kentucky is not interested in funding a research and development effort. As a consequence, our application for continuation of project support was rejected, and we will have to find einother source of support in order to continue the project when current funds are expended. In spite of the difficulties we have encountered, we remain convinced that the ARTS system can provide the services originally claimed for it, but we now know that a developmental effort of considerable magnitude will be required before full implementation can be achieved. The time required for this effort will depend upon the level of support we are able to obtain, but with adequate funding, it can be accomplished within three years.

A first draft of a proposal has been written, and we are now searching for a funding agency that might be receptive to such a proposal. In the meantime, with remaining project funds, CSB plans to use the applications computer that is a part of the ARTS system for the transcription of print to braille and, if time and money permit, for the generation of tangible graphs and maps. In our judgment, both types of material are needed by the blind persons served by CSB, and with the addition of the appropriate accessories to the basic system already owned by CSB, we can easily develop the capability to produce these materials.

The Transcription of Print to Braille

The NOVA 800 computer that is a component of our ARTS system is adequate for the recording that is required for the transcription of print to braille. The print that is to be transcribed into braille is typed on a typewriter that creates a punched paper tape. This tape provides input to the computer. The output of the computer is used to operate an automatic braille page embosser. The system can be made more flexible if provision is made for storing the output of the computer so that it is not necessary to operate the page embosser on-line, and so that the output of the computer can be preserved for future use. In the system we have planned, teletypewriters will be used to punch the paper tapes that provide input. We have one teletype-writer, and another one is on order. We have also ordered the automatic braille page embosser manufactured by Triformations. This machine embosses braille a line at a time, and it can emboss 120 lines-per-minute. We will be using cassettes to record the output of the computer, and these cassettes, when reproduced on a transport connected to the Triformations embosser, will control its operation.

In addition to the hardware components of the system, software will be needed to transcribe the input code to an output code that will cause the Triformations machine to emboss Grade 2 braille. CSB will contract with Glenn Smith, the former director of CSB, for the creation of the recoding program that is required.

The system we envision is one that will depend heavily upon volunteer services. A system manager will insure that the computing machinery is functioning properly, and will assist volunteers in the use of that machinery. However, volunteers will prepare input tapes, operate the braille page embosser, and bind brailled pages into volumes. Because we are using a minicomputer, our product will not be perfect. There will be occasional violations of the rules governing the use of contractions in the Grade 2 braille code. However, the occasional departures from perfect form will not have serious consequences for the reader. In fact, it is likely that many of them will be unnoticed by the average reader, who does not subject his copy to the scrutiny of a proof reader. Our primary objective is rapid response. We intend to satisfy the ongoing needs of students, and of blind persons engaged in professions and occupations. Under present management, there are well developed facilities for the volume production of braille, but there is inadequate provision for responding to individual needs. A blind person may request a group of volunteer transcribers to prepare a book for his use, but with the methods they currently employ, he will have to wait months for the delivery of that book. It is frequently the experience of the person who needs a book to meet educational or occupational requirements that, by the time the book can be delivered to him, the need for it has passed. If the system contemplated by CSB functions properly, it should be possible to bring about a very significant reduction in the time required to fill orders for transcribed braille reading matter.