In 1964 the MIT SAEDC undertook the systems design of a programming complex adapted to a more ambitious and flexible braille utilization system, than had previously existed(1). The system was dubbed DOTSYS (the DOT SYStem) and is described in some detail in the Proceedings of Braille Research conferences (1,2) and by Goldish (4).
DOTSYS ability to translate teletypesetter (TTS) tapes into grade II braille was demonstrated twice during 1966. The first demonstration converted news service tapes into brailler code punched paper tapes. These tapes were then run on the MIT High Speed Braille Embosser, the predecessor of the BRAILLEMBOSS, to produce the braille. The second demonstration converted tlw TTS punched paper tapes used to print a textbook into stereograph punched cards. These cards were sent to the American Printing House for the Blind where interpointed zinc braille plates were made on their card driven stereograph. The braille was then embossed in the standard fashion.
DOTSYS consists of a number of program co-routines or "boxes" each which manipulates the information being processed in response to computer directed requests from successive elements in the computation chain. This segmented approach to the programming of DOTSYS was predicated on certain projected advantages. Flexibility is achieved since new "boxes" can be introduced progressively into the system with but minor side effects on the rest of the system. Thus, new input media can be assimilated as it becomes available, the translation program can be upgraded, and new braille production techniques can be accommodated. Adaptation to computers of different sizes is facilitated since an overall processing operation can be segmented into blocks which fit the available computer, producing and storing intermediate results for batching operations. Finally, from a program writing and testing point-of-view, the "box" approach divides a very big overall job into digestible portions which individuals can program separately while maintaining effective communication with their co-workers, and the individual segments can be independently tested and debugged.
During the summer of 1967, the necessary parts of DOTSYS were written or modified to permit computer translated braille to be generated remotely from the computer. The necessary Input/Output (I/0) boxes were written for a time-shared computer (CTSS, an IBM 7094 at MIT).
The material to be brailled was typed into the computer by a typist using a model 35 KSR teletype. Then the typist completed typing the material, (or when a maximum of 60 lines were typed), the typist, via a typed command, initiated the translation. The Grade II braille was sent to an MIT High Speed Braille Embosser through the teletype. (During the time the Embosser was printing braille, the Teletypewriter was printing meaningless hash.) The braille is correctly paged and of the standard format.
It was this system which was demonstrated at Perkins during the winter of 1968. The operation of the remote braille production system was taught to approximately 48 members of the upper school faculty. Enthusiastic approval of the concept around which the system was_designed -- the production of braille material by an individual who was not familiar with Grade II braille -- was unanimous. Hany of the teachers were familiar with special forms of braille, such as the Nemeth mathematics code, yet understood only superficially the English encoding. Others could read it quickly, but were comparatively slow transcribers. Still others simply did not have time to mnke several braille copies themselves and were dissatisfied with the quality and necessary waiting period for volunteer supplied braille material. The most encouraging result of the demonstration was that well over half of those who used the equipment stated that were it available, they could continue to use it several times a month, even without any further modifications.
Development of DOTSYS was not continued further for several reasons. First, the program is in Fortran Assembly Language (FAP) for IBM 704 and 709 computers. These computers are now obsolete and have been superceded by the System/360. FAP is a machine language and cannot be readily transferred between similar computers and cannot be used on the 360 series without complete reprogramming or by emulation (now not readily available).
At the time of the Perkins demonstration the then current version of Embosser could operate at only one-half teletype transmission speeds and then only with frequent attention of the experimenter.
These limitations have been overcome. DOTSYS II has been written in a higher level computer language, COBOL, a nearly universal language. This language is available on most large computers regardless of manufacturer. Further, the BRAILLBEBOSS has been developed to the point where it works reliably for long periods of time at MODEL 35 Teletype Speeds.
PERKINS EXPERIMENTS WITH COMPUTER PRODUCED BRAILLE Benjamin F. Smith, Assistant Director DURING the week of January 22nd, Professor Dwight Baumann of the Mechanical Engineering Staff at Massachusetts Institute of Technology and Mr. William Greiner, a graduate student in Mechanical Engineering at the same University installed two pieces of special equipment in an oftice on the Perkins Campus. One of these pieces of equipment was an ordinary, electrically-driven teletype machine which has a keyboard with the same arrangements as a standard typewriter, but with a few additional buttons on it and a telephone dial attached. The other piece was a braille embosser console which embosses automatically when electrically stimulated. A telephone line previously installed connected this equipment to an IBM 7094 computer on the M.LT. Campus in Cambridge. Mr. Greiner typed onto the teletype machine some printed material from a sheet of copy. On completing the typing, he reached over and pushed another button. For about a minute, there was dead silence. Then suddenly with a series of muffled staccato reports, the electronically harnessed brailler began to operate. Without the touch of a hand, the embossing head traveled back and forth across the brailler leaving in its wake lines of perfect Grade II Braille upon a roll of braille paper moving upward from the carriage of the machine. Quickly, the electric brailler filled a large page of paper with braille characters and then returned to silence, waiting for more instructions. It had reproduced in Grade II Braille in an incredibly short time, the print copy material which Mr. Greiner had typed onto the teletype console. It was then possible, had he wished to do so, for Mr. Greiner to instruct the computer to repeat this operation and to produce further copies. This particular hookup of teletype machine and braille embosser to put print material into Grade II Braille is the brainchild of Mr. William Greiner and other students before him. Under the direction of Professor Baumann. Mr. Greiner has adapted this machinery as his Research Project for his Master's Degree. Here is an interesting way in which the modern computer can assist in the education of blind children. Teachers who wish to obtain quickly two or three pages of braille material to put in the hands of a blind student need know only how to type and, in a matter of minutes, they can have in their hands the required braille reproduction. The use of equipment of this kind, of course, is not inexpensive. The building of an electronically controlled braille embossing machine itself is costly and a modern teletype machine is not cheap. Funhermore, the rental of time on any of the busy computers in use today is still very expensive, though the costs are coming down. How long it will be before systems of this kind will be economically feasible for schools for the blind is still unknown. This particular equipment will soon be removed from Perkins and tested elsewhere. However, the results of the experiment at Perkins would seem to indicate that the general concept is entirely practical.
PERKINS ENTERS THE COMPUTER AGE WITH our participation in the Instant Braille Project, described else- where in this issue of the LANTERN, Perkins makes its first venture into the Computer Age. During the ICEBY Conference at Perkins last August, it was pre- dicted that before the next meeting of the Council in Madrid in 1972, some of the schools represented would be using computers in one or more ways. We did not anticipate that in five months rather than in five years we would be "sharing time" on a computer at Massachusetts Institute of Technology. The instant production of braille is only one way in which computers are likely to serve our students. We are investigating the possibility of participating in programs such as the one at the Kiewit Computation Center at Dartmouth College. In this program a computer is shared by a number of high schools and colleges with pupils learning to solve problems not only in mathematics and science, but in the humanities. Not only are computers likely to improve our educational programs but computer science is already opening up employment possibilities to blind youth at various levels of accomplishment. Special courses for blind students in computer programming are being offered at several American universities. One deaf-blind college student in Ohio is now employed as a programmer at a high level. The computer is taking society into new worlds. We are encouraged in our belief that blind people will share fully in this exciting field of human achievement. Edward J. Waterhouse, Director Mrs. Sally Stuclcey issues instructions on tele- type machine to computer at Massachusetts Institute of Technology five miles away. Michael Silver reads the braille the computer produces.