Society's tendency has, all too often, been to treat handicapped persons as second-class citizens. In particular, blind persons are often relegated to jobs well below their intellectual capacity.

In the mid 60s, the University of Manitoba created a program to teach blind persons how to program in a bid to address this gap. The Canadian National Institute for the Blind expressed interest, agreeing to fund the costs of sending four students, with the university providing teaching and facilities.

The program faced a big challenge - at the time there was no documentation suitable for blind persons, no established way for a blind person to draw or read flow diagrams, and all computer outputs required vision to understand. It was also deemed important for their future employment prospects that students operate independently in the commercial world, so university staff were told to help blind students only in an emergency, to which they agreed.

The pilot course began with an extensive orientation, followed by the instructional material. It was also stressed in the orientation that, hard as it might be, others in the area had been instructed to help only in emergency situations, so students could work independently when they were employed in the sighted world.

Everything about the course had to be innovative. One enterprising, sighted staff member created a method for the computer to print in Braille by placing a rubber band behind the print head so dots caused a readable indentation, adequate for short-term test results.

Another problem were the Flow charts, which described application logic and which typically contain symbols with decision points that can alter the flow when the application is run on the computer. The symbols, of course, were useless. So a technique was developed where only one decision was made on each page of computer code. The page numbers, symbols and computer code would be dot coded so the student could follow through the program logic. It worked out better than expected.

Believing it would be acceptable to employers a target of 10 per cent sighted help was set (to assist in reading output). Not a lot has changed in recent years but the eighties did see the introduction of voice output (text to speech) and, later still, voice input. Today a blind person can get by with much less sighted help.

Initially, only students who already had some academic qualifications were accepted to the program. These students tended to be the most frustrated in the jobs they were able to get, such as serving in canteens or monotonous factory work. They received an immediate morale boost and knew that after completing the course they could make a significant contribution to society. This was particularly true of those who had lost their sight as adults, and whose companies would usually transfer them to positions way under their intellectual capacity.

The memory of blind students seemed to compensate somewhat for their disability, because in developing their computer applications, the students very rarely made mistakes (an obvious bonus for their future employers). In fact, their accuracy led to an unexpected problem. Errors do occur in computer work, and therefore it was necessary to ask the blind students to make programming errors so that they could learn how to handle them in future work.

The courses lasted about four months (leter expanded to ten months) and, because of the very careful selection process by the CNIB, combined with guidance from the university, the success ratio was close to 100 per cent. Close to 250 persons took the course over the 24 years of operation, which closed its doors in 1988.

Next came persuading persuade the IT industry to hire them. In approaching potential employers, it was stressed that the graduates might need 10% sighted help but otherwise they were independent, and that help time would be more than compensated for by their accuracy.

Most potential employers responded by asking, "what if they do not work out?" Our answer: then they don't work out. They are not to be treated differently to other employees. Of the four who took part in the pilot project, three found jobs, while the fourth, Don Keeping, was asked to give future courses, which he did admirably. Keeping continued to run the program for about 15 years, (at which point others took over) with an intake of between eight and 12 students per course. After relinquishing the teaching role he became a regular programmer. One of the initial four eventually worked in Ottawa supervising a team of sighted programmers.

I spoke with Nick Donaldson, who graduated from the course in the 1984/85 session. He indicated that the main emphasis at that time was still on main frames, using slightly adapted terminals for access, with text to voice capability, and with Cobol and PL1 as the main languages. Nick went straight to the course from High School. He himself taught part of the course for a while and is currently employed in the Academic Computing group of the University, as a senior programmer. He indicated that while many of the graduates did move to programming jobs quite a few went in to customer and system support roles. He said the attitude to place blind persons in jobs below their intellectual capacities is still with us.