The heart of the discerning acquires knowledge,
Proverbs 18:15 (NIV)
for the ears of the wise seek it out.
My return to Bradford following the year working for ICL in Wakefield began with me, for the first term, renting a room near Lister Park, about 2 miles away from the main campus. For that first term I got myself a bicycle and as a result kept quite fit. Bradford isn’t known for it’s cycling community by virtue of the fact that like Rome, it’s built on 7 hills. It was quite scary because it seemed like most of the local minicab drivers had never seen a pushbike before and they didn’t understand that you are supposed to give way to them, rather than kill them.
I found that The University had got rid of the old ICL 1904 mainframe (by then it had become seriously outdated) and replaced it with a Control Data Corporation (CDC) “cyber” series machine. (CDC was one of the main American mainframe computer manufacturers of the era). The new system had on-line computer terminals all over the campus, rather than just in the computer centre, and it was a tool for all the departments of the University. For those of us coming back from placement years, it meant that we had a new system to learn. It was quirky but reliable and fast. I remember seeing the operators playing something looking like “Space Invaders” on the central operator console.
What was unusual about that first term was that, due to the way the subjects had been scheduled over the year, most of my choices didn’t start until term 2 and I had practically no lectures. I therefore used the time to get ahead in my final year project.
We had been given a number of suggestions about the subject of a project. The one I chose interested me because it was in the field of Artificial Intelligence (AI) and the opportunity to do some research in that field was too good to miss. You may have heard of the “Turin Test” proposed by the brilliant mathematician and computer scientist, Alan Turin. Very simplistically, the idea was that a human being (the observer) could have a conversation (by exchanging text only rather than voice) with another player (which could be a human or a computer) in a second room. If the observer could not reliably tell whether the player in the second room was human or computer, then the test is “passed” and the computer deemed to have some kind of Artificial Intelligence. I don’t think this test is very helpful at divining true intelligence. For example, back in 1966 , Joseph Weizenbaum created a program called ELIZA, which would analyse key words in the sentence submitted by the observer to create an apparently inciteful response. ELIZA really fooled some observers into thinking it was a human but of course there was no actual understanding by the computer of what was being said.
My brief was to take this a little further and create a natural language interface into some real-world data. So, a person could ask a question of the computer about the data in plain English and the computer would then provide the answer, again, in plain English. This means that the computer is “intelligent” in that it knows the data and can answer questions about it but it can only be so in a very restricted field. So, if you asked a question about something else it would not understand and it also would not be able to comprehend the relationship of the data to the “real “world. You might think this was limited, but remember it was 1982 and there is not much fundamental difference to what I was doing to modern day Alexa or Siri! (Save their ability to recognise speech).
The solution I developed was to break down the problem into two parts. Decoding the syntax of what was being said and then, using the knowledge of the syntax, relate this to meaning. That is what the computer was being asked to do, the symantics. The result was two specialist computer languages; one to cover the syntax of the English Language (or at least, a small part of it) and a language of semantics telling the computer how to answer the query.
One day I might see if they still have a copy of my paper in the archive.
The lop-sided timetable over my final year enabled me to break the back of my dissertation before the second term and I could then use the last term to concentrate on what I needed to know for the final year exams. I had moved into halls of residence just about ½ mile from the main campus which was very convenient. It had a half-board arrangement which included unlimited free food for breakfast and dinner, which was extremely welcome.
My dissertation was very well received and helped me achieve a First-Class degree. One or two others in my year did the same and significantly, this included one of our overseas students. On our course we had a number of students from the far east. Although we collectively referred to them as “The Chinese”, I think most were from Malaysia, Taiwan, South Korea and Hong Kong (this was before Hong Kong was ceded back to China). Language was a great difficulty for them and those of us who had helped them over the years were rewarded with a fantastic authentic Tai meal! Unlike us English students, these overseas visitors had to pay fees for their course. This, and the fact they had to spend so much time on the other side of the world from their home, in a foreign land, is testimony to that fact that we had received, over the years at Bradford, a truly world class education.
I received a fantastic grounding in so many subjects: statistics, economics, business accounting, databases, networking and not forgetting the ability to touch-type.
So there ended my university education, and an amazing character forming time at Bradford. The faculty offered me the opportunity of continuing as a PHD research student. The area of research they offered didn’t fill me with excitement in the way working for ICL did and so I made the decision to go back to ICL at Wakefield and continue a career there.
Richard Mann 