Tech Lunch

At Leeds on Friday (see yesterday’s entry for more background) the main keynote speech was delivered by Dr Andrew Herbert, Director of Microsoft Research, Cambridge, who was a student at Leeds in the early 1970s. His talk was entitled, “Why everything I learned at Leeds in 1972 is no longer true” and he declared it to be, in part, an opportunity to have his revenge on his lecturers of thirty years ago. Continuing this blog’s interest in all things lunch oriented, Andrew noted that revenge is a dish best eaten cold, and that in this case, given the thirty-five year wait, it was “glacial”.

His main point was that there have been enormous technological changes and much of what was being taught in the ’70s was focused on “overcoming the limitation of the machine” e.g. speed of processor, size of memory, poor quality displays, use of punch cards etc. Some of Dr Herbert’s points were more obvious, for example: the day of the stand-alone computer is over, we live in the age of the networked computer. Secondly, software is much more complex than it used to be: testing code by simple ‘desk’ review (people who work in publishing can equate this to being a bit like proofreading) is no longer sufficient. Also, we need new ways of organising and visualising the large amounts of information on our systems – hierarchical file systems have had their day.

Other things were more complex. For example, the questions affecting Artificial Intelligence research have changed – the issue now is how far are we prepared to put our faith in computer algorithms that demonstrate ‘intelligence’. What if computers can’t explain WHY they’ve made a decision?

He also made the point that we are still in the process of defining the subject of computer science and there is a need to make it clear to people today that there is a definite discipline of computing and it is not just a group of techies playing around with gadgets. He said: “we have some way to go in persuading other disciplines that we have a theoretical underpinning”. This was very interesting to me as I feel quite strongly that we need to keep the ‘science’ in computer science. And when I say that, I mean we need to keep hold of the importance of ideas and the spirit of exploration and investigation, rather than turning into a purely vocational course of study.

I spent an interesting day last week in Birmingham (or Brum as it is affectionately known) at a workshop on Business Process Modelling given by Balbir Barn of Thames Valley University. Birmingham has certainly changed since my youth there and the workshop was held in the completely redeveloped Brindley Place area of the city. This used to be a decaying network of stinking canals, collapsing Victorian warehouses and rat-infested walkways. It is now home to flash hotels, bars, offices and a series of conference venues including Austin Court, where the workshop took place. Lunch included chocolate covered strawberries which were extremely tasty, although probably ethically dubious given that this is March.

The workshop itself covered Business Process Modelling Notation (BPMN) a diagrammatic notation for representing workflows in the business environment. This probably sounds fairly dull, but it is quite interesting in that it was aimed at higher education and is a sign that universities are becoming more aware of the methods used in the commercial sector. Towards the end of the day, one subject of debate was the likely uptake and impact of using such workflow tools in higher education settings. It can certainly be argued that there are parts of the university system that are akin to the bureaucratic functions of a business (HR, payroll, student registration, course validation). But what of more non-traditional areas like library repositories or e-learning systems? Delegates were certainly interested in debating the potential return-on-investment for groups of developers within the education community who have spent time learning and mastering these kinds of workflow tools.