UniServe*Science News: Newsletter of the Science Software Clearinghouse Vol. 3, March 1996

From the Director

Dry Labs (with a dash of bitters)

Every university in the country is facing pressure to increase teaching efficiency. Our funds are being cut, our student numbers are going up, and we are all being expected to do more administration. If we are to continue to have the money, and more importantly the time, to do our research we must find ways to streamline our teaching, if possible without sacrificing quality. Of all the teaching we do, experimental laboratories are the most expensive. Equipment is costly, staff : student ratios are high, and community groups worried about safety and ethical issues have to be satisfied. Who of us has not toyed with the idea of replacing at least some of our experimental ('wet') laboratory teaching by something more automated?

Bring on 'dry' labs!

There is no question that people can be taught useful skills, and taught them well, by the use of computer simulations. Are not all our airline pilots trained on flight simulators long before they are let anywhere near a real jumbo jet? And have not most of us seen computer packages which train the most ham-fisted student to acquire necessary laboratory skills - like, as just one example, the splendid Virtual Microscope developed by the Open University. And these are getting better and better, as computers increase in speed and graphical power. It won't be long now before we have virtual field trips, designed along the same principles as Doomreg. employing millions of (real) images. Is it any wonder that some departments in this country have already taken the decision to replace some of the costly and time-consuming laboratory experience they previously offered their students, with some kind of IT package? And is it any wonder that even more departments are asking themselves if this is the way they should be going?

But there are worries about this trend. Experiment is, after all, the very keystone of science. That is how scientific knowledge differs from other kinds of human understanding -- all our theories must be tested against what happens in the real world, in real experiments. Therefore if we give physics students an exercise to test whether a simulated radioactive source decays in a random manner, what will they have discovered at the end of it? Certainly not that real quantum events are indeterminate. Or if psychology students learn, by patient keyboard manipulation, to "train" Sniffy the Virtual Rat to perform some action in response to an external stimulus, what exactly has been established? Certainly not that Skinner's theory of learning by operant conditioning can be applied to real rats (let alone people). In either case the student has simply discovered something that the software designer coded into the program. There are many, many academics who harbour the gravest of misgivings about simulated laboratory experiences.

I personally believe that we cannot afford to keep teaching experimental science as we have done in the past to all our students. I also believe that, if we do replace it with something else, we must not throw out the very thing that makes science science. These are issues that need to be brought out into the open and discussed. Therefore UniServe Science will hold a one day workshop devoted entirely to the subject of DRY LABS. It will be held at the University of Sydney on April12. It will bring together academics from all round the country from departments which have successfully replaced some laboratory teaching with different experiences. They will describe how they organized the exercise, what problems they faced and how successful they feel it has been. You can pick their brains and decide if it would work in your department, and maybe even take some software back with you.

So be there on April 12.

Ian Johnston