Over engineering learning computer solutions?

I would guess that there are still software engineers and developers who harbour the ambition of engineering the human out of more and more processes. You often hear the retorts that a process would work really well, if only people were not involved. Doubtless there are people too, who would like to be engineered out of particularly tasks and procedures to a point where everything is automatic. But is that really a desirable objective?



The car is now massively advanced from the days when cars had to be cranked up to start by hand. You could imagine we will get to the point where it would be possible to point our car at a destination and, with a combination of satellite navigation and automatic controls, get to where we want to go almost without thinking. Is that really what we want? Isn't part of the enjoyment of a car journey and what captures our interest having to make decisions.

Making decisions can therefore be good and it is important to remember this when engineering computer-based learning solutions. Take e-portfolios for NVQs as one example. There are some systems that in essence make the decision for the assessor, if you have this and this, the following is automatically satisfied. There are others which require an assessor to make a genuine judgement. From a learning pedagogical perspective and in terms of motivating the assessor and learner the latter approach must be better. Providing helpful tools and guides is good, as is removing unnecessary paperwork but if we are over-automating any learning process we are in real danger of turning off all those involved in the learning journey. 

Part of the creativity in developing computer software for learning lies in managing that difficult balance, between automating processes to reduce what the learner has to do but not to the extent where the learner becomes disengaged from the whole process.