For a long time now, educationalists have been keen to harness the attraction that computer games hold for many of us. “People spend hours and hours of their own time playing games!”, say teachers, “If only they showed that sort of interest in my lessons!”.

Off the shelf games are already being used, in an ad hoc fashion, in many classrooms, to support existing teaching, and provide background, motivation, and opportunities for debate. Using an existing game as a tool is a great solution, no different from using any other physical object or fun activity as part of learning.

However, there have also been many attempts to create “educational” games. Most of these have been misguided in my opinion, and when I hear people from the educational world getting excited about games, I start worrying that it’s the precursor to another attempt to make a great educational game. My response to this would be: “No! Don’t do it!”.

The idea that one can harness the power of games as some sort of magic bullet to encourage learning is just as valid (and invalid) as the idea of using anything else that people love; music, say, or sport, or chocolate!

One might well be able to base a lesson plan around eating chocolate - it would have a certain novelty value and it would probably be pretty popular with the students, but only if it was nice chocolate! The same is true with games - the key, always, is that you must start with a compelling experience - something that people want to do.

Making a good game is a process with just as many nebulous, uncertain variables as writing a novel or writing a symphony. To make a great game you need a good idea, some great hooks, a slice of luck, and a team of expert programmers, designers, musicians and artists who are focussed on making something cool.

I don’t believe that many developers of successful games think in terms of their “audience”. Most will be in their twenties and thirties, but they are still likely to share the same taste in games as their users (who are probably a lot younger). The developers are just trying to create the game that they want to play.

By adding an extra educational objective to a game, the link between developer and consumer is often lost. Instead of asking themselves, “what would be cool here”, a developer is suddenly forced to ask “how do I make latin verbs fun?”, or some other equally inappropriate question.

The process of development changes from something driven by gut instinct, emotional responses and the subconscious, into something conscious and intellectual. The motivation is almost inevitably weakened as a result, and the game suffers.

I’m not saying that a good educational game is impossible, simply that the barriers to making an engaging game are already high. To make a good game is hard enough already, and adding a set of educational objectives make things worse not better.

Immersive, engaging, fun educational software is certainly being written, and sometimes it provides an experience hard to distinguish from a game. Crucially though, the stuff that works best is generally designed first and foremost to be good educational software, and not a leading edge game.

I absolutely agree that the games and learning industries should be talking to each other, but hiring game developers and asking them to make something educational is probably misguided.

Instead, we should be encouraging developers to just keep making great games, but to make them more adaptable, scriptable, and extensible. They should be encouraged to build their games as frameworks and shells that can be taken by educators and re-used in different contexts.

Good developers do this anyway, as it has a number of advantages, usually improving the robustness of the code, increases the speed with which the developer can generate their own content, and helps to promote a whole community of add-ons and customisation around the game which extend its life.

Not all game developers are great engineers though, and even the good ones have to move fast, and meet deadlines. As a result their tools aren’t always up to scratch, and are almost always not ready for release to the wider world.

If there is any government money available for “games and learning”, then this is where I think it should be focussed. We should pick a few great games which are already designed with extensibility in mind, and spend some resource helping to clean up and document their tools for use by educators. This would allow the educational community to benefit from games technology without having to try to recreate it, and it would offer a clear benefit to the developers too, who would find a way to spend the time on their tools that they know they ought to, and often want to, but generally can’t justify.