I came across this paper from Waller Hart recently, and I quite like it. It draws a parallel between what architects do and how the authors intend to carry out learning design processes. But I don’t think it takes the parallels quite deep enough.

In spite of the fairly widespread use of architectural terminology in the e-learning industry – we often talk of blueprints, learning architectures and so on – I’ve not seen a particularly clear discussion of some of the more important ways in which our practice as learning designers is similar to, or could learn from, architectural practice. So here’s a quick attempt.

Think “experience in an environment” not “content in a course”. An effective piece of architecture is about the experience that people have within it. It’s not about walls, doors and windows. Similarly, an effective piece of learning design is about the experience that people have, not the “content”.

I'm sorry if that's painfully obvious but my experience is that many in the learning design business seem to find this approach difficult or threatening. Why? Firstly, there seems to be an assumption that the link between what we design and what people learn is a strong one. But it isn’t. Like architects, we create an opportunity for people to have an experience; if they choose not to behave as we want them to, there’s not much we can do about it. Our designs guide, but cannot determine, their experience. This lack of direct control seems to worry those from traditional instructional design backgrounds. Secondly, on the subject of personal backgrounds, many learning designers have arrived in the industry as writers, (and technical writers at that) not designers. They prioritise words, and the conveyance of information, not the creation of experiences. Finally, and most importantly, common conceptions of knowledge and learning amongst traditional learning designers (at least in the UK ) appear to me to be pretty crude. I’m not going to go on about this much here; my views on this are spread around this blog and this site.

Something I’ve always loved about how (good) architects think is the way they’re driven by the experience that is likely to result from the spaces they create. Where I’ve worked with learning designers to cultivate this mind-set, the results have generally been surprising and positive.

Humanise re-use. I’m not hearing too much any more about the crude, mechanistic view of content re-use that emerged from the learning object movement of a few years ago. We’ve come to recognise that the Lego view of the world was never going to work. But sadly, I still find many learning designers and companies who have a simplistic approach to templating as a form of re-use. Most template systems I’ve come across are functionally crude and pedagogically ropey, resulting in neither the cost savings they’re intended to deliver, nor effective learning experiences. Have a look at my posting here if you want to read more about this.

One approach that’s been knocking around for many years is the use of “patterns”, a term originally coined by architectural theorist Christopher Alexander. His approach is based on the premise that it’s possible to look at the kind of challenges architects face as collections of familiar, repeated problems assembled in different ways. For example, when you’re designing an entrance to a building there are certain things you always need to consider: its size, aspect, materials to use and so on. Interestingly, Alexander’s ideas don’t seem to be highly regarded among architects these days (see Bryan Lawson, in “How designers think”). But they’ve been widely adopted by software designers and developers – particularly those applying agile (extreme, SCRUM…) methods. Talking to software developers who use such approaches the main thing that comes across is that re-use is essentially a human process, not an automated one. It’s about understanding ourselves as designers, and mining our professional experience. It’s not about menus of prefabs to be snapped together without thought. Of course, in an increasingly complex world, it’s really helpful to have menus of suggestions, models and tools to prompt thinking, but this is very different from the painting by numbers approach adopted by so many e-learning content providers.

When I’ve worked with e-learning designers and developers to cultivate a pattern-based approach to re-use, we’ve generally found that discarding mechanistic approaches is not only more fun for designers and appealing to clients; it’s also genuinely cheaper and more efficient.

Design in layers. What (good) architects do supremely well is think and design in layers. They look at the big picture – and the medium-sized picture…and the little picture…and the detail. There are lots of rather obscure academic models of this (for a simple one, have a look at Brand's model on page 14 of this Word doc), but the main point is this: however you define the layers, they have to make sense with each other. So in Brand’s model, the building’s "skin" must work well with the "site" and the "structure". All layers have to work towards solving the design challenge the architect has been set. Likewise, in designing a learning experience, the designer must consider layers such as the physical environment the learner is working in, their broader IT system, the LMS/PLE/VLE they’re working through, individual elements of the learning system, the interfaces of each component, the content within each…and so on, ever inwards towards the detail.

I’ve found that thinking in terms of architectural layers really helps to ensure that whatever the learning solution is, it makes sense to the learner. That is, it works.

Blur the design process. Thankfully, we’re coming to realise that traditional, linear instructional design processes were only really relevant for the environments they were originally designed for (i.e. the military – although my experience suggests that even there they were not always carried out particularly effectively).

Indeed I’d hardly call traditional ISD "design" at all. It’s more akin to low-level engineering than design in that it assumes that the designer can get a firm grasp of the learning problem early on in the process, and build from that towards a clearly definable solution in a series of linear, discrete stages.

Real designers – particularly architects – don’t work like that. They explore problems through the designs they create. They mess about, iterate, experiment and develop their understanding of the problem through what emerges. Indeed Architectural writer Bryan Lawson suggests that the three traditional stages of design – analysis, synthesis, evaluation – are in many cases almost entirely blurred together. Innovation companies like IDEO think the same. And many would argue (have a look at Michael Schrage’s writing on this) that you start with evaluation, for example, of a crude prototype, synthesise your ideas from this evaluation, spin round that cycle a number of times and out pops your analysis.

The problem with this approach for many in the learning design world is that it’s an uncertain process. It’s not the predictable, highly-structured, engineering-like approach so beloved of traditional instructional designers. And the main challenge I’ve found in trying to cultivate more genuinely design-like behaviour amongst learning designers is that they don’t handle uncertainty particularly well. This is, to say the least, a bit of a hindrance in such an ever-changing, uncertain world.

One architect I interviewed last year as part of an investigation into design practice told me: “if you think you know where your design is heading when you start the process, the only place you’re heading is disaster”.

* * *

A final word. Sometimes when I go on about "blurring the design process" or "humanising re-use" I get asked whether I’m some kind of new-age innocent. Ok – so compare the day-rates and profit margins of e-learning providers with those of architects and professional designers. I reckon that thinking and working like architects is one way we’ll increase the rather lowly status of learning designers. Oh – and maybe help more people to learn better.