If all our reasoning is based on metaphors (as argued by Lakoff and Johnson in Metaphors we Live By) then we system designers might as well get good at choosing them.
I find most single grand-sweeping metaphors to be corny. They rarely embody insight. How can we do better? I recommend this process:
1. have insight
2. capture it in metaphor
3. repeat
Let me elaborate.
A quick launch into programming will surface the need for insight. Use it. Stay attune to the slightest difficulty. Study it. Where have you seen it before? Where has it been avoided? Why? Look for the forces and their resolution. Patterns. Reflect them back on your current effort. What thinking needs to change? When you find you get to the same place through multiple paths, then you are on to an insight.
You will want to share your insight without asking others to do so much thinking. Can you summarize your insight as "like" something else? Something familiar? Or better, something from an area you already tap for metaphors? Remember, you're not looking for one metaphor, you are constructing a productive system of reinforcing metaphors. When you do this, over time, you can truthfully say you helped a productive architecture emerge. You've used your collective minds well.
Be careful. You may find that the metaphors you want to use clash with the metaphoric system (or systems) already in place. This means you have more work to do. Analyze the existing metaphors. Understand what they have contributed. Was it good? Does it have limits? Are you at a limit now? Even worse, could it be that your own thinking is caught up in obsolete metaphors? Are you having trouble finding insight in the work you are doing? Do the metaphors you already use blind you to the insight you need? Are you pushing ahead anyway? There be dragons.
I offer as example my life long struggle with weight. I've made some progress by reexamining many other things I do. I gave an ignite talk (5 minutes) on the experience without once mentioning weight or weight loss. There is video online. The talk is dripping with metaphors, many visual, that I hoped to slip past my audience's conscious mind. A recurring theme is: you think X is like Y, but it's really the opposite, and your going to have to struggle to get that opposite into your head.
http://c2.com/~ward/ip7/
Best regards. -- Ward
__________________
Ward Cunningham
503-432-5682
On Feb 19, 2011, at 1:20 AM, wil.pannell wrote:
> Metaphor has been the least tangible of the concepts deemed important
> in software development. Most discussions around metaphor detail the
> effect that a good one has on development. And only the most
> experienced practitioners (Ward, Joshua, Kent, et.al.) can point to
> effective ones. Does it follow that software metaphors are not easy
> to discover, and given the importance placed on them, the discovery of
> bad ones can do more harm than good?
> For those who can point to effective metaphors in their practice, I
> ask: what was the inspiration? What allowed you to come up with the
> metaphor for the product?
