As an addendum to Dom’s previous post on conversation, something we noticed repeatedly during last weekend’s brainstorming session was our attempts to define the tool’s functionality through interface design. This is a natural way to approach software development (”When I sit down to use it, what can it do?”), especially since we’re both designers, but we had to pull back from this and take a higher-level approach.

We have already decided that we want the tool to be adaptive and it’s function to be emergent. This means consciously avoiding anything that might contribute to a sense of determinism in the tool.

One of the basic tenets of our approach is faith in the wisdom of crowds; that the collective users who employ the tool will be smarter than us, and that through using it, they will be able to suggest it’s function better than we would ever be able to enforce it.

We discussed how for any new user, the function of the tool will be implied by the existing content. As such, should we include “hard” GIS data (which we have access to from the Wildlife Trust) as a default bed of information that users can build upon? If we do, will they be overly-influenced by the nature of this data and hold back on contributing “softer” data, like stories and personal experiences?

There’s probably a fine balance to be achieved, but there’s no way right now of us knowing what that might be. Rather than make a judgement call that might influence use of the tool, we’ve decided to hold back. Like Dom said, unless we can provide a strong argument for including something, it doesn’t make the initial release. With a small initial feature list, we’re agile and adaptive. By holding back, we’ve got our ‘reserve of ideas’ ready as potential responses to feedback we get, and the development of the tool becomes a conversation with our users.

I think this is a good example of listening to the underlying principles of the project, and allowing the interface to be informed by philosophy.

Perhaps this is more of a philosophical debate than something that will change how the tool is built, but I think it’s an interesting topic, and one that we’ll really only be able to discuss beyond hypothesis once the users are in there and the whole gig starts rolling.