On Designing Interesting Tools

My brother recently introduced me to the Holga, which is a low-end Hong Kong product that emerged in the early 80s to make cameras available to working-class Chinese.

Aesthetically, what’s interesting about the Holga is that its cheap design gives rise to a host of imperfections in the structure of the camera itself. Light periodically leaks in from the outside, switches on the body of the camera are periodically nonfunctional, the camera itself often scratches the negatives as the film unwinds, among other things.

The odd, random distortions that these defects create on the film has cultivated alot of interest among photographers, and a pretty enthusiastic community of artists has developed around it. The defects are unique to each camera, and there’s alot of customization that’s done on them.

The Holga is, of course, not unique at all in this respect. In the world of film, it is cultural cousins with the LOMO, Diana, and the continuing community surrounding the instant-developing Polaroid camera (like the SX-70 and its ilk). In the video world, there’s (most famously), the PXL-2000.

Varied, imperfect tools that generate eccentric, unexpected effects lay the groundwork for nerdiness. (this is just a corollary of other nerd communities, think: wine, music nerd-dom, Magic cards, hardware, etc) They make a tool interesting in the sense that there are many subtleties to be learned, unanticipated effects that cannot be controlled, and that each tool has a unique personality.

Mastery of an interesting tool (as opposed to what you might call a streamlined tool), then, depends on more than just a well-developed technical knowledge about the tool in general. (like you would learning how to use a keyboard) There’s a need to develop a less well-defined familiarity with how your specific tool performs (as in the case of computer owners that learn how to compensate with their machine’s increasingly dysfunctional hardware).

And, unlike a streamlined tool, there is a great deal of randomness that ensures that one never really masters it. To that end, the user’s approach to the interesting device is different: not so much precision control to achieve a particular effect, but rough approximation to achieve a generally expected outcome.

In the case of the Holga, its eccentricity emerges incidentally from the need to keep costs low, but I think an interesting question is whether or not interestingness itself can be the product of intentional design. How one might go about developing tools that are interesting in this way? How do you, in a way, organize for random defect?

There’s a great adage in the CS world called Conway’s Law, stating that:

If you have four groups working on a compiler, you’ll get a 4-pass compiler.

Essentially, that any piece of software reflects the organizational design which produced it. This “law” is applicable to product design more broadly. The iPod, Zune, and Neuros for example, reflect the organizational structure (and, in turn, the values) of the companies that produced them.

Or, more germane to this discussion, a car is designed the way it is because there is a department that handles engines, and another that handles brakes, and another that handles wheels, and so on. How the machine handles as a unit is a function of how well these teams communicate with one another in developing ways for these parts to interface with one another.

The Holga is interesting since these parts in fact don’t interface well with one another. Even the body of the camera, which you think would be easy to assemble as one piece, is consistently flawed. Together, these imperfect interactions create interesting, indeed, pretty cool tools.

I wonder if organizational development can be arranged along the same lines when the aim is to produce interesting tools. It’s interesting because Conway’s Law suggests that one might intentionally design an organization that aims not to share information well interdepartmentally or randomly reveals engineering goals during development. That is, so long as aiming to emphasize randomness as a feature of the product is understood, elements that you might traditionally think as counterproductive might end up producing a more interesting tool.
Or on a smaller-scale product, one might attempt to approach every unit of the tool as a different product, redesigning from the ground up. Eccentricities would evolve, then, as a result of team members actively trying new things with each “release” of the item. For example, interestingness is a feature of most custom-built cars, or custom-created software.

There is, of course, the fine line between being interesting and being entirely dysfunctional. (No one would say that the Ford Pinto, for example, would be interesting to use, merely dangerous) But, organizationally, I think there are ways of mitigating this tendency to slip into just creating a bad product: establishing the goal of the product as interestingness, for example, or consistently testing the ability for the tool to achieve its purpose. An interesting product is forever elusive in performance, but never truly frusturating in the sense of not functioning at all.

Just a thought.


1 Response so far »

  1. 1

    […] Conway’s Law? — a group’s product is inevitably a reflection of the organizational structure that […]

Comment RSS · TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: