Wikipedia, Design, and Consensus

Co.Design has a short piece on the attempted redesign of Wikipedia. The design team was open and transparent, produced a handsome proposal, and the reactionaries poured out of the woodwork. In the end, they changed the title font and called it a day. It’s an object lesson in collaborative design work.

I’ve argued before that open and collaborative efforts are not incompatible with good design. I focused on the fact that good design is difficult generally, and that open and closed projects suffer equally. This Wikipedia story has me thinking more specifically about how community norms affect decision-making.

The first comment on the piece is really insightful:

The redesign exposed a common error in open projects: confusing your audience with your contributors. The audience is who will benefit from the design. The contributors help create it. Wikipedians are very solicitous of the contributor community, sometimes at the expense of the millions in the audience. The hermeneutics of the moderation process is one symptom of this, and the abortive redesign is another. This solicitousness isn’t an accident. It’s rooted in a sense of fairness: the contributors put in the effort and so deserve a voice. It’s their reward for showing up.

So: Conway’s Law. Wikipedia isn’t built to produce a well-executed design for an audience, it’s built to produce something the contributors are comfortable with. If you have any doubt about this, wander into the hellmouth of the typography discussion.

Matthew Thomas gets close to explaining what’s happening on that page:

Every contributor to the project tries to take part in the interface design, regardless of how little they know about the subject. And once you have more than one designer, you get inconsistency, both in vision and in detail. The quality of an interface design is inversely proportional to the number of designers.

It’s possible to overcome these structural challenges. Design is, after all, a fundamentally social process. It’s just a matter of finding a collaborative process that’s appropriate for the task: using the same process to filter bias in Wikipedia articles and choose a font face for the homepage is probably a mistake. At work, we’ve successfully used a process called Design Thinking, which encourages collaboration and incorporates the “critical thinking” energy of a group without letting it run wild. I recommend it.

When we think about collaborative design, it’s easy to summon straw man arguments about the virtue of a single designer and the hazards of the crowd. It’s more difficult, and more fruitful, to talk about the biases in the organization itself, and how alternative collaborative processes can help overcome them.