The Hundred-Year Chair · a chapter

A Chair Without Me

The cherry chair had gone together without fitting, but I was the person who cut and assembled it. Someone else would bring different habits, tools, and assumptions to the same drawings. The next test should put those drawings on another bench.

Tim Gray, my longtime editor, raised the practical question while replying to a request about the prose. He wondered whether he could build the chair with what he called carpentry skills one step up from Lincoln Logs. I liked the idea of giving him a design to try. An interest expressed in an email is not an agreed build.

Repair would ask a similar question. If a joint cracked or a panel split, a future owner would need to know how the chair came apart and how to make the failed part. A photograph of the finished chair would not necessarily tell them.

I want the knowledge to travel in more than one way: in the printed instructions, in a construction someone can inspect, and in people who have made or repaired the chair.

For a builder such as Tim, I need to separate what a particular build requires from what I recommend or leave open. A dimension can be required for two parts to fit without being the best dimension for every chair.

The habit of ranking the parts of a pattern by how sure of them you are I owe to Christopher Alexander. His A Pattern Language marks each pattern with the confidence that it names a real invariant.

Within the named Keel family, I treat some of the arrangement as fixed: the two triangles and the shared keel. That is a boundary I choose for the family, not a discovery that other arrangements are wrong. The Panel and Truss versions still need their own dimensions and construction details.

Some choices are recommended, with the reason and the limits of my experience attached. For the Keel Panel, I recommend the dining posture I began with, informed by the literature and the people who have tried it. That is not evidence that it suits every body or use. I also believe that fitting parts by scribing can be quicker than transferring measurements, but I have not put a stopwatch to the comparison. A recommendation should say what I tried, what I observed, and what remains a belief.

Other choices are flexible when developing another version: species, thickness, finish, color, fasteners, the shape of the bracket, how the ends are cut. Every one of those has already changed at least once. That history is not permission to substitute parts indiscriminately in a checked recipe. A new version needs its own checks.

The useful distinction for a builder is between a dated recipe and the wider design idea. The recipe needs an unambiguous account of the parts and connections that were checked together. The wider idea may suggest another chair, whose usefulness and safety remain to be established.

Making the drawings useful is one part of letting the chair out of my shop. Being willing to show an unfinished design is another. I remember an encounter in which secrecy came first.

When I was a new assistant professor, my senior colleague Dave Staelin approached me with an idea for revolutionizing personal transportation. He asked me to sign a nondisclosure agreement. Billions of people, he explained, needed a better way to travel a kilometer or two. I was already passionately interested in personal transportation and waited eagerly for the solution.

Dave smiled and whispered, “rollerskates.” He produced a prototype that looked like small electric Formula One cars for the feet, with a power cord leading to a battery worn in a backpack. My enthusiasm ebbed. I pictured people traveling through city streets at thirty kilometers an hour on fifty-millimeter wheels, balancing and steering with their feet. The need was real. I doubted that these skates were the answer.

The venture went ahead. After a year on the skates, the company turned to a lightweight electric scooter, which struck me as a much better idea.

I think about that encounter when deciding what to show of my own work. Dave had asked for confidentiality before revealing the design; once I saw it, my concern was whether it would work for the people he hoped to serve. For my chair, too, I have to ask which risk deserves more attention: that someone copies the design, or that I get the design wrong. Showing it to other people gives me a chance to discover the latter. Their indifference might be more useful than their promise to keep a secret.

I want the finished pattern published in full, free for anyone to build, sell, or improve without asking me. A drawing in my drawer depends on one custodian. Instructions that people use can be copied, corrected, and taught. Rietveld's sheet had reached my bench; I still needed to find out what someone could make from mine.

Giving away the drawings would not remove my wish for recognition. I would like people to know where the design came from. I would also like them to make chairs. Those wishes can diverge. Someone might admire a beautifully presented drawing and never build from it. Someone else might make a useful variation, pass it along, and lose my name in the process.

It is easy to applaud that second possibility at a distance. It is harder to know how I would feel on seeing a version I disliked offered for sale by someone who did not mention me. I have not had that experience with this chair. I can decide what freedoms to offer without claiming that every use of them would please me. Nor does offering a design guarantee that anybody wants it.

There is practical work between a generous intention and useful access. A prospective builder needs to know which files belong together, which instructions are current, and whether a change is a correction or a new variant. Someone making a modification needs a way to distinguish it from the version I built. An attractive photograph with a download link could invite participation while leaving most of that work undone.

I would also need to make the permissions explicit. Saying here that people may build, sell, or improve the chair is my intention for the finished pattern, not a claim that every photograph, borrowed drawing, or piece of software associated with this book is mine to license. The project's own drawings and instructions need clear terms. Material supplied by others needs to retain its own attribution and permissions.

I have not settled all of those terms. Tim's question gives me something more immediate to work on: can he find what is needed and understand it? Recognition might bring someone to the drawings. After that, the usefulness of the drawings has to carry the work.

As of September 29, I still need to choose the design and instruction version, ask whether Tim wants to do it, and prepare a dated packet. If he agrees, the useful record will include what he understands unaided, where he has questions, what I explain, and what has to change.

Even a successful handoff would be evidence about one maker using one version of the instructions. It would help me find what I had failed to communicate. It would not certify the structure or tell me how long someone would keep the chair.

The finished appendix should be a dated recipe for a particular build. The website can hold other versions, but a change online should not silently change what the book tells someone to make.

I was also pursuing a much less testable question: what a thousand-year horizon would ask of a different chair.

A chapter of The Hundred-Year Chair, drafted and refined in public. The same text lives in the book; the interactive tools and the fully-sourced technical notes live here. home · the book · the technical notes · hundredyearchair.com