Prototyping: A Design Thinking Skill

Last night, parent Dan S. and I built a couple of tables for the New Design Lab project that our head of school announced a few weeks ago.  The lumber had been sitting in the room for a couple of weeks, and we were both eager to get these structures done. 

It's worth mentioning that Dan had a lot more carpentry experience than I've had.  I've built a couple of Adirondack chairs, and assembled a worktable very much like this one.  He's built some cabinets, some furniture, some other projects, and more.  He's also got a much larger tool-set than I have. I have an impact drill and a circular saw, a set square and a long level; in addition to that, he's got a mitre saw and some other quite useful tools for cutting and shaping wood.  Even so, we'd build a couple of things like this before, and how much trouble could it be?

Nonetheless, there were a couple of times when we stopped working, stared at the parts we'd cut and the screws and drills and measuring tools lying about in the sawdust. Anyone looking through the windows would have assumed we were a couple of dummies who'd just made a major mistake.  We weren't dummies, though. We were thinking.  Hard.

The IDS Design Thinking Manifesto assigns the heptagon to the mental mindset of prototyping.  The seven-sided polygon is one of the most difficult shapes to create, geometrically.  It requires an estimation technique called neusis to complete; and if the geometer doesn't estimate correctly, the heptagon looks lopsided and weird.  We were going through our own process of neusis as we built those two tables — using straight edges, right angles, levels, and other tools to confirm that the frames of our tables were built at right angles to each other and to the floor; that the table tops would be level; that the legs would all rest on the floor and there wouldn't be any wobble (or at least, not very much).  

And that's prototyping in a nutshell, really.  We built two tables in three hours, and under normal circumstances you'd assume that meant an hour-and-a-half per table.  But really, it was two and a quarter hours for one table, and forty-five minutes for the second.   A lot of that extra time, we were looking at pictures and reading explanatory text in the manual; neither of us had built a table quite like this before; and just because it was going to be for a workshop space was no reason not to do it right.

Even so, we still got it wrong.  

As we prepared to cut the plywood panel down from 4x8 feet to the four tabletops and lower shelves of the two tables, we discovered that the frame of our first table was not square.  Could we fix it by hammering on the frame? No.  Could we twist the frame by hand into the right shape? Sort of, but not really. How was this going to affect our final result?  Should we unscrew the frame, and try again? 

All at once, a simple solution occurred to us. Maybe we should measure the frame and see how much the actual construction was off from the intended design.  The answer surprised both of us — a little less than 1/8".  We could see the distortion in the frame when compared with a totally square piece of plywood; but once the plywood was cut, the distortion would be invisible. Our obsession with neusis — close, deliberate, estimation — had paid off.

We cut the plywood.  As I did so, the circular saw drifted offline, and I wound up cutting a deep gouge into the sheet of plywood — the stuff we needed for the other table!  Oh no!  We stood there for a good long while, trying to figure out how to solve the problem; Dan said a couple of placating words, but you could tell he was disappointed in me for screwing up.

We stood there for a while trying to figure out how to work around the mistake.  My mistake.

Then we shrugged, and moved on.  We re-allocated how the plywood sheet would be cut. This time, Dan would cut the plywood, and I would manage the process of weighting down the sheet and holding it stable for the saw.  Again, during the main cut, the circular saw drifted off the true line, and this time Dan cut a bit off that 'ruined' the sheet of plywood.  

Cutting plywood is a maddening experience, you may guess.

We re-re-allocated the remaining plywood, shifting those two remaining panels from being the tops of tables to being the lower shelves; and reserving our two original pieces (which looked BEAUTIFUL, by the way) for the tops of the tables.  We decided to save the second sheet of plywood for another purpose, which I'll describe at another time.  

In the midst of this process, Dan asked me if we should assemble the other frame to the point that we'd reached with the first frame.  I dithered for a while, I admit.  It would be nice to have two tables almost finished.  

At the last moment, though, my eye landed on a paper along one of the windows, a quotation by former NASA engineer Scott Billups.  Scott is, in many ways, a 'nobody'.  You won't find a celebratory biography of him on a website, nor a Wikipedia entry, nor anything of that sort; you might not even find his Facebook page.  But something he said to me resonated, and it's a rule I've tried to adopt in our Design Lab ever since: "Build the whole prototype, well past the first mistake. That way, you'll discover ALL of the mistakes, and learn how to avoid them."  

I told Dan, "Let's finish this table.  And figure out what we did wrong by comparing with the manual."

So we did. We finished the first table.  Then we compared it with our original build instructions.  Along the way, we discovered that all of our original mistakes, that led to the slight twist in our first frame, were the result of not following the directions.  We followed the directions the second time through; and behold!  The second table took less time, and was more exactingly constructed — and even so, it's the table with more wobble in it.  

How did that happen?

Well, it turns out that materials aren't always perfect.  A plywood board, or any wood at all, can have a bit of warp in it. The 'perfect' materials that are found in instruction guides are rarely found at your local hardware store.  You have to learn to adjust, and move on.

That's prototyping in action.

Your Kids and Prototyping

Children go through the same process that Dan and I went through while building these tables.  There are times when they stand around, looking at what they've built. There are times when they have to look at the materials and equipment and tools, and times when they have to stop and ask questions. There are times when the assembly process hasn't gone according to plan.  There are times when the work has to be completely taken apart, and re-constructed. 

This is normal.

It's not failure to go back and correct your mistakes.  Even with the most careful estimation, the most careful measurement, mistakes still happen.  Your frame is going to be off by an eighth of an inch.  Sometimes that miniscule distance matters a lot. Sometimes it doesn't matter at all — but you won't know until you finish building your prototype.  

It's important to keep going.  Even when you get stuck, even when the effort seems to be a failure.  Build the whole prototype, and learn from the mistakes, and then figure out which mistakes matter and need to be corrected; and which ones don't.

That's the lesson we're trying to impart when we teach Design Thinking to you and to your children — that it's hard to tell which mistakes matter until the work is done.  And once the work is done, it's usually clear that the mistakes are part and parcel of the success.