This may come as a shock, but there are no software developers in San Francisco. They have gone the way of the honeybees and vanished. What we have instead — in spades — are engineers.
If you program computers you are not a computer programmer. If you develop websites you are not a web developer. If you code you are not a coder. We are all of us engineers.
My fellow engineers, what in the actual what? Who do we think we’re fooling? The university I went to doesn’t even have an engineering school. The only thing I’ve ever fabricated is my job title. I’m no more a software engineer than an author is a book engineer. Plus, hardly anything about my work is physical. There are dozens (hundreds?) of computers running my code and I don’t even know where they are! (Somewhere in Virginia, probably!) Do we think this is a problem that other engineers are having? “I just built a lovely bridge,” says the civil engineer, “but it’s the damnedest thing, I have no idea where it went!”
I find our engineering obsession peculiar. What’s behind this job title shift? Why do we software developers now insist on calling ourselves engineers?
I have a theory. I think it’s because we love what the word “engineer” has come to mean. To be a software developer is to wear jeans with sneakers and spend your days behind a computer. To be an engineer is to build things. Which would you rather be?
Now, don’t get me wrong. Writing software is a wonderful challenge. You get to do cool stuff like design systems and command fleets of computers.
But sometimes writing software isn’t enough. Sometimes we want to build things, real things, with our hands. I once met a software engineer who took apart his entire car just to see if he could put it back together (he could) and who dedicated a room in his house to an industrial-sized lathe to experiment with replaceable parts (don’t ask). I feel myself being pulled in a similar direction. Last year I rented a textbook titled The Fundamentals of Building Construction. I spent a recent Friday night gluing pieces of wood together. I am beginning to take my job title literally.
Across San Francisco, developers are turning into engineers — and some not only in name. This essay is about what happened when I decided to put down my mouse and pick up a hacksaw. It is the story of three sleepless days in which I 3D-printed, sanded, lacquered, measured, cut, laser-etched, painted, and glued my way to engineering glory. Proceed with caution, my friends. This is a hard-hat area.
First, some background. I work at Plaid, a technology company that connects software developers — err, engineers — to banks. A love of building is central to what we do. We build stuff so that other engineers can build cooler stuff on top of our stuff. We have a strong love for building, which we show in many ways.
Take, for instance, our latest engineering team event, which was Lego-themed. By this I mean we bought a bunch of Legos and then sat around playing with them.
Or consider the Plammer, a giant sledgehammer and legal liability. If one of our engineers does something heroic like rewrite a complicated tangle of code or fix a bad bug, we bestow upon them the Plammer. “Plammer” is short for “Plaid hammer,” because at Plaid everything is a portmanteau — even “plortmanteau.”
Our altar to engineering is our workshop, which somehow avoided being plortmanteau’d into “plorkshop” and is instead called “Tartan Manufacturing.” Our head designer, Nick, may be more of an engineer than any of us engineers, for it was he who dreamed up the workshop and brought it to life. It exists despite the fact that Plaid has no business owning power tools. For goodness sakes, the workshop has a working garage door — and knives! How is legal OK with this?
Perhaps the only thing more surprising than the workshop’s existence is the fact that it gets used. It’s proven to be a great space for drawing, writing, and tinkering. People dip in and out of the workshop at all hours. I spend my free time there playing with an awesome tool: a 3D printer.
The 3D printer needs explaining. A few months ago a fellow Plaid engineer and I got to talking about Rube-Goldberg machines. We decided that it’d be fun to build one together, and soon we were talking about designing and printing parts with a 3D printer, and pretty soon after that there was a 3D printer in the mail room. I bought it with my own money — 3D printers have greatly come down in cost; mine was $400 — and decided to keep it in the workshop. That way other Plaids could play with it. Plus, I could slack off by printing stuff.
Legos, the Plammer, and our workshop testify to Plaid’s love of building, but the ultimate example of our obsession is Plaiderdays, our thrice-annual three-day hackathon. Like many tech companies we put on hackathons to give ourselves the opportunity to work with new people on interesting projects. Hackathons are typically unstructured free-for-alls and at Plaid that’s certainly the case.
Last Plaiderdays I worked with two other engineers on an internal tool to make debugging easier. This time around I thought it’d be fun to make use of our fancy new workshop — and our delightful 3D printer — to build something. But what?
Fortunately, two of my coworkers already had something in mind. Ryan and Haley, who helped plan and execute Plaiderdays, came to me with a question: Could we use the 3D printer to make trophies for the winners of Plaiderdays?
After all, no hackathon is complete without prizes. Plaid got its start by winning the TechCrunch Disrupt hackathon back when we were only three people, and today we gleefully award five prizes during Plaiderdays. They all have stupid names.
The Meritorious Feat of Industry: for the best non-technical project. The Golden Mixer: for the most cross-functional team. The Technical Award of Science and Technology: for the most technically impressive project. And The Burner of the Midnight Oil: for whoever sleeps the least. (My team accidentally won this one last Plaiderdays by working on our project until 4am.)
But there’s only one prize that really matters: The Honorable Golden Plaidypus. Named for our beloved mascot, the HGP is awarded for the most impressive project overall. The winning team members get their names emblazoned on the HGP trophy and receive individual trophies for posterity and bragging rights. Honorable indeed.
I loved Ryan and Haley’s idea but had my doubts. Plaiderdays was already upon us. More importantly, 3D prints look and feel cheap. I wanted the winners to get awesome trophies, and if I were to merely print something for them then it probably wouldn’t turn out well.
Nevertheless, and in the can-do spirit of Plaiderdays, I agreed.
The concept of Plaiderdays has existed for almost as long as the company itself. I’ve been at Plaid less than a year, but I’m told that the early Plaiderdays were held with little fanfare. Back then there weren’t enough resources to make them special. Plaiderdays simply provided an excuse to work on projects that were in need of some love. The inaugural Plaiderdays was just three people working in a cramped office, and nobody even remembers where the name “Plaiderdays” came from.
Plaid has changed a lot since then. We’re about 130 people now, which means we can do cool things like buy balloons and print posters. And did we ever!
In harmony with our space theme — “Into A Brave New World” — 2001: A Space Odyssey and Planet of the Apes played in the café. Nick, our fearless design leader, held a rocket-building workshop to teach fellow Plaids how to make rockets out of cardboard. And of course, we got t-shirts, long-sleeve ones with the Plaid logo styled to look like NASA’s.
Ryan and Haley wanted to design the trophies based on our theme, so we decided that the trophies should be shaped like rockets. I logged on to Thingiverse, a website where people share 3D models they’ve created, and found an awesome collection of rockets. I downloaded them and printed a test batch.
Truth be told, this wasn’t my first foray into 3D printing. About four years ago my dad bought me a kit to build my own, which, at $300, was the cheapest option available. The kit came in hundreds of pieces and took roughly eighty steps to assemble.
It was then that I learned how 3D printers work. For all their perceived magic they’re surprisingly easy to understand. Imagine a hot glue gun pointed downward onto a flat surface. Pull the trigger and draw a shape on the surface. Great — you’ve just laid the bottom layer of your creation. Now raise the gun a little higher and repeat, so that each layer builds on the one before it. When you reach the top, you’re done.
That’s exactly what 3D printers do, only they use plastic (or wood, or metal, or chocolate) instead of glue. The only components you need to make a 3D printer are a nozzle that gets really hot, a motor to push plastic through that nozzle, and a few motors to move the nozzle or the print bed itself.
To stay cheap, my first printer cut corners wherever possible. The larger parts were made of wood and had to be punched out of a pre-cut sheet. The tiny motors moved the print bed by turning string around plastic spools, which I coated with sandpaper so the string would stick to them.
Sadly, I never got the thing working properly. The software had a bug that caused it to freeze in the middle of prints and I eventually gave up. My enthusiasm for the hobby died then and there.
Or so I thought. Four years later I find myself the proud owner of a 3D printer that cost about the same as my old one but that worked out of the box. It’s a step up in every way. Its body is made of sturdy injection-molded plastic. The motors drive the extruder using rubber belts instead of string. It has a bigger printing area, a touch screen, WiFi, and cupholders. Like any other technology, 3D printers have dramatically improved in quality and reduced in cost since their introduction. OK, I lied about the cupholders.
For my test run, I printed all five rocket models to see how my budget printer would handle the finer details. Most turned out well but some had design or print flaws that I worried would prove troublesome. We voted on our favorite design and declared a winner: the SpaceX Falcon Heavy.
An added bonus of choosing the Falcon Heavy was that its wide, smooth boosters would provide ample space for text. My initial idea for the trophy was a rocket with some text on it, plain and simple. But a couple test prints showed that 3D text, which appeared so cool in my mind’s eye, didn’t actually look good in plastic. I scrapped the idea and sync’d with Ryan, who’s a designer, to brainstorm.
First, we tried to figure out how to make a cheap 3D print look and feel substantial. 3D prints weigh very little and come out looking like dull plastic. We considered solving the weight issue by filling the prints with sand. We figured that we could print the rocket in two pieces, fill ’em up, and then glue ’em back together. But this seemed doomed to fail because it would be difficult to fill the rockets completely, so they would probably feel more like shakers than trophies.
We tabled the weight issue for the moment and considered what kind of base the trophy should have. Some materials, like polished rock, would be beautiful and solve our weight issue. (Enter two-birds-one-stone pun here.) But it would probably take weeks to order them and we only had two days.
After considering more readily available materials we settled on wood. Wood doesn’t have the same heft as stone but it’s easier to work with and extremely cheap. And, as Ryan informed me, we’d be able to laser-etch designs into it. Wait, what?
I’m convinced that the design team has the most fun at Plaid. They recently ordered a laser cutter so they could make branded items for recruiting purposes. At least that’s the reason they gave me. Anyway, I don’t care why we have the thing, I’m just excited that we do. On the first day of Plaiderdays my design team gave to me, one laser cutter!
Sort of. It was still in its box and, like my old 3D printer, required a ton of assembly. A college friend who happened to be visiting for lunch decided to help Ryan and me put it together. I’m not sure what she had planned for that afternoon but I assume it had little to do with building a laser cutter.
Laser cutters are similar to 3D printers in several ways. An entry-level laser cutter like ours can be had for around the same price as a 3D printer. Plus, a laser cutter is practically identical to a 3D printer except there’s a laser instead of an extruder and there’s no movement along the Z-axis. Once completed, our laser cutter worked without issue. In minutes we had our first successful etching.
By now Ryan and I had a rough idea of how the trophy would look. Rather than print the rocket in one piece, we would to print it in two pieces — top and bottom — and separate them with a thin piece of wood. That way the trophy would have a more complex shape than a straight line, and it would look like the rocket were bursting through a wall. Ryan, with his eye for design, insisted on slicing the rocket a third from the bottom. “Better go with the Rule of Thirds,” he told me. Sure, Ryan, whatever you say.
I loaded the model into my slicing software, which converts a continuous 3D model and slices it into layers so the printer knows how to print it. The software I use has a fancy option that lets you slice a model into pieces. I set the slice at the desired height and then placed several copies onto the virtual print bed. That way I could avoid extra trips to the printer to remove finished prints and start new ones.
While the printer got to work I realized that I had no idea how many trophies I needed to build since teams range in size. Ryan and I decided to hedge and build five. If need be we could always make more.
Ryan had just come back from the hardware store with two long, flat pieces of walnut wood. In our first use of trigonometry since high school, we figured out how long to make the pieces. We wanted the trophies to make an “X” with the rocket and the piece of wood meeting in the middle, so we decided to cut a 45º angle into one end of each piece so it would lie flat. We measured twice, clamped the wood down, and cut the pieces with a comically tiny saw.
After sixteen hours of printing, the first set of rockets was done. I’ve printed dozens of things but I’m still amazed when a prints finishes successfully, especially when it takes so long that I have to leave the printer on overnight.
One tricky aspect of 3D printing is handling how the plastic interacts with printing bed beneath it. If the plastic doesn’t stick, the first layer will never get laid down properly and the print will be ruined. A common remedy for this is to rub the printing bed with a glue stick to give it some tack. But there’s also the opposite problem. When the plastic cools it forms a strong bond to the printing bed. It takes a lot of force to break this bond, so delicate prints can easily break during removal. To reduce the risk of breakage you can use a flathead screwdriver and hammer to gently chisel the print away from the board.
Despite my best efforts, a couple of the rocket bottoms broke when I tried to remove them. I queued up some more prints and eventually had ten pieces: five tops and five bottoms.
A 3D print isn’t perfect when it comes off the bed. Its surface is jagged from slight variations and mistakes. That’s fine when you’re printing something like a replacement part, but details matter in the trophy-making business. I’d earlier found an online guide to finishing 3D prints and had ordered the required materials. It was now time to put the finishing touches on my rockets.
First, I sanded them to smooth out imperfections, working my way from coarse grit to fine. Next, once I was sufficiently covered in plastic dust, I prepared for step two: lacquering.
Nick, who practically lives in the workshop, recoiled when he saw me reach for the XTC-3D lacquer. Nick is a tough-looking guy who wears only black and adores skulls, so I was surprised by his reaction. “Be careful with that stuff,” he warned, his eyes wide. “You don’t want to be breathing that. And don’t use it in here, either. Go somewhere else.” At this point he grabbed a surgical mask — I’m telling you, the workshop has everything — and escorted me to a conference room with big windows. “Open these and get gloves from the kitchen. That stuff is toxic.” What the heck had I bought?
Poison, apparently. The box for XTC-3D contains a three-page pamphlet of warnings. Apparently XTC-3D is exothermic, skin-melting, eye-disintegrating death in a bottle. Two bottles, actually. The active ingredients can’t safely coexist in one bottle so you have to mix it yourself. I gladly followed Nick’s advice and suited up, donning plastic gloves, a surgical mask, and a pair of red-tinted glasses that came with the laser cutter.
Carefully, and with music by 2 Chainz to guide my hand, I applied two coats of liquid death to my rockets. They immediately looked smoother and shinier.
I left them to dry and walked to the nearest sink with my mask, gloves, and glasses on, unwilling to risk touching my skin. I held up my hands in that don’t-touch-me-I’m-sterile way that TV doctors, and therefore probably real doctors, always do. I felt awesome and dangerous.
With the gloves off and my heart rate back to normal, I joined Ryan and Haley to sand down the pieces of wood. We then began experimenting with the laser cutter and figured out how to change the laser’s speed and power, which together determine the quality and darkness of an etching. We made other discoveries along the way. Lesson one: don’t look at the laser. It’s pretty and blue but it’s bad for your eyes, hence the red glasses. Lesson two: wood + laser = smoke. When you’re using a material as flammable as wood, etching is just a fancy name for controlled arson.
Ryan toyed with a few designs in Adobe Illustrator but in the end we decided to use the NASA-inspired logo that we’d used for our t-shirts. We spent a couple hours tinkering with designs and laser settings, but the etching itself took only a few minutes per piece of wood.
By now it was about midnight on Thursday, and most of the other teams had gone. With the lacquer dry, we gave the rocket pieces one last sanding. Then it was time to paint.
In a shocking display of common sense for midnight, we brought the pieces outside before spray-painting them and even laid a tarp under them. How responsible of us!
The paint provided an amazing finish for the rockets. We gave them two coats for good measure and brought them back upstairs for final assembly.
Five rockets, five pieces of wood, one bottle of super glue. It all came down to this. Using as many tools as possible — including clamps, a quick square, and several cans of paint as weights — Ryan and I nervously joined the rocket pieces to the finished wood slats… and… liftoff!
We were beyond excited. We’d gone from concept to finished product in two days and the trophies had turned out splendidly. Dazed, we only had enough energy to stare at our creations in awe. I headed home around two-thirty and dreamt of rockets.
A week later the winners were unceremoniously announced in an email. The winners of the HGP — and of our fabulous trophies — are… Kevin?
Not “Team Kevin” or “Project Kevin” — just Kevin. By himself. Kevin worked on several projects. He wrote sketches for our SNL-inspired skit “Plaiderdays Night Live” and took lots of excellent pictures including some of the ones in this essay. But never mind why Kevin won the HGP. What are we going to do with all those extra trophies?
I’m actually not worried about it. The trophies shall forever remind me of the time I journeyed into the brave new world of trophy making. I’ll be glad to have them around the office wherever they land. Come to think of it, there might be room for one on my desk.
As for the whole “developer vs. engineer” debate, I don’t think it’ll be settled anytime soon. One could always argue that software developers don’t deserve the title of engineer. Let’s just say I’m building toward it.