The Linker is a new blog from Christopher White and Elecia White, hosts and creators of the weekly embedded.fm podcast. The Linker discusses embedded software and hardware engineering, education, and entrepreneurship. The goal is to dive deeper into the tangential topics brought up during the show.
Sophi Kravitz is a Mythical Creature. She writes an element14 blog but always about other people so you might not have realized her fabled status. Normally an electrical engineering consultant, Sophi is one of those people who get to know everyone and then links them together. Think of it as matchmaking but with a bent toward getting things built. Since responding to a job post for a Mythical Creature, Sophi has been working on the new Hackaday Prize.
Last year, Hackaday had a big contest for connected devices. The top prize was a trip to space (or ~$200k US) and there were other prizes totaling well over $100k US. I got to judge. It was great: I learned an incredible amount just by reading the entries.
This year, Hackaday will have another engineering contest focusing on solving problems that matter. The top prize remains about the but there will also be some new prizes. I get to judge again. Since Sophi leaked the details on last week’s embedded.fm show (proper announcements here), I’ve spent some time thinking about what I want to see. It has made me wish I could enter.
On the show, chatting with Sophi and Chris about soft robots, I said I wanted a robot worm to eat radioactive waste. Realistically, a soft robot to clean up ground or water pollution would be great. Why limit it? I’d love to see any device to clean up any pollution.
We talked about systems that would save the world from asteroids (ok, in recording, I accidentally said the Earth needed to be saved from astronauts but I’m sure everyone understood, right?). On the other hand, there are more things here that we need to fix first: hunger, clean water, cheap power, education, medical care, etc.
There are many companies working on these problems and no one expects them to be fixed overnight. Hackaday is about encouraging the maker community: helping people work on projects outside their day jobs. They encourage people to learn new things, to share what they’ve done, and to teach others.
If you keep that in mind as you enter, I think you’ll do well. On the other hand, since Sophi leaked the Hackaday Prize details for me, I feel it only appropriate that I leak a few suggestions on how to win my personal judge-y votes.
Like submitting a proposal to a conference, entering a contest is odd. It isn’t necessarily difficult but it may not be what you expect. Also, this advice is not limited to Hackaday, there are lots of other engineering contests. Our host, element 14, often is running a few. As of this writing, the current contests are In the Air, about using The Internet of Things to fight air pollution, and Sudden Impact, about making sports safer for young adults. They are also teasing a SciFi Raspberry Pi contest, starting soon (I hope it will involve at least a little steam punk). Since the contests change often, look at the Design Challenges page for the latest. Some of this advice may be useful for those.
First, remember that I (and the other judges) will be looking at twenty to fifty entries at a time. I need to understand your idea quickly. Figure out how you might explain it in as few words as possible (i.e. how would you describe it on twitter). With lots of entries to juggle around, a short, clear summary helps me remember you. Don’t get buried with detail and why yours is awesome, be concise.
Once you have me hooked with your brief summary or tag line, introduce the idea as though you were talking to a high school senior. With a mix of software, hardware, industry, and academic judges, you can’t be sure who you are confusing.
It may seem strange that an engineering competition comes down to communication. It does. If you show me the working cold fusion reactor that you built in your basement and I can’t understand the source of your glowing light, well, I can’t score it highly.
Use all of the communication methods open to you: diagrams, words, flow charts, pictures, and videos. While perfection in grammar and presentation would be lovely, that’s not what I’m scoring on. Instead, I want to understand. Help me get to know what you've done.
Also, I really liked the projects that made me feel like I could build them. There was a scoring category for Hacakday last year (“reproducibility”). Not only did the finalists describe what they’d done, they told me how I could do it too. Going back to the fictional high school senior, how would you tell them to build it?
Once you get past the introduction to the device and dig into the details, don’t hide the errors. Failure makes for a good story. Contests are about learning. I want to see the progress, if only because by this point in your amazing project, I’m feeling like a slacker for not having started building one for myself.
Also, in the communication section, you don’t need to suck up to the judges. Even as I was amused at a few attempts to do that, it wasn't a good use of the contestant’s energy. Jokes would be good though. Feel free to pull a few heartstrings (but not all of them, it is an engineering contest not a sob story). While not at all a requirement, it is nice to feel like I’ve gotten to know you a little bit through your project.
If you can post a video, sometimes your personality can come through most there. I don’t like making videos and I don’t much like watching them either (heresy, I know). Still, if it is a contest requirement, you must do something. Except, don’t have silence with a floating hand showing the device, pressing buttons. Music doesn’t cover the silence the way you hope it will, either. If you hate your voice, use a speech to text engine so you sound like a robot. Or have a friend do a voice over. And you can show your computer screen (screencast) as well as the floating hand with your device. You can remain anonymous (one of the Hackaday finalists last year did).
I recommend that you make something for yourself instead of for someone else that you don’t know. Solving faraway problems sounds like it is the perfect thing: you have all this tech and can use it to do things, things that other people surely need. However, if you aren't there, having the problem, the chances you understand it thoroughly are slim. So, this is purely my preference, but it is based on experience: solve problems near you, then expand on them for other people.
Next, if you are judged on openness, make everything you can open, even stuff you think is silly. Using open source tools is helpful but not required (in my book): it’s not worth the effort of redoing your project just to get away from a tool you are comfortable with. On the other hand, publish everything. For hardware, that means schematics, Gerbers, and a BOM. Also, PDF versions are needed, even if you are using open source tools (some of us are software engineers, you know). Fo
r software, you need all the code, even that little bit of glue logic on that auxiliary processor that isn’t really doing anything. In
clude project or make files so someone else could build it. Don’t forget readme files to tell us how. Please put the CAD files out there too (again with PDFs for those without tools). Finally, if it doesn't say what license it is or if you say “we plan to make this open source”, it isn't open source. I’m not going to judge your code harshly but if you want all the points in the openness category, please let me check off all the boxes.
Here’s a piece of advice you may not like but is the most important: don’t aim for the top prize. Aim to learn something and to build something you are proud of. Use the contest as a way to get things done, to give yourself deadlines. Do this for yourself and for that little bit of the Earth you’ll be saving. Don’t be afraid. Even if you don’t win or don’t finish, the experience will teach you something for next time. Be brave and get started on that idea.
- Hackaday Prize
- Element 14 Design Challenge Page
Dev Kits (to give you ideas)