Find all your DIY electronics in the MakerShed. 3D Printing, Kits, Arduino, Raspberry Pi, Books & more!

The following was written by Michael Rosenblatt, founder and CEO of Seamless Toy Company, maker of ATOMS.

Hard Choices When Designing an Electronics Construction Set

Over the last few months, I’ve had the opportunity to work with a great team to design a new electronics construction set called ATOMS. The design goal for ATOMS was to create a plug-and-play construction set that required no experience in electronics or programming. We designed ATOMS to be simple enough for a 5-year-old, and powerful enough for a professional — which means that making table-top creations isn’t enough. You have to be able to build “real” things with it — like a cake with motorized elements, or a nightlight that you can switch remotely with a the shake of a “magic” wand. We designed ATOMS to enable more people to make great things, and particularly with kids, we hope ATOMS will seed a curiosity for how things work.

However, there is an inherent contradiction in these goals — tools that help users extend their capabilities almost always abstract away the true nature of how things work. ATOMS is no exception, and that leads to a number of design decisions we struggle with on a regular basis. We want ATOMS to both empower users, but also be transparent, and oftentimes those objectives are at odds.

Take, for example, the most basic project a user might build with ATOMS — a knob that controls the brightness of an LED. ATOMS uses a “signal flow” analogy throughout its architecture: signals flow from “arrows” to “boots.” and routed with a wire (in our system this is a 4-terminal cable that carries serial data, power, and ground). The arrows and boots blink when a signal is present, and blink faster when a signal is stronger. A stronger signal is generated by turning the knob up, and a stronger signal to the LED ATOM makes the light shine brighter.


Using ATOMS to control the LED’s brightness

ATOMS provides a great logical representation of what’s happening that is easy to understand, but is a significant abstraction of what is actually happening on an electrical level. Still, it’s a start. By thinking in terms of signal flow, we think that kids and non-electrically minded makers can get started faster, and build more powerful creations sooner.


What’s really happening to control the LED’s brightness

Take a more complicated example, such as a self-balancing robot (think Segway). This is typically a challenging project, even for an experienced maker. With ATOMS, by abstracting the complexity of signal processing and motion control into our modules (each has a microprocessor), this project can be built with just two connections from opposite sides of the Accelerometer ATOM. If I’m falling one way, run the motor clockwise, if I’m falling the other way, run the motor counter-clockwise. So you can see, ATOMS can be a powerful maker tool (which we want), but currently we are adding a lot of abstraction (which we probably don’t want).


Using ATOMS to build a self-balancing robot

Even so, I think ATOMS is off to a good start. We launched in November, and hit our kickstarter goal to fund production in a just over three weeks. We have introduced 13 modules, ranging from a Record-Playback ATOM, to a Bluetooth Smart ATOM, and we’ll be building out the library in 2013.

Our challenge, as we grow and expand the ATOMS system will be to enable our users to also grow and peel back the layers of abstraction as their comfort level with electronics and their curiosity increases. I think Little Bits may be onto something by making their modules open source – and that model might fit us too. We’re considering several design directions to enable our users to “get under the hood”, but for now, we are just happy to be helping more people play and make things.

Mark Frauenfelder

Mark Frauenfelder is the editor-in-chief of Make magazine, and the founder of the popular Boing Boing blog.


Related

Comments

  1. Ben Young says:

    I don’t think “abstraction” really means “obscuring the true nature of how things work”. The abstraction *is* the true nature of how things work. Any object designed by a human is modular in precisely the way you describe. Kids (and adults!) *need* things to be broken down into simple systems before they roll up their sleeves and start solving problems.

    I mean, look – you yourself drew a diagram with black boxes labeled “UART” and “Microprocessor”, which are surely abstractions too. I’m sure you weren’t thinking about the transistor layout in the UART when you were designing the ATOMs! So I guess what I’m saying is that you have nothing to apologize for. We should be proud of good abstractions; they’re the key to solving problems.

    That said, the kids who do master the toy, get a little older, and grow curious about the next level of abstraction – “how does this ATOM work? Can I have a new sort of atom that does X?” really should be supported in learning about that too. You want the kid to pry open that box with her screwdriver and see, rather than a confusing morass of electronic components, an assembly of other atomic pieces which they could understand, if they wanted.

    Of course, as you’ve observed, open-sourcing the design is by far the most honest way to do that… If you hide the design then you really are getting in the way of the curious kid who wants to learn more. Just sayin’. :)

  2. djkindberg says:

    Frankly I was a little confused by the abstract explanation. As most of us think in a linear fashion the A2D Input (should be Analog to Digital Converter or ADC) should follow the potentiometer which is an analog device. The microprocessor, is a digital device and cannot use the signal from the potentiometer without conversion.

  3. rocketguy1701 says:

    I have the same issues with this as I did with littlebits, it’s too packaged. My mom wired up components like bulbs and switches to nails driven into a wood block, soldered up tiny jumper cables with gator clips, and gave me the rawest possible introduction to electrical circuits. Nothing hidden, you immediately got what was going on, and it was *real*. Went from that to the radio shack spring-based 101-project kits, then breadboarding digital circuits, and now I’m a network engineer. The two things that may have been the most important for me to learn is that A) it’s not magic and B) I can mess with it.

    Sometimes less really is more. If it’s wrapped in plastic, frankly I’m not interested, I’ll build another set of blocks for my niece instead.

  4. I believe your goal should be to simply inspire the inquisitive traits that would drive a child to dig further. Hiding by abstraction is just the way we do things. If a child goes somewhere else to understand nature shouldn’t matter to you; whether they were driven to do that from your stuff should.

In the Maker Shed