Soundbyte 396: The Art of Having Bad Ideas

26 May 2019

Okay, I’ll admit that I shamelessly stole the title of this Soundbyte from a wonderful keynote session I attended this week from Tobias Ahlin, lead designer of Minecraft and Spotify/Github alumnus. In our profession, where there’s often no physical component to our work in sight, it’s all about having the right ideas. But how do you arrive at the right ideas? According to the thesis of the keynote I alluded to: by having many bad ideas first.

That, in and of itself, is already hard for me. And I refuse to believe I’m the only one. We tend to think of ideas as these perfect thoughts that need to be formed. Aching to find an immediate solution to problems at hand, great ideas seem hard to come by. And, when we have ideas, we hold them to a very high standard before we dare share them. Almost as if every idea should fit in a TED talk (‘Ideas worth spreading’). Clearly, the bar needs to be lowered for ideas to flow.

In Tobias’ keynote, one of the tools he offers to help comes from improv comedy. Start with any idea, say it loud, and the next person should continue with ‘yes, and …’. All ideas are equally valid, and lead to new ideas being developed on top. This is not the phase for disagreement or discussion. Even if you’re not literally doing this with a group, it can be a good mental tool to apply yourself. Instead poking holes in ideas, build and improve upon them.

Another striking observation is that silence is underrated in meetings. When an idea is offered, we feel compelled to discuss its merits and pitfalls immediately. What if, instead, we would create five minutes of silence? Then, everyone has the opportunity to organize – and maybe write down – their thoughts and ideas before continuing. This also gives people of the more introverted kind some breathing room.

Even then, being able to generate many divergent ideas is only half the battle. The next question is: how do you find the signal in the noise? That brings me to another crazy idea I encountered this week. One of these out-there ideas that actually seems to work is the artificial pancreas for diabetics. Scott Hanselman (of Microsoft fame) presented this idea in another keynote, perfectly describing the dire fate of diabetics where it not for hard- and software supporting them. The only problem is: hardware manufacturers refuse to open up interfaces to for example insulin pumps and measurement devices. Therefore, closing the measure/intervention feedback loop using open-source software is incredibly hard. Scott shared he has to use a 10 year old insulin pump he got from Craig’s List, because newer versions disabled external control. Still, he and others persisted in reverse engineering the hardware. After many years, it finally seems like the FDA and other authorities are paying attention and new initiatives around open medicine start becoming reality. If you’re interested in these developments, I definitely recommend checking out Scott Hanselman’s talk ‘Solving Diabetes with an Open Source Artificial Pancreas‘.

I’ll definitely try some of the ways to generate more ideas going forward. It’ll probably feel counterintuitive at times, but if it leads to better outcomes, why not?

Now, time for some music. I’ve selected ‘Wait for the train’ by Joe Robinson, and you can probably guess why. Hopefully the strike won’t affect our commutes too much coming Tuesday!