January 26, 2017
Let’s start from the beginning. It’s February 2016 and I’ve just been hired at WATTx.
One day, we have a company-wide meeting and someone suggests it would be really cool if we could all come up with our own ideas for what to work on next. Preferably something associated with the Internet of Things (IoT).
If not IoT, the ideas should at least be related to hardware, software, or data science. And here I was, thinking about perfecting meatballs, one of the few things my native Sweden is known for.
If I’m being honest, by this stage, I had only a vague understanding of IoT. I’d read articles about how it was the next big thing, with headlines that generally went something like: “X says that in Y years there will be Z billion smart devices”.
A data scientist at work?
Software development was something I was kind of familiar with, having worked for e-commerce companies, but being able to write a bit of CSS is not the same as coming up with the next big SaaS-idea. Hardware just sounded scary and I still think that you need to wear a white coat to work in data science.
How could I come up with an idea that would be worth working on?
The answer, at least for me, was exploratory research.
Some of you reading this might be familiar with the concept of exploratory research but for those of you who have never heard of it before, it is research that is done on a problem that hasn’t been clearly defined.
In academia, secondary research often means that you will go through sources (often very thick books), look up scientific articles, and generally search for every piece of information or data that can be found on the specific topic. Of course, you can and should qualify your quantitative research with some qualitative research like informal interviews. Basically, you build up your knowledge of a subject and then you use qualitative research methods make sure that your understanding of a subject is correct.
So how can exploratory research help you to come up with ideas? Well, for me, ideas often start with a gut-feeling. I read something and feel like I’ve read the same story before. I’m out running and run past a farmer stacking hay in the exact same manner they would have done it 200 years ago. And I get a gut feeling that there must be a way to solve this perceived problem. And that’s where exploratory research comes in. After all, it is a way of researching a potential problem that isn’t clearly defined.
Having decided to give this idea generation thing a try, I started to follow up on hunches I had. I spent a couple of hours looking into agriculture tech. After reading about voting fraud in, I think, Zambia, I decided to look into that, too. And you know what, sometimes it worked. I started to come up with ideas. Many were dead on arrival and, frankly, not very good (a lot of them involved the “blockchain” - the deus ex machina of technological problem-solving). But some, I thought, were at least interesting.
What follows are three ideas I came up with by doing exploratory research. All three of these ideas were eventually killed, so nobody ever worked on them. However, I hope that they are useful as an example of how you can come up with and develop ideas.
Gut-feeling: There’s a lot of gun violence in the world What I found out: There are a lot of products and measures for gun safety. However, most of them are analog. But I found some digital versions of products like gun locks and gun safes. What I thought: There must be a new way, using IoT, for us to reduce gun violence. My idea: Use a camera connected to some sort of image recognition software and then connect that to a gun. Basically, I wanted to build a gun lock that wouldn’t fire on something that was recognized as a human shape. Why it failed: Before I even pitched the idea, it started to fall apart. Technically, it seemed to be possible. But reading up more on the subject I found quite an active resistance to gun control in general and new technology enabling better gun safety in particular. I decided to go straight to the source, and posted on a subreddit for gun enthusiasts in the US, asking them what they thought about these new technologies. Without going into too much detail, I got a very negative response and decided to abandon my idea.
Gut-feeling: Agriculture doesn’t seem to make use of modern digital technologies. What I found out: There are a lot of digital initiatives in farming. One of the big trends is the growing importance of data collection. Drones, especially, seem to be very useful for collecting agricultural data in an inexpensive and efficient way. However, the data that’s collected can be hard to understand for farmers. And the way you access that data in most cases is by either bringing a laptop out in the fields or by collecting data during the day and then going over it in the evening, at home. Looking for new ways to present data to farmers, I found a couple of interesting articles on using augmented reality in agriculture. What I thought: There are some interesting research findings on the use of augmented reality in agriculture, but no company that has taken the idea and really run with it. I thought that it would be very interesting for us to create that company. My idea: Give farmers real-time data visualization when they are out in the field. That could help with everything from detecting harmful insects to making sure that you fertilize your fields in an optimal way. Why it failed: Generally people were interested in the idea, but felt that we needed to focus on ideas closer to our areas of expertise. For this idea to work we would need to build up knowledge in precision farming, drone data-collecting, and augmented reality, which would demand a lot more resources than other projects that were suggested.
Gut-feeling: The act of voting is the pillar of democracy, yet the process is not waterproof against attempts to change results and manipulate voting systems. What I found out: Electronic voting machines do exist, and are used in a number of different countries (namely in the US and throughout Southeast Asia). Voters themselves say that they want more modern and efficient ways to vote. There is also a big market with a number of large companies that have been making electronic voting machines for a long time,. However, there’s a lot of identified security flaws and ways that electronic voting machines could be compromised. What I thought: There must be some way of building a safer voting machine. My idea: Over the last couple of years, we’ve seen more and more companies make use of technology like facial recognition and fingerprint scanning. By incorporating these technologies in a voting machine and making sure that the machine does not have the same vulnerabilities found in existing voting machines, you should increase the ease-of-use to vote, leading to higher turnouts and increased public trust in results. Why it failed: As you probably can see above, the idea was a bit half-baked. In hindsight, I should have spent more time thinking about exactly what we could do to increase security. At it’s core, there’s a kernel of a good idea here, but to build a voting machine you would have to research legislation, cutting-edge security measures, and then actually build the machine in itself. Like augmented reality in agriculture, that would mean that we would have to commit a lot of resources to this one project, which we didn’t want to do at the time.
I don’t think you can come up with a truly creative solution without understanding a problem in depth. That’s why exploratory research is such a useful tool. It helps you build that understanding which let’s you come up with ideas. Maybe not all of your ideas will be doable. But personally I was surprised with how much you can learn doing a couple of hours research and how that enabled me to think up somewhat realistic ideas on how to solve problems I didn’t even know existed when starting my research.
My impressions from the two days I’ve spent at a ‘un’-conference
Innovation - is it an art or a science? We say that it is both....
In this blog article, will go through the most important preprocessing step when analysing time...