Originally interviewed October 11, 2012
API Economist: You have quite an interesting background. I love the combination of the artistic and technical. In many ways, you are the archetype of the API economist. Tell us a little bit about yourself, including your stint at the startup, Boo.com.
Cameron Hickey: I went to Bard College back in the '90s. I majored in Photography, but I always had an interest in both political science and economics. While I was there, I taught myself to program and build web apps and I actually left halfway through my senior year. I started working in New York City, first freelance, and then at one small Internet company and then I got a job at Boo.com. I spent about a year and a half working at Boo.com based in the New York office. I worked on a variety of projects, but primarily helping build and launch the first and second versions of that ill-fated website.
API Economist: CNET actually listed Boo.com as one of the greatest dot-com busts. Those failures are generally badges of honor…and that’s what I love about the start-up environment.
Cameron Hickey: I worked there pretty much until it collapsed. I left about a month before they actually closed the doors. It was definitely a really interesting experience. I was a kid. I started college at 16, so when I left I wasn't even old enough to drink. When I worked at Boo.com, I was neither old enough to drink at the parties that they had or even contribute to my 401k, because I was too young.
API Economist: By the way, how old do you have to be to contribute to a 401k?
Cameron Hickey: You have to be 21.
API Economist: I didn't know that.
Cameron Hickey: After Boo.com, I did a bunch of different freelance gigs. Then I got into film making and computer animation, working on a couple of small films as well as a Miramax film called, Naqoyqatsi, where I did about seven minutes worth of the film created in 3D animation. After September 11th happened, my girlfriend, who is now my wife, and I decided to leave New York, and we spent two years living aboard making our first independent documentary film.
Because making independent documentary films is not something that is easy to fund, but it is quite expensive, I continued to do a lot of freelance web application development. I've developed in ASP, I've developed in Java, and when Ruby on Rails came around I switched over.
Now, I basically don't do any web development for money, I just do it more in the service of the documentary and journalism work that I do. I work quite regularly, although not full time, for the "PBS News Hour," WGBH, and other public television stations.
API Economist: Do you think we are seeing a new class of professionals who, in addition to mastering their professions, happen to be pretty good hackers?
Cameron Hickey: In the journalism industry we now call it a one-man band. It used to be that a news crew would have a producer, an associate producer, a cameraman, a sound guy, and then maybe also a correspondent. Now, I travel around the country with a correspondent and I am the camera and sound. I’m also the producer...and the correspondent also sometimes operates a second camera.
In the world of web development, the same is also true. You used to see big teams. When I was at Boo.com, there was a massive software development team. Today, that entire site could be built by two people in a couple of months.
Within the scope of journalism there is a big drive towards data-driven journalism, which is built on the accessibility of a lot of these API's. Things that use to be really difficult or complicated to do are much easier to do, and the project that we developed at the Election Hackathon 2012 is a perfect example of that.
For the Election Hackathon, we built a project called Capital Clout in which we pulled all these different APIs to collect statistical data, analyze it, and try to find something new that we hadn't really realized before regarding the quality of our legislators. It's interesting, in the last couple of days I've actually been discovering new things that I didn't know and I didn't understand. When I posed the question, "Why is that that way?", I can't even find news articles about it, which suggest to me that people haven't had an opportunity to look at this information in this way yet, so no one's posed that particular question. And now because of this data and the easy access to it, a lot of new questions are rising up.
API Economist: How did you approach the design of the site?
Cameron Hickey: Thanks to the kinds of technology that we have, the platforms and frameworks make it easy to build things with a lot of complexity really quickly. Thinking about web site design from an historical perspective, it has grown quite a bit from the early days. The first generation of web applications looked really crappy. Some of them I think still do. But today, if you look at what gets funded by VCs in the Silicon Valley, you see that design plays as big a role as the quality of the software development.
So the people who are doing the programming and back-end coding need to have a lot of comfort with making things look good from the beginning, or people aren't going to take your project seriously.
That was actually something we employed in the first version of this app (Capitol Clout), and that's part of the reason why it was successful. I think if we did the same thing and the site was ugly, I don’think anyone would've noticed it in the way that they did.
API Economist: Speaking of the Election Hackathon, and as background for of our readers, your project, Capitol Clout, was the winning app. The Hackathon was held by the Washington Post, and also sponsored by NPR and the Sunlight Foundation. This was the first time the Washington Post had ever held a hackathon, and they just launched their first ever developer portal. How did you get involved with this Election Hackathon?
Cameron Hickey: I've been passionate about technology startups since I was in college, and while I've moved forward with my journalism career, I've always followed the industry quite a bit. Back in, I think it was the spring, I decided to participate in the TechCrunch Disrupt Hackathon. It was the first time I had ever done anything like that - spending 24 hours building something. They had a dozen different APIs, and I met someone on-site who I had never met before, and we built a hack that we were able to present. We didn't win anything, but it was really a fun, exciting experience.
Then I decided to participate in another hackathon, which was the AngelHack Hackathon, which is really a hackathon that's geared toward trying to get angel investment in a business. At that hackathon we built an app called MediaWire, which was sort of a curation platform for breaking new media. It started to bring together my interest in journalism, breaking news, international news, with building an app.
The app that we built at AngelHack in New York actually was one of the six winners. So we got a free trip out to the Silicon Valley to compete in the AngelHack finals, which, sadly, we didn't win, but it was still a great experience to meet investors and other entrepreneurs, and stuff like that.
I'm not sure exactly where I first found out about the Election Hackathon. I think maybe somebody had participated in one of those events, or maybe I got on a list or something, but I got a notification, "Hey, we're doing this Election Hackathon, you ought to come join."
My wife works in journalism at Bill Moyers and they employed in some of the work they've done, Sunlight Foundation and Sunlight Labs tools. She was notified about it and said, "Hey, this seems like something you should do, or maybe we could do together." So my wife and I decided, "Why don't we make a weekend out of it?" and made into an uber-nerdy vacation to Washington, D.C.
API Economist: Last year at SXSW (South by Southwest) in Austin, we attended the Backplane’s Managers Hack and had the opportunity to catch Deepak Chopra open up the hackathon with some pithy words of wisdom. I don’t remember all of his words but I do remember him clearly characterizing hackathons as a form of creative chaos. I thought he nailed it.
Cameron Hickey: It's an interesting way to look at it. Hackathons bring together the idea that you want to accomplish a goal, you want to accomplish it really quickly, and you are going to do it not in the way you normally accomplish a goal. You're going to do it fast and dirty, so it's definitely chaotic. What would never be the right way to write code is a fine way to write code at a hackathon. It doesn't matter how complex the code is that you've written, how clever it is, how many different features it has, you really have to start with, "What's my demo going to be? What am I going to describe as a problem and my clever solution to it?"
That's actually where the most creativity comes in, figuring out what's a simple problem to describe, and a solution that people can understand to that problem. At the previous hackathon, the AngelHack hackathon, where we won, before we wrote a line of code, we actually devised our entire pitch presentation. Then, everything we built was in the service of the presentation, so anything that didn't need to be built didn't get built.
We employed that same strategy at the Election Hackathon. We came up with exactly what our problem was and how we were going to describe the solution.
The hack we ended up building was really an incredibly simple thing, it's basically a two page hack, "Collect all the data, process it, and show it on an individual page for each legislature," but it demonstrates a lot of potential for the things you can do in the future, and I think that was why it was successful as well.
API Economist: I want to get your perspective as a developer, so keep your developer hat on here. There's the great line from "Field of Dreams" back in 1989 where Ray Kinsella (played by Kevin Costner) hears a voice that whispers, “If you build it, he will come.” Obviously, that worked in baseball, but it doesn't always necessarily work with APIs. If you build an API, will the developers come?
Cameron Hickey: I came for one reason, because there was a Hackathon to use it. That was the impetus. Every Hackathon you go to there are API presenters offering prizes for people to try to build something with their API. For most developers, it's going to be the first time they try and use that particular API. That's definitely a great carrot to trying to get your API out there to some subset of developers that may eventually use it in a more practical use case. I think that everyone doesn't need to have an API. I think that APIs are valuable when the data or functionality that they reveal can be re-purposed outside of the original organization's use case, when it can be mashed up or combined or give broader access to something.
News articles, especially those related to politics, are less relevant and important the older they get. And people built a lot of apps at this event and other events that I've seen where they pull out content. It just doesn't matter. I don't need to see a list of old articles about something. It doesn't actually help me that much. In certain cases it might, but I think sometimes content based APIs aren't going to give you very much that you care about.
I feel like going beyond just building it so they can come is either thinking through what features of an API are you going to give to somebody that will allow them to either create new and valuable ways of looking at information or new and valuable ways of making money. If it's not easy to make money off an API that you built then there's no big reason to use it.
I was trying to think of a different way to tackle this question. There's got to be more than just giving you access to the API, creating code libraries to use the API. There's got to be a reason for it. For example, Stripe has a great API that makes it easy to accept credit cards on the web without a merchant account or gateway. Their API solves a problem that everybody has that you couldn't solve easily before.
API Economist: Cameron, this is our last question for you. Maybe I can have you put your prognosticator's hat on here. You're an independent filmmaker and journalist. How do you see APIs evolving in your line of work?
Cameron Hickey: I'm incredibly interested in data driven journalism. I think that data visualization presents a lot of potential for reviewing patterns that you couldn't otherwise recognize or that some insider wouldn't necessarily know about. I think that the more kinds of APIs we have available to us, the better able we are to analyze them quickly to prove a hypothesis about some fact or some trajectory that the world is taking. I also think people respond really well to simple graphics or simple summations of information and APIs give us the ability to do that a lot more effectively - especially within the context of open government. Many different elements of our government are making content available to the public. They're trying to release as many documents as possible.
The way those documents are being released is not always in a clean, easy to digest API format and so it's still quite hard to deal with this stuff. Entities like Sunlight Foundation and GovTrack processing that information and making it easier to consume in a software-oriented way, there's a lot of new potential in terms of what we can uncover and what we can describe to the public. I think it's going to have a tremendously positive effect and we're just at the beginning of it now.
API Economist: Thank you for sharing your insights.
Cameron Hickey: Great. My pleasure!