API Economist: How did you become the API Evangelist?
Kin Lane: I was the VP of Technology at an events management company, WebEvents Global, leading their technology and architecture. I ran all SAP events, including Sapphire, for two years. I was also involved with TechEd and a lot of the North American events. I was brought on to move them out of the data center and be more elastic in the cloud, and meet the demands of the global events. I moved it into the Amazon Cloud and re-architected the whole system using APIs and Amazon APIs. I loved APIs. But I wanted to do something else. I started studying the API space. I quickly realized that there are a lot of technical pundits in this space. But no one was keeping an eye on the business of APIs, the myriad of tools it takes to be successful, approaches to evangelism and marketing to developers, and the whole politics of APIs. So I launched API Evangelist and just started studying this space. Three years later I'm still doing it.
API Economist: Life as an app developer seems to be getting easier and easier, especially with the explosion of mobile backend services, or what's also being called BaaS or MBaaS. Why is backend-as-a-service so exciting?
Kin Lane: The motive of BaaS is to bring together valuable and essential resources that developers use in building cross-platform mobile apps, and empowering them to do this as quickly, easily and efficiently as possible. Most of those resources tend to be available in API form. When you talk about user management or database key value stores, a lot of the social features, those are all things that are readily available through APIs. Really, what BaaS is doing is just aggregating these resources into meaningful stacks for developers. Right now they're hand-rolling a lot of the core pieces. But if you think about it, you could just pull together the best breed of any API resource and build an efficient backend framework that caches and gives developers single authorization across all of the services.
That's backend-as-a-service. It's just a natural progression of the API space. Rather than 8,000 single APIs out there that you have to weed through, people are assembling them into very targeted and meaningful stacks that help developers be productive and successful.
API Economist: Are you seeing that improve app development? In other words, making mobile app development more accessible to developers and increase the number of apps that they can release?
Kin Lane: The beauty for developers is that you don't have to think about all the nuts and bolts of the backend: rolling your own database, figuring out how to set that up, file management, image management, getting all the social fabric such as login and sharing and friends…all those pieces. If you don't have to go out and pull together all of that, you're going to be a lot faster.
There are also valuable cross-platform aspects to BaaS. Many of them are allowing you to deploy on iOS, Android, Windows Phone, and some BlackBerry devices. They let you roll those out a lot quicker because all your resources are right there.
Another key advantage is not worrying about how to scale your app. You don't have to think about, "Oh no, we have a thousand new users! What are we going to do?" The backends are going to do that for you. So really you can focus on what you do best.
Kin Lane: There was definitely a mix of attendees: one third novice, one third intermediate, and one third advanced. There were equal parts open developers and the other half being enterprise that were very business-focused. There were some big brands there such as Capital One, Target, and Best Buy. There were a lot of enterprise people represented there. But there were also a lot of thought leaders, and people bringing expertise from the open space. So really, it's not a question of, "Hey, is API a real thing? Is it something we should be paying attention to?" I think we achieved that in 2012. Now it’s, “How do we take the best from the open space?” These are all concepts that grew in the petri dish, the experiment that has been APIs for the last seven to ten years.
We are clearly in the “figuring out” phase. That is, how do APIs work in the enterprise, in more secure and hardened environments? How do you innovate around developers in these ecosystems? How do we do it in logical, sensible ways that work, across many different industries? Answering these types of questions is what the conference was all about.
API Economist: With regards to OAuth, the open standard for authorization, what is the state of OAuth today?
Kin Lane: I'm still processing it. There's a great discussion on the Google Group, API Craft, about the recent turmoil around the standard. OAuth 2.0 is ratified. Everyone's behind it and feeling pretty good about it. But a lot of the original folks and people who really believe in it felt like the enterprise got heavy handed and got a lot of stuff put in there. It made the spec more bloated then it needed to be, when it comes to certain identity practices and things that the enterprise wanted. But I think we got there. It's ratified. It's good enough. There's wide industry support. We need to run with it and we need to use it. We need to spend time educating folks about what OAuth is and why it's important.
Regarding adoption, most of the big top-tier providers are running with it. For example, Google has been moving all of their APIs to standardize around OAuth 2.0.
API Economist: In your book, Business of APIs, you and Audrey Waters aimed to bridge the technical and business aspects of API development. Who in business are you exactly targeting? Who cares about APIs in traditional organizations and enterprises?
Kin Lane: One, it’s companies looking to be more competitive and agile in this fast-changing environment. It’s organizations that want a healthy digital strategy that is not just mobile and tablets, but is equal parts social and using the cloud. For example, cloud services such as SalesForce, Drop Box, Google, and Box.net. And having an overall healthy digital strategy across social, cloud, and mobile takes APIs. If you want to be competitive and be healthy in that way, you've got to understand APIs and why your business needs APIs. These are business executives, marketers, and business people from all walks that have to solve day-to-day problems. They can't get the time of day from their IT departments.
API Economist: You've looked at President Obama's presidential directive that states, "Every federal government agency should have an API." As a matter of fact, you list all the executive departments or agencies that have published a digital strategy. While there are a number of agencies that have, there were way more that have not. Why did you start tracking this?
Kin Lane: In the public, open API space, I saw the success of John Musser from ProgrammableWeb doing a fantastic job of showcasing the growth of these new APIs rolling out, who's using them, and the mashups that are being built. The storytelling around this space is why we have an industry in 2013. I saw the value of me coming in and further extending that, teaching people how to build APIs, understand best practices, and understand developers. I've seen it influence people's decisions. For example, you tell someone in a space, "Oh look at this. Company A launched a travel API." Their competitors are going to take notice because they read, in an independent source, that they launched an API and they're finding success. So a lot of this competitive approach to storytelling really gets the juices flowing.
At the federal agency level, I really wanted to track who was implementing the digital strategy and the programmatic guidelines it sets forth. I'm able to discover which agencies have deployed APIs and what their plans are for APIs. So I built a tool to showcase it, in the hopes that, by showcasing the top 20 that are doing cool things, NASA, FCC, Department of Commerce, Department of Labor, and the others will say, "Oh. OK, I get this." They understand it in a context that makes sense.
Hopefully they'll start using Twitter more, or at least launch one dataset as an API, or throw one hackathon—just one baby step towards understanding this a little bit more. I think that process is critical to get to a point where all 200-plus agencies are machine readable by default as an output. Rather than PDFs, they're outputting JSON and API, and changing the game.
API Economist: John Musser's recent presentation at the API Strategy and Practice Conference focused on API business models. In particular, I was interested in his point that an API strategy is not an API business model. What does he mean by this?
Kin Lane: Just because you build an API and a platform to support it doesn’t mean you’ve figured out how to make money. You better have a clear business model for your API. What John's trying to do, by walking us through all the different possible scenarios, is bring to light that you should be thinking about a business model that adds value and can be successful. And that the API just comes from that. You should really have a strategy for building a business and being successful in adding value in some way.
You shouldn't focus on just having an API. You shouldn't just have an API strategy. You should have a business model and then have as part of that strategy an API.
API Economist: What excites you the most about the future of APIs?
Kin Lane: It would have to be in the realm of the Internet of Things. I'm talking about environmental automation: home, business, and buildings. I'm talking about 3D printing. It’s basically everywhere the API and digital are going to meet the real world. The potential for transformation there is going to be mind blowing. One of the keynotes at the API Strategy Conference was Jeff Meisel of LabVIEW Tools, from National Instruments. It’s amazing when you hear the scope of what they're doing: providing a system design tool used by hundreds of thousands of engineers and scientists worldwide. They are the API core for diverse applications ranging from Lego Mindstorms to the SpaceX control center.
It’s truly incredible to think of the potential that the Internet of Things holds. The best case scenario that I have for the physical world and APIs meeting and changing how things are done is the parking meter.
When you think about what's going on with the parking meter, paying by a credit card was the first step. But now you can up your parking while you're still in the restaurant using your mobile phone. Some places, you can see what spots are open based upon sensors reporting whether there's a car parked there or not. So I am very excited about physical real world objects being API-enabled and allowing new types of interaction and new types of data that will generate analysis and understanding of the Internet of Things. You apply that to health care, green energy, you name it. I can just go on and on. It gets my head spinning.
API Economist: What mobile devices do you use and what are some of your favorite apps?
Kin Lane: I have an Android Nexus tablet as one of my devices. My primary device is my iPhone. The apps on my iPhone 4 that I depend on are Evernote, Google Reader, Twitter, Foursquare, and Tapped. Tracking my check-ins and tracking my beer check-ins are very critical to me. But I'm in the process of switching. I don't think I'm going to go to the iPhone next time we talk.
API Economist: Kin, thanks for your time!
Kin Lane: Thank you!