Posts Tagged: API
I’m fascinated by the Open Data movement, particularly when it comes to government where the concept of open data hinges on the belief that data is a public asset – just like highways and parks – and should be made available to all.
We’ve already seen compelling examples of people hacking into this available data for the public good. For example, at the recent Open Data Day here in Vancouver, projects ranged from a visualization of Vancouver’s budget to an analysis of test scores at local schools. One project even took Vancouver’s open elevation data to recreate the city in MineCraft.
Making data available, discoverable, and usable will help improve the effectiveness of governments, strengthen our democracy, as well as create opportunities to improve everyone’s quality of life. Open data encourages innovation from both the inside and outside:
- Innovation from the outside: Putting information in the hands of the public makes it possible for innovative citizens and companies to create groundbreaking solutions and improvements to some of the biggest challenges facing the community. For example, on a small scale, one website monitors and analyzes Vancouver’s bike accidents, revealing the trend that bike accidents decline along roads with bike lanes and increase where bike lanes end.
- Innovation from the inside: When government data is accessible and easily usable, other governments and government agencies are able to compare experiences and share best practices. Although government data is public today, most state and city governments do not readily share information. The Open Data movement can unlock this data, so local governments don’t necessarily have to recreate the wheel. They can look at examples and models of what has already worked.
Where are the start-ups to take on this opportunity?
Making data publicly available is just a first step. In order for the Open Data movement to truly bring about change in government, that data needs to be searchable, discoverable, and usable. But so far companies have mainly focused on helping governments transform data (usually financial data) into actionable information. For example, there’s OpenGov, Open Data Soft, and Junar.
In addition to these players, we need applications and websites that put the data into something useful for both governments and citizens – whether by visualization, integration, or notification. We need a company that can take all the existing public datasets and make them accessible and comparable through a single platform… similar to what Clever does in education. It’s an incredible opportunity to make a significant difference: the question is who will lead the way?
- Palo Alto looks to use open data to embrace ‘city as a platform’ (radar.oreilly.com)
- Open Data Day at Vancouver City Hall this Saturday (blogs.vancouversun.com)
Virtually every start-up dreams of becoming a platform at some stage. After all, it is the most powerful position in the ecosystem. What has been the one rule to becoming a platform? Develop a killer app that gets you to scale: and then open up the platform once your reach is attractive enough for other developers to develop specific apps for your platform. This game plan has worked for consumer (Facebook, Twitter) and enterprise (Salesforce, Shopify) apps alike.
But recently, we have seen a new breed of platform emerging: platforms that are targeted at developers. Their mission is simple: solve the pain of integrating all the cumbersome and fragmented legacy systems, by providing an easy-to-integrate-with and easy-to-manage frontend for developers.
For example, we see horizontal platforms like Twilio that replaces all the telecom hardware with an API for phone, VoIP, and messaging. There’s Stripe and Braintree for payment processing, and Lob for printing. This is what Garry Tan has called the API-ization of everything: “Where there is paper to push, a call to answer, or a purchase to approve, there is an API coming to replace it.”
In addition to these horizontal platforms, there are also very interesting vertical players emerging: Spout for finance, Rets.ly for real estate, and Clever for education. While horizontal developer platforms are usually built around features, vertical platforms are usually built around data.
The traditional platform approach (i.e. Facebook, Salesforce) is still alive and well, but it feels like we are on the cusp of tremendous innovation in the field of developer platforms as the web is evolving into being entirely programmable.
I recently spoke with a founder and CEO of a start-up that had just crossed the 200-employee milestone. Like so many others lucky enough to reach that level, he was complaining about how progress was coming to a halt as the organization grew and became more complex. I myself experienced this as COO at AbeBooks. It’s a cruel irony in business: the more people you add, the slower you get.
To help overcome this scale stagnation, I have three pieces of advice:
1. Overinvest in your management team
An ineffective employee is bad for the company, but the impact of an ineffective manager reaches across the organization, multiplied by the number of people he/she manages. For this reason, you need to hire the absolute best people you can get for managerial positions. Invest both money and time in developing their managerial skills, and have a strong employee feedback system in place to make sure you can identify and correct the “bad apples” right away.
2. Listen to Amazon: two-pizza teams and internal APIs
Few would argue that Amazon is one of the most entrepreneurial companies around and CEO Jeff Bezos has an uncanny ability to see patterns long before others. Two of the reasons Amazon is able to keep its momentum even as a 97,000-employee giant are pizza teams and APIs…
- “Two-pizza teams”: Bezos structured Amazon as a decentralized company where small groups can innovate on their own and be free from the inherent problems of groupthink. Company-wide, he introduced the principle of the two-pizza team. If a team can’t be fed by two pizzas, then the team is too large.
- Internal APIs: Every internal product should have an API, just as if it were developed for an external client. This decouples the speed of development between different product teams, as you can have a clean hand-off between the two.
The added bonus is that APIs lay the groundwork for the productization of internal tools. Bezos mandated that all IT assets be exposed as APIs. To quote a Forbes article, “That single, simple declaration created an IT (and cultural) architecture that catalyzed and stoked the stunning growth of Amazon Web Services, which is thought to be a billion-dollar business unit after only a few short years of growth.”
In short, small teams can run fast and innovate because of their size and the fact that they’re not reliant on the technology from other teams.
3. Focus your culture on speed over perfectionism
“Hackers try to build the best services over the long term by quickly releasing and learning from smaller iterations rather than trying to get everything right all at once…We have the words ‘Done is better than perfect’ painted on our walls to remind ourselves to always keep shipping.”
Prioritizing speed over perfection can be a very powerful way to keep momentum up as you scale, particularly as you add layer upon layer to the organizations and require multiple signatures to get things done.
One key message that can apply to any start-up is to never end a meeting without a decision (unless everyone determines that more data is needed to make that decision). If a meeting begins to succumb to groupthink and indecision, you need to ask why we can’t make a decision today instead of another day. After all, you just need to move forward.
- The 3 Books Amazon’s Jeff Bezos Asks His Senior Managers To Read (embargozone.com)