The API Economy is a phenomenon that is starting to be covered widely in technology circles and spreading well beyond, with many companies now investing in API powered business strategies. There are also a number of definitions of the term API Economy that are useful (including here and here). As the term catches hold however, it makes sense to step back and reflect what the API Economy actually is and what effect it might have on organizations, individuals and the Internet/Web as a whole.
To do this, we’ve tried to describe the nature of the API Economy in the form of five axioms we believe are true and a number of conclusions that can be drawn from these axioms. We’ll be publishing the axioms and our conclusions in a series of posts here to try to make them more digestible. The thoughts here are somewhat raw so we’re hoping for feedback and debate. The axioms and conclusions are a result of discussions between Craig Burton, Steven Willmott and others.
While producing a set of Axioms may seem a rather theoretical exercise, we believe it is a useful endeavor since, when talking about something as complex as the interactions between many new types of service providers and consumers, solid foundations are very valuable. The API Economy is already turning out to be clearly different from the human powered “Web Economy” we see today – transactions are happening faster and at greater scale, but they are distributed differently.
We begin with an overview of the API Economy, name the five Axioms and finish with the details of the first Axiom. The remaining Axioms and what we can derive from them will be posted here in the next few weeks.
The API Economy
As software and technology become ubiquitous in today’s business processes, the means by which organizations, partners and customers interface with this software and technology is becoming a critical differentiator. The Application Programming Interfaces (APIs) that form these interfaces are the means to fuel internal innovation and collaboration, reach new customers, extend products and create vibrant partner ecosystems. The shift in the way business works this creates is giving rise to huge new opportunities for both individual organizations and global commerce – it is difficult to understate the importance of this change.
Added to this, the computing trifecta of mobile, social, and cloud computing is changing the landscape of computing and business. The combination is having a huge impact on the nature of data and control flows between servers, resources, devices, products, partners and customers. Many previous notions of how an organization might set up its internal systems or deliver services / products to its customers are having to change, with APIs becoming the what enable an organization to define how data passes between their internal systems, to and from their partners and to / from their customers.
Lastly, and most importantly from the API Economy perspective, it is becoming obvious that the APIs deployed by one company do not just affect its own pre-existing internal teams, customers and partners. Very often APIs open the door to large new product, partner and customer opportunities. When deployed, these APIs then also become part of a wider ecosystem of available building blocks for other organizations to build new exciting services on top of – many with, in turn, their own APIs.
As a result of these shifts, it is critical that organizations understand the emerging API Economy, its implications and what actions should be taken to embrace its rapid emergence.
The Five Axioms of the API Economy
To try to capture the dynamics of this world, we’re starting with five axiom we think are true today or rapidly becoming true. Things which underpin the notion of the API Economy
- Everything and everyone will be API enabled.
- APIs are core to every cloud, social and mobile computing strategy.
- APIs are an economic imperative.
- Organizations must provide their core competence through APIs.
- Organizations must consume core competences of others through APIs.
This is the first blog post in this series and we’ll follow up regularly,with more blog posts covering axioms #2, #3, #4, #5 and then the consequences of these Axioms.
Axiom #1: Everything and Everyone will be API-enabled
In other words, every crafted object, structure, and every individual person as well as many natural objects, structures, objects and beings will become addressable by API and make use of APIs. Not only that but the infrastructure used to create and manage APIs for everything and everyone will also be API enabled.
On the surface this looks like an impossibly bold claim – why would this occur and how? On closer examination however, it becomes obvious that this is the inevitable outcome current strong technology trends.
There are three forces driving the “api-ification” of everything:
- The emergence of wireless mobile devices as the dominant form of interface technology for most computing systems.
- The explosion in connected device numbers, configurations, types and form factors – from large to incredibly tiny – driven by the Internet of Things.
- The extension and adaptation of integration technologies previously only used in large enterprises into more open forms that permit cross department and cross organizational systems integration at a fraction of the costs of previous technologies.
The first driving force means that networked communication at distance without wires has become ubiquitous in almost all modern environments. In other words devices can communicate and act at distance near seamlessly. The second force has made it possible for many computing form factors to be deployed in these wireless environments – some in the form of large objects such as televisions or cars, some in powerful small mobile devices such as phones, others as tiny passive sensors or identifiers. The last force is the increasing ease of software-to-software communication between systems operating on these diverse devices and/or on remote servers many miles away.
Devices tie into the identities of individuals and objects, sometimes even just to physical spaces and are able to transmit and receive data. This data in turn permits not only information capture, but also control of actions to be taken (for example thermostat to automatically change a temperature setting using settings inferred on a cloud server from data captured the previous day or the proximity to home of a smartphone using owner).
Each compute enabled device, each software application running on such a device or a remote server, and each networked environment provides interfaces to the outside world that can be identified and accessed.
Mobile and Wireless Computing
Data about mobile and wireless growth abounds, for example an article in Forbes by Cheryl Snapp Conner provides shows what is happening in the mobile segment of the market.
- There are 6.8 billion people on the planet at present. 4 billion own mobile phones. (But only 3.5 billion use a toothbrush. Oy!) (Source: 60SecondMarketer.com).
- Twenty five percent of Americans use only mobile devices to access the Internet. (Source: GoMoNews.com)
- There are 5x as many cellphones in the world as PCs. (Source: ImpigoMobile).
- 82 percent of U.S. adults own a cellphone. (Source: Pew Reports 2010).
- There are 271 million mobile subscribers in the U.S. alone.
Gartner also recently published these numbers.
World Wide Devices Shipments by Segment (Thousands of Units):
These numbers are staggering: 2.4 billion devices in 2013 and 2.5 billion in 2014. Gartner also predicts that there will be over 81 billion apps downloaded in 2013.
In terms of raw volume of wirelessly network devices, we are clearly well past the tipping point of adoption and on the road to near ubiquitous deployment of such technology.
A Myriad of Devices
High mobile computing growth is one of the most important forces driving growth in both wireless networking and devices. As the data indicates, the mobile device is rapidly becoming the preferred entry point to information and the Internet. In today’s world, information rarely remains static, it is in a state of constant change. People use many tens of mobile apps to keep track of and manage all sorts of relevant information for their day-to-day routines. This information includes location, intention, scheduling, contacts, meetings, business, context, shopping and relationships to name but a few. This information is always changing, using mobile software apps on mobile devices to track and manage these changes.
However, smart phones are only the tip of the Iceberg in terms of what computing devices are being distributed in the world. The number of devices that will be a part of the Internet of Things will be even greater. These numbers do not include the devices shown in table 1 part but are “things” that will be connected to the Internet. These devices have form factors of everything from lightweight plastic fitness trackers such as the Fitbit, to Low Power Bluetooth beacons such as the Estimote. Many everyday objects now come with sensors and networking capability embedded. Sensor density is also rapidly increasing with compact gadgets such as the wireless Quirky Sensor now packing Humidity, Light, Sound, Motion sensors into a package for under $50.
In addition the new “hobbyists and maker community” — driven by the Raspberry Pi, Arduino and other high function and affordable micro-controller devices.
Cisco predicts that by 2020 there will be 50 billion devices connected to the Internet of Things.
This is a huge number of potential compute enabled endpoints.
Since its emergence and mass adoption, interactions on the World Wide Web have been fundamentally human driven. Human users browse information, upload content, download and browse data, click controls in order to generate effects: both virtual (such as a video playing) and physical (such as a purchased book being shipped from a warehouse). This setup enabled a huge diversity of applications since on the basis of a few simple rules and patterns a vast number of different services can be built and human users were flexible enough to understand the purpose of the sites/applications. They can then act accordingly to operate the service.
Today however, human driven Web Activity is rapidly being caught up by software-to-software activity: traffic between systems with no human in the loop and automated transfer of data and control actions. This is occurring because:
- Mobile devices are increasingly acting autonomously and pre-emptively to fetch data from and deliver data to remote systems.
- Many new device form factors include no keyboards, screens or other means of data input/output – and are hence their onboard software is only able to communicate with other specialized software on other devices or on remote servers.
- Many modern applications require high rates of highly accurate transaction flow – with millions of similar calls per day which must be carried out stably and efficiently in a way no human could drive.
- Almost no modern software application developed today – Web Applications and those running locally on a device – functions in a completely standalone manner any more. That is, it is very often the case that some part of the functionality requires direct, coordinated communication with some other piece of software.
Why APIs will be Ubiquitous
As these trends continue they all continually reinforce the need for APIs for every device and person. For devices, almost all devices will require:
- A hardware substrate of some kind.
- Control software.
- An interface by which the device can be reached.
- A number of remote interfaces the device may call from time to time.
- One or more identities by which the entity can be addressed.
For a natural person (or potentially another creation or natural object), they may become associated with one or more digital devices that can receive incoming messages and conversely send data to one or more remote systems.
Existing identity mechanisms such as email addresses, twitter handles, phone numbers and the like already permit data exchange and as devices become more customized and embedded about the person, the ability for an individual to receive information (which is some cases will cause them to act) and stream data to a remote location will increase.
It is hard to imagine, how with this growth we will not enter a situation where almost every individual and object will have some kind of addressable set of interfaces associated with it.
As we’ve already pointed out the claim that everything and everyone will be API enabled is mind-boggling within itself – at least it seems impossibly bold. However as we can see above, it seems inevitable as we move down the path of universal connectivity collaboration interoperability that this example come to pass. That is not to say there are not huge obstacles in accomplishing this grand vision but it seems nonetheless this is the path we are on.
Axiom number 2 “APIs are core to every cloud, social and mobile computing strategy” is up next.