Blog archive

2016 Dev Predictions, Part 2: Cognitive Computing, IoT, Cloud, More

We're only a month into 2016 and it's already shaping up to be another lively year for enterprise developers. Mobile, cloud, DevOps, IoT, microservices, the API economy, cognitive computing, virtual reality -- all are reshaping organizations in fundamental ways, and it looks like devs are going to have a large role to play in that change.

Analyst Clive Howard, who keeps an eye on the mobile and IoT space for UK-based Creative Intellect Consulting, expects 2016 to be the year most companies figure out what all this stuff actually means for their businesses. Projects that have hovered mostly around the edges will find their way into the heart of the enterprise, and a few -- IoT, cognitive computing and cloud -- will mature as a "progressive few" organizations "begin to shape them into exciting new products that will start to appear in 2016 but really emerge in 2017+," he said.

Howard also expects IoT to continue to pull in developers, both consumer and enterprise, with the action heating up on the enterprise side. Mobile B2B and B2E (Business to Employee) will grow significantly, he said, which means and more developers within organizations will be involved in mobile—which means lots of people are going to have to buff up their skill sets.

Meanwhile, non-developers are likely to be more involved in enterprise development, Howard predicts, via so-called low-code tools and services. Given the current skills shortage, it's likely that companies will create strategies that embrace these non-developers, he said.

The coming year will not, however, see the hype around connected cars and wearables live up to the reality, he said. "I think there will be little activity [around connected cars] outside of those already involved in the car industry," he said. "Cars are probably heading to the top of the hype bubble. And wearables will go nowhere in 2016, certainly in terms of the consumer. Industrial use cases may see some interesting developments, but not at significant scale."

Another UK-based analyst, Gartner's Gary Olliffe, sees the growing interest in microservice architectures as something of a harbinger, signaling a rediscovery of the value of service orientation. "Enterprise developers are waking up to good old fashioned architecture principles that aren't new," he said. "Microservices are a kind of beacon, showing them the benefits of those principles and how they can apply them to their own work.

"Developers are excited about microservices, because they allow them to simply their development stack, chose optimal technology, and not have bloated middleware forced upon them," he added.

Microservices have gained enough mindshare that even though most organizations are not trying to, say, replicate a Netflix-style microservice architecture, they're learning from that example, and feeding that knowledge back into the business, Olliffe said. In 2016 microservices will have an impact on most enterprises delivering anything that needs to be exposed or managed as a service, he said,

One bump on this particular road: no true microservice platform. "The tools to help you get your feet wet are easily accessible, and have become more so in the past 18 months," he said. "But the vendors have yet to step up and really provide a true microservice platform for developers. We're not seeing what you might call the next generation of app servers, the platforms onto which I deploy my units and which handles all of the complexities of the outer architecture."

Microsoft's Azure Service Fabric comes closer to providing more rapid developer productivity in that environment, Olliffe said. But to actually manage and operate tens or hundreds of instances of multiple microservices in production is a whole different game, he said. The app lifecycle teams that increasingly include coders, integrators, architects, operations, and QA, have yet to really get their arms around these environments, he said.

Microservices are poised on the edge of the phase of Gartner's hype cycle known as "The Trough of Disillusionment," Olliffe said, which follows "The Peak of Inflated Expectations." During this phase, people focus on a technology's shortcomings and limitations, and a few products fail. But he expects developers to dig in and figure out what really works, prompting speedy progress to the "Slope of Enlightenment," followed by the "Plateau of Productivity."

You can read more about microservices in Part 1 of this series ("2016 Dev Predictions, Part 1: DevOps, APIs, Microservices, More"), and I'll soon share additional observations about the year ahead in Part 3 ("2016 Dev Predictions, Part 3: Mainstream Microservices, Reactive Streams and Containers-as-a-Service.")

Posted by John K. Waters on January 29, 2016