What's the Future of Integration and APIs?
More systems are being connected more seamlessly thanks to the integration of artificial intelligence (AI) and machine learning (ML).
Join the DZone community and get the full member experience.
Join For FreeTo gather insights on the state of integration, API design, and API management, we spoke with 19 executives who are familiar with integration and APIs and asked them, "What’s the future of integration from your point of view - where do the greatest opportunities lie?"
Here's what they told us:
More Connections
- The opportunities that arise when two or more companies are serious about collaboration and working together to build a solution and are willing to brainstorm a solution that combines the expertise of both. Co-created integration is a mindset that starts creating value that blows customers’ minds. Look at the end user and how they integrate. Go upstream with customers’ partners and partner with them.
- Integration and APIs are designed to make life easier. There will be more intelligent systems ingesting data in real time. Make the functionality available to all different verticals. Take in metadata to measure real-time risk and make informed decisions with regards to security and orchestration.
- This is a rapidly changing space, but I’m convinced that real-time streaming data flows are going to be at the center of innovation in integrations for the foreseeable future. Core business logic is meshing more and more with analytics and business intelligence concerns, and the demand for in-the-moment information and insight is only going to grow.
- Transition around API base for system-to-system communication. Provide 100% visibility of the ecosystem to see the data and the explosion of the lines of business. Change to best of breed cloud-based apps. Our clients have the goal to be 100% cloud-based – even oil and gas.
- An explosion of APIs to change the face of integration. API management is a feature of integration. Management is just one component. Who builds it? How to create APIs? If you know where the data is, we can access it, create the API and URL.
- Integrations will start to become more multi-faceted. In other words, an integration between two SaaS providers won’t meet enterprise customers' needs in the long run. Rather, SaaS products need to play together in a kind of mesh network. This involves not only content and data flowing between systems, but also actions being triggered across different providers in more than a linear fashion. There is a lot of complexity from a systems perspective that needs to be presented to users in a straightforward way.
- A major opportunity is where businesses could bring their legacy systems and applications into the API mold so they can integrate with modern applications. This can be done by building API bridge layers to translate modern API calls into legacy calls. This ensures the investments done on those apps are not unduly wasted. The other opportunity is where we have the likes of Microsoft Flow, which acts as a broker/proxy API layer. This way, individual apps don’t have to worry about integrating with each other. Rather, they all integrate and publish their APIs with the Flow service, which then enables integration between any two participating apps seamlessly.
- Integration platform where every vendor has the option to write their own plugin. A lot more stuff is becoming open. Authentication and authorization is an essential part of the application plugin and only does what it needs to do. No tokenization, authentication, authorization. More simplified across multiple platforms results in more integration.
AI/ML
- Less connected service systems. We’ll use AI/ML to predict future integration needs and they’ll produce more efficient connections.
- Concepting around microservices. It’s grown a lot in the last two years and adds a lot of complexity. Another area of growth is function as a service or serverless (Lambda). Run in an AWS environment without worrying about the underlying virtual machine. Serverless will hide the complexity of the integration. API gateway links act like a firewall with fine grain services. How do you secure communications between services? You will need to microgate ways to provide functionality in the API gateway. What tax are you willing to pay for core functionality?
- Self-service, low code, no code. Flexibility for developers to jump in and out of code. Citizen developers can develop and publish. Reduce the number of steps needed to get projects done.
- How APIs enable automation and artificial intelligence (AI). More devices are talking to each other. APIs enable this communication to occur. The more hyper-connected economy requires developers to look at APIs from a use case perspective.
- Simplifying integration of on-prem, hybrid, and cloud environments. Will make gains with AI/ML with self-defining and self-healing integration. A lot to be gained through crowdsourcing especially in mapping data. We’ll see six to eight times greater productivity. In the discovery of data for processing, AI/ML will provide recommendations for the next best objects. Apiary.io designs APIs correctly for net new revenue, net new experience, and operational efficiency. There’s a need for API as design. Design correctly first to see the net effect.
- The real benefactors of integration are the users, who get a more robust experience as more services are integrated. The greatest opportunity lies in creating seamless, expansive experiences that we can enjoy from one interface. With the explosion in machine learning, we should expect to see a massive growth in artificial intelligence services. Application developers will be able to exploit the complex logic and large datasets from these providers – letting them build apps they could never have imagined.
Other
- Figure out how to test from a security aspect. API has dynamic analysis. UI is built with API authentication. Look at the results. Ensure things are in the right order. Automated security for APIs.
- To me, normalization and/or abstraction would be a huge step to API adoption. If you could skip the cumbersome authentication processes (dealing with tokens expiration, endpoint changing, small variations on OAuth implementation...) and normalization/mapping and talk to a single third party to access any API, API integration would be way easier. It's something Samsung is doing, for instance, with Artik cloud even though it's more oriented toward devices.
- The shift to the cloud necessitates a reassessment of data and application integration strategies for on-premises and cloud applications and a new approach to how integration is delivered. The nature of the systems and applications to be integrated demands a hybrid integration strategy that supports ongoing integration as businesses transition their applications to the cloud. 1) Lines of businesses (LOBs) that adopt cloud applications do so with an expectation that they will obtain more control, agility, and flexibility than they’ve had in the past with on-premises applications deployed or built by IT. They also expect the same kind of user experience and access to information from applications, whether these reside on-premises or in the cloud. 2) Any approach needs to account for how data is shared and applications are integrated and make it possible for a variety of cloud and existing on-premises applications to coexist. It is also important that applications can share information and data is kept consistent whether accessed from the cloud or on-premises. 3) What is really required is the development of a new paradigm that combines data integration strategy with application integration strategy – the what and how of application integration regardless of where applications reside. This new approach must address the need to relocate on-premises applications to the cloud over time without incurring penalties during the transition. The integration platform and capabilities employed to accomplish this need to be hybrid in nature.
What's your vision for the future of integration, API design, or API management?
Here’s who we talked to:
- Murali Palanisamy, E.V.P., Chief Product Officer, AppViewX
- Kevin Fealey, Director of Automation and Integration Services, Aspect Security
- Max Mancini, VP of Ecosystem, Atlassian
- Shawn Ryan, V.P. Product Marketing, Digital as a Service, Axway
- Parthiv Patel, Technical Marketing Manager, Built.io
- Chaitanya Gupta, CTO, Flock
- Anwesa Chatterjee, Director of Product Marketing, Informatica Cloud
- Simon Peel, CMO, Jitterbit
- Keoki Andrus, VP of Products and Steve Bunch, Product Manager APIs and Integrations, Jive
- Rajesh Ganesan, Director of Product Management, ManageEngine
- Brooks Crichlow, Vice President, Product Marketing, MongoDB
- Derek Smith, CEO, Naveego
- Guillaume Lo Re, Senior Software Engineer, Netvibes
- Vikas Anand, V.P. Product Management and Strategy – Integration, Oracle
- Keshav Vasudevan, Product Marketing, SmartBear
- Kevin Bohan, Director of Product Marketing Manager, TIBCO
- Pete Chestna, Director of Developer Engagement, Veracode
- Milt Reder, V.P. of Engineering, Yet Analytics
Opinions expressed by DZone contributors are their own.
Comments