Happy holidays. At Pivotal, our mission is to transform how the world builds software. Every year we bring together Pivotal experts, customers, and leading technologists to forecast transformative trends for the New Year. We hope these insights will help readers like you stay ahead of the curve. Enjoy!
—Rob Mee, CEO, Pivotal
The Lean, User-Centered Design, and Agile/XP communities will continue to converge. Lines will increasingly blur; practices will be de-dogmatized, in favor of principles and values. The role of the Product Manager as "mini-CEO of the team" will continue to wane, replaced by Balanced Teams. Collective thinking, and thus innovation, will rise accordingly. Psychological Safety exercises, tests, and consulting will balloon in 2019, as agile teams embrace psychological safety as a fundamental prerequisite to success. SAFe will continue to be sought out by enterprises looking for a quick, ready-made agile transformation solution—but they will continue to wrestle with sustainably delivering value to their users.
—Matthew Parker, Head of Engineering, Pivotal Labs
Applied Artificial Intelligence (AI) and Machine Learning (ML)
There is no shortage of hype about artificial intelligence, but we’re still in the early stages of AI deployment in the enterprise. This is especially true when it comes to deep learning, barely 5 years into the current explosion. The AI realist, who should become more vocal in 2019, might ask: “That’s an impressive demo of image recognition, but how does it translate into helping my business?” AI projects will require different implementations, different data (a lot of it), and different skill-sets for which companies will have to aggressively hire. Fortunately, help is coming in the form of better software products and services to make AI more accessible to the enterprise. One example is tooling for model management, where many of the same patterns from building and deploying complex applications and microservices apply. Startups in this area will continue to attract funding in 2019, as well as those applying AI to specific use cases within verticals.
—Frank McQuillan, Director of Product Management on Applied AI & Machine Learning, Pivotal
Blockchain will fall short in financial services but be utilized more in supply chain tracking in 2019. The initial use cases in Blockchain were heavily focused within financial services and have yet to take off. Currently, there are large profit margins made by intermediaries of financial transactions, i.e. custodians, as well as heavy regulation introduced by Dodd-Frank demanding central exchanges to be intermediaries for control. Until the value chain is disrupted completely and/or regulation adjusts its view, the sector will continue to stall in the area of blockchain. The supply chain tracking use case is becoming the more relevant use of blockchain and will speed up in progress in 2019 because all parties benefit from the utilization of the distributed ledger and immutable transaction benefits provided by blockchain. Consumers can guarantee that what they order is what they receive and sellers of the final product can identify down to the individual item the origination and path the item takes. In the recent lettuce recall situation, grocery stores could have barred only the specific farms that were infected by the recall and not everything all at once, whereby avoiding loss of sales and revenue.
—Jesse Bean, CIO of the Field Services Group, Pivotal
“... We have been using a centralized model for years, and at MasterCard, we have been following what they call the ‘zero trust model.’ We've been doing it for years, and over time, the only way we thought to be most effective was having centralized security controls, and we had been trying to build that into the containers and build it into the network level and into every single component that we can touch. But, over time, it generates a lot of complexities, and it can be hard to figure out where the problem is. And so it's very difficult. So, now we're kind of shifting the models more toward the platform-managed instead of container-managed. It's not really the container; it's more like the application server at that point, but now we're trying to push that functionality into the platform.”
—Jenny Zhang, Principal Consultant, Mastercard
This text was taken from an interview Pivotal did with Jenny at SpringOne 2018. Watch it here.
Data continues to be one of the most important assets that an organization has at its disposal, and the ease with which insights around it can occur will rapidly accelerate in 2019. New application development will really start to involve artificial intelligence and machine learning co-developers, thanks in large part to the continued trend of decoupled microservices, which will finally start the realization of “smart applications.” This will also mean that the experiment in data lakes will come to a conclusion for many organizations as they switch back to familiar products and platforms that are either open source or managed by an IaaS or PasS in order to show some actual ROI. All of this points to an aggressive demand for qualified and experienced data engineers and data scientists, who will be critical in this next wave of innovation.
—Jacque Istok, Head of Data, Pivotal
In 2019 organizations will invest more in design leadership and we will see more Chief Design Officer and VP Design appointments across industries. This underpins their transformation into customer-centric product organisations and follows a trend to bring design capabilities in-house. As design practices mature they need appropriate executive leadership support to continue to grow the breadth and depth of their practice, scale to match the need of the organization, develop a core design ops capability, and keep hard-to-find talent engaged.
—Martina Hodges-Schell, Head of Product Design, Pivotal Labs on Design
“... We want to enable our developers as much as possible and give them choice of tools where appropriate, but we have to strike that balance between compliance and efficiency. There are lots of tools that the developers are able to choose from on their desktop and within their teams, from team to team, but when it comes to something like a CI/CD pipeline, that remains within a centralized team. But one of the things that we're doing now is pairing up those development teams with that centralized team when we make changes or enhancements to make sure that we're getting real-time feedback and making decisions that will benefit the development teams.”
—Kurt Glore, Director, Cloud and Deliver Engineering, Express Scripts
This text was taken from an interview Pivotal did with Kurt at SpringOne 2018. Watch it here.
IT Modernization / Replatforming
In 2019 I expect to see IT organizations modernize larger (like “system of systems”) and more diverse (like Mainframe) workloads to our rapidly evolving set of cloud abstractions. Doing this will require good decision making that balances technical suitability (the "what"), business criteria (the "why"), and organizational (the "how") factors aligned with a set of strategies (like Re-Host or Re-Factor) that deliver immediate results (like better security posture). These teams will work incrementally and move as many applications into a full production state as possible. By reducing the operational burden of their existing portfolio, organizations will unlock precious dollars to improve customer experience and drive innovation forward.
—Edward Hieatt, Senior Vice President, Customer Success, Pivotal
2019 will be the year of PaaS and CaaS convergence. Enterprise IT’s accelerating appetite for containerization of applications is going to take Kubernetes further up the stack and the blending of capabilities from PaaS environments will commence. I expect many modernization efforts within enterprise IT throughout 2019 will target a Kubernetes runtime, and convergence will be welcome. A single environment that accommodates new cloud-native applications and provides more control over traditional workloads will appeal to many large enterprises.
—Alan Flowers, VP and CTO EMEA, HCL
“...Right now, Kubernetes is cool, teams at T-Mobile are being told to containerize their applications, so for a lot of them, that looks like PKS. We're trying to make sure that they're placing themselves there appropriately and not just doing what is the cool thing, but is the right thing for their applications. If they don't need to maintain the CI/CD for that docker image for the next two or three or five years, they're much better off just trusting the build pack and letting that take care of the problems for them.”
—Brendan Aye, Cloud Foundry Platform Architect, T-Mobile
This text was taken from an interview Pivotal did with Brendan at SpringOne Platform 2018. Watch it here.
Some would argue that Kubernetes has crossed the chasm, as evidenced by 8,000 people attending the final Kubecon of the year in Seattle in December. And while the conference saw a marked increase in presentations coming from users (rather than only from open source contributors and vendors), most of these still came from cloud-native companies—those that were born in the web era (say within the last 10 years). Most of the enthusiasm around Kubernetes is still coming from the developer—the consumer of Kubernetes. 2019 is the year that Kubernetes will cross the chasm in the traditional enterprise. Early adopters are seeing value coming from their use of Kubernetes and are beginning to speak publicly about it. And the people who provide infrastructure to those developers asking for Kubernetes are actively looking for ways to offer that service. Of course, security and compliance reign supreme in the enterprise; and Kubernetes has matured to the point where it is possible to meet these requirements, however, the level of complexity remains a challenge. 2019 will see emphasis placed on making Kubernetes more accessible to exponentially more consumers—this will truly enable the crossing of the chasm.
—Cornelia Davis, Senior Director of Technology, Pivotal
“... Everyone has a mobile device now. Smartphones are ubiquitous. Having the ability to take a native mobile client and to connect it into the same server that you might use to present your website was absolutely critical. The next logical step at Merrill as we were thinking about what should we do, was microservices, because that's really my view of the best way to handle complexity in today's world. What it means to me is you can split parts of your system in these swim lanes, to make sure that there's isolation. In the old fashion sort of consolidated monolithic systems, if a reporting job sucked up a lot of database CPU, you might impact people that are doing something totally different on the database, and that's not good. Now, of course, you can try and engineer around it, but if you keep everything separate within a microservice from the get-go you're built with that isolation, you're built with that sort of stability in mind. All things still need to scale at the same rate, so if you can split things into these segments, into these microservers, you can scale as appropriate, and add more resources where necessary. It keeps the code base small, so the smaller things are the easier they are to understand, the more likely that you're going to be able to maintain and support it and grow it over time.”
—Thomas Fredell, Chief Product Officer, Merrill Corporation
This text was taken from an interview Pivotal did with Thomas at SpringOne Platform 2018. Watch it here.
Consolidation around open source software will continue into 2019. In late 2018, IBM and VMware announced acquisitions of Red Hat and Heptio, respectively; and Microsoft acquired GitHub, which hosts millions of open source projects. In addition to more acquisitions, additional open source projects will consolidate under foundations. This is a continuation of what we have recently seen with many open source projects, including the Ceph Foundation, moving under the Linux Foundation, for example. Additional jockeying for position within the open source Kubernetes project is likely to continue as part of this consolidation in 2019. The acquisitions of Red Hat with their OpenShift Kubernetes platform and Kubernetes company, Heptio, strengthen IBM’s and VMware’s already strong positions in Kubernetes. More of these Kubernetes-related acquisitions are expected to continue in the future.
—Dawn Foster, Open Source Software Strategy Lead, Pivotal.
“...We're completely invested in growing technology at DICK'S Sporting Goods. We've touched on it a little bit earlier, but the concept of the balanced team within each of the product teams is so critical and that's something that we're going to continue to focus on. I have a great partnership with the product manager inside the organization and the design side of the organization and leveraging the experience of our teams working in the immersive experience of Pivotal Labs as well as the teams that we have onsite, really looking to grow 31 product teams that we have, the next 30 after that.”
—J.P. White, Director of eCommerce Application Development, DICK’S Sporting Goods
This text was taken from an interview Pivotal did with J.P at SpringOne Platform 2018. Watch it here.
“...Safety is paramount for us. We want to make sure that we are building products that are doing exactly what they're supposed to do, no fuss, no fray on that. That also makes us inherently a very, very conservative company, which means we check everything once, twice, three times, four times. Then we have other people that check our things as well... We're a 102-year-old company that worked very well for us for the first 102 years. That is not going to work well for us going forward. We need to fundamentally make some changes, whether it's changing the way we do software development, the way we want to actually think about things in order to be able to come to market it a lot faster. We have to inherently change, which also means that we need to move faster without compromising the safety… The idea of test-driven-development, the idea of us sort of building in those safety checks along the way, really kind of helps to answer that question of how do we continue to build safe products that we want to have out there? I think there's always been this idea that we need to be able to move faster, we need to build to provide value to our end customers, but what's nice is now we have a framework that we can actually use to be able to go forward and have that conversations. That's been really powerful.”
—Sophia Bright, IT Director, Boeing
This text was taken from a SpringOne Platform presentation. Watch it here.
“So, within West we are starting to see security teams becoming more of a partner to the product teams. We think about it as going from governance to guidance, or being much more a part of that team, so if you start to see the evolution of things like DevSecOps and things like that, that's really where we start to see it. Our relationship with a security organization is that if we can standardize things within the platform as much as possible, that gives us an opportunity to be a lot more consistent, with the right way also being the easy way.”
—Thomas Squeo, CTO, West Corp.
This text was taken from an interview Pivotal did with Thomas at SpringOne 2018. Watch it here.
“... If you think you're going to patch weekly or patch daily or repave daily and you're not going to do it through automation, you are sorely going to have thousands of people, which is going to introduce human error…. Stop treating your servers like it's something that you want to stick around. Treat them like they're ephemeral and immutable. And then, redeploy often, even when you don't think you have to or need to. It's just good practice. Do it as often as possible… We probably have in the neighborhood of 15 to 18,000 virtual appliances that are spun up at any given time in all of our PCF environments… when you deploy an application and that application has a number of instances, and those number of instances follow a cloud-native practice, that they have the ability to be patched, the platform be patched without causing customer impact. We've seen this since we do it once a week right now, and we're going to be going to once a day by the end of next year.”
—Lance Rochelle, Product Manager, Wells Fargo
This text was taken from an interview Pivotal did with Lance at SpringOne 2018. Watch it here.
The terms "serverless" and "FaaS" (Functions-as-a-Service) are often used interchangeably. However, the characteristics of serverless are more generally beneficial, allowing developers to be more productive since they can focus on code that solves business problems rather than "server" concerns such as load-balancing and scaling deployments to zero and back. The specific benefit of a FaaS is that it narrows the developer focus to an even higher level of abstraction. When describing application frameworks, such as Spring, we refer to that as Inversion of Control: as the framework and/or platform takes control of more concerns, the developer has fewer responsibilities and thus more focus. It also means the platform can provide and patch even more of the stack, such as a security vulnerability in an application framework dependency. Five years ago at SpringOne 2013, Paul Maritz described IaaS as the new hardware and PaaS as the new OS. In 2019, we expect more developers to consider FaaS as the new application framework for those use cases where a single-responsibility event-driven function is a good fit, while still benefiting from "serverless" characteristics for their full-stack cloud-native applications, increasing developer productivity at any layer of abstraction.
—Mark Fisher, Senior Staff Engineer, Pivotal
As we enter 2019, the role of the software engineer continues to rapidly evolve. Today’s software engineers are being asked to do more than ever before as they manage the full application development lifecycle from end-user interviews and developing MVPs to production deployments and continuing support. While this has produced more effective, higher quality software products and increased customer satisfaction, it has also created a tighter labor market for engineers that possess the necessary skills for the evolving software engineer role. Enterprises are now investing massive amounts of time and money to attract and retain employees that are able to deliver on the promise of being a “full lifecycle” engineer.
—Ryan Johnson, Associate Director, Accenture
About the Author