DevOps Practice

DevOps Predictions for 2018

The new year is upon us, and that can only mean one thing: outrageous tech predictions! This year, we have a few that are easy to believe and a few that are completely off-the-wall. See if you can pick them out.

Serverless Overtakes Containers

Serverless is the future and containers are dead. Some (or maybe even a majority) of the industry will come to realize this. Serverless is largely defined as using functions as a service (FaaS) as the backbone of your architecture. These services are a layer on top of containers that provide an opinionated way to run code that scales with demand and ties true compute usage to costs at sub-second billing increments. Much like virtualization in the data center was dwarfed by cloud providers doing opinionated virtualization a decade ago, serverless offerings will overtake the container market.

DevOps Backlash Gets Loud

DevOps has been happily spreading through the industry as an extension of Agile to the rest of IT. But in 2018, the pushback comes. In the early days of DevOps, there was a movement called “No-Ops.” It was championed by some, but then later, reversed back to DevOps. This year, we will see stories of failed digital transformations and DevOps attempts at conferences across the globe. At first, they will be marked as “not doing DevOps right,” but as more and more happen, they will be branded with the old-new stigma of No-Ops—and they won’t come back.

Chaos Engineering Overtakes the Airwaves

You remember the Chaos Monkey Netflix released to the IT world that captured the hearts and minds of software engineers everywhere? Well, this is the year when chaos reigns! We’ve already had a taste of it through the new open-source project ChaoSlingr on the Modern Security Series show at Signal Sciences, but my guess is that we are far from done. With the release of Nassim Taleb’s book “Antifragile” in 2014, this has been brewing for a few years. Expect to see more chaos-themed engineering books, conference talks and podcasts in the coming year.

Software Flaws Mean Prison Time

Unit testing used to seem like such a good idea, but this year it’s going to become illegal to release software without testing rigor applied. Congress has been on the edge of threatening software accountability legislation for years now, but at this point it looks unavoidable. Will this be the year that a landmark legal battle happens and an engineer who didn’t do proper testing gets thrown in jail? Don’t laugh. We have already seen the first signs of this with the VW diesel scandal; just imagine if it was a pacemaker failure. In 2018, lawyers make the new rules for software engineering.

Promise Theory Finds New Adoption

Mark Burgess brought physics to IT through his groundbreaking book on Promise Theory and its implications for the way our applications and infrastructure will be built over the coming years. Instead of making assertions of obligation from machine to machine, networks will be composed of voluntary agents making promises. At the time it was released, Promise Theory received a lukewarm reception. But with the new services-based architectures such as serverless and even the rise of container orchestration platforms, the industry is now ready to adopt Promise Theory thinking. Networks, applications and systems are already functioning as voluntary agents working together through promises. This year, we will see more and more mainstream adoption of the concepts and more examples of how to get it right. Get ready to learn how systems can be self-organizing.

Application Security Becomes More About the Application

The perennial web application security checklist is to avoid XSS, SQLi and all the rest of the OWASP Top 10. Application security will finally shift from language and framework vulnerabilities to more conversations around how applications are getting attacked. In the new year, the conversation hinges on account takeovers, fraud detection and abuse inside the application. This is good for the industry because we will move past fighting automatic scanning by adding real instrumentation to detect bad actors.

Happy New Year!

The next 12 months are certainly going to be an adventure, and I hope you have a happy new year.

P.S. Please don’t get thrown in jail!

James Wickett

James Wickett

James Wickett is the Head of Research at Signal Sciences, a web protection platform that high performing DevOps teams love. He is the author of the most popular courses on DevOps topics in the Lynda.com and LinkedIn Learning platforms. James lives in Texas and has helped run DevOps Days Austin for the last six years. In his spare time he is trying to make a perfect BBQ brisket. Follow him on twitter: @wickett

Recent Posts

IBM Confirms: It’s Buying HashiCorp

Everyone knew HashiCorp was attempting to find a buyer. Few suspected it would be IBM.

10 hours ago

Embrace Adds Support for OpenTelemetry to Instrument Mobile Applications

Embrace revealed today it is adding support for open source OpenTelemetry agent software to its software development kits (SDKs) that…

18 hours ago

Paying Your Dues

TANSTAAFL, ya know?

20 hours ago

AIOps Success Requires Synthetic Internet Telemetry Data

The data used to train AI models needs to reflect the production environments where applications are deployed.

2 days ago

Five Great DevOps Jobs Opportunities

Looking for a DevOps job? Look at these openings at NBC Universal, BAE, UBS, and other companies with three-letter abbreviations.

3 days ago

Tricentis Taps Generative AI to Automate Application Testing

Tricentis is adding AI assistants to make it simpler for DevOps teams to create tests.

5 days ago