Industry executives and experts share their predictions for 2022.  Read them in this 14th annual VMblog.com series exclusive.

How the way we code will change in 2022

By Brian Rue, Co-founder, and CEO of Rollbar

The way we code is going to change in some fundamental ways, all driven by one core underlying force - the need for developers to produce code faster.

Faster

The need for code to be written and released faster is already happening, but that need is going to increase. The primary driver for this is competitive pressure. Software teams need to iterate faster. It's not a choice -- teams have to go faster to survive because customer expectations require it. Things are moving to digital faster than ever, and more products and services are moving to digital. Because some teams are going to go faster, if your team doesn't, you will fall behind. Ultimately it comes down to market pull. Consumers want things to get better -- and if they don't get better, you will lose. So, every developer has to put out better code faster.

Getting code out faster will lead to a couple of other things happening. Release cycles will shorten even more, and smaller releases will become the norm.

The way to go faster in development cycles is to be more iterative. It comes down to momentum = mass x velocity. You get the best overall velocity when you do it in smaller chunks. By comparison, bigger releases are more risky and are almost always delayed. Therefore processes will reorient to be more iterative in smaller chunks. Feature flagging, faster test times and a move from manual to automated testing help advance the goal to make releases easier and faster.

Another way to support going faster is automation. A great way to do anything faster is to have it done instantly by a machine. So, we will adopt automation of any repeatable manual tasks during development. We also will see more integration so that automation is more effective. 

As code is built to operate in more complex environments and with more and more microservices engineers will work to integrate their tools and workflow so that cross team and individual communication, data and visibility is improved.

More teamwork

With better communication and tools, more code will be built by teams working together and rapidly releasing and iterating. This is in contrast to siloed efforts.

Engineers and DevOps will take advantage of new tools in the areas of observability and operations to make the SDLC more iterative as it moves faster, since new tools make it less risky to release code faster and thus include the understanding and fixing of bugs as an expected part of the cycle instead of striving for perfect releases.

This cycle and the steady increase in release times is part of what Rollbar refers to as Continuous Code Improvement. This approach embraces the move to faster releases and more iterative approaches combined with integration and automation. The same philosophy is behind Google's adoption of DORA metrics and the use of the mean time To release (MTTR) metric.

##

ABOUT THE AUTHOR

Brian is the CEO and Co-founder of Rollbar, a SF-based provider of real-time error monitoring Software as a Service, where he leads the company’s overall strategy and direction. Brian founded the company with Cory Virok in 2012. Prior to Rollbar, Brian was the CTO and Co-founder of Lolapps, a leading publisher of independent games on social networks and mobile platforms. Brian attended Stanford University where he studied Management Science and Engineering.