Exploring through forums brought me to a question I had heard before; is programming in your 40s a dead end? Over the course of a career, do programmers have to choose between a move to management or becoming outdated resources?

As programmers and software engineers, the only constant in the field is change. Good code doesn’t necessarily change often, but the tools and times always do and programmers need to take account to stay relevant in the field. Just as doctors constantly look for more effective, safer, or quick procedures to treat illness, engineers must stay at the edge of technology to keep their solutions viable as the field and their careers keep moving.

To me, this means that being a software engineer is a never-ending pursuit to understand how the field evolves, and figuring out which of those changes are impactful.

Without this ambition to stay ahead of the curve, engineers may find themselves being pushed in a direction they might not appreciate; however, many find this effort exhausting and might find it more comfortable to dive deeply into research or academia. Even then, not understanding how to keep up with any field is programmer’s death sentence.

Naturally, what I wanted to know was "how can we keep growing as developers?". And this lead me to wonder where anyone would start.

How can we stay aware of the field?

Staying on the edge requires commitment and passion! Building on that commitment, there’s a few resources that are worth checking out habitually. Check the end of the post for a few suggestions.

The list gives a couple of places to start on some broad subjects, but there’s always the more focused newsletters on useful products or even discussions/news on the programming subreddit.

With all the awareness in the world, however, there’s no replacement for getting hands-on with new tools and technology. Depending on time and interest, that can be as simple as a hello world and as complex as an entire project from scratch. It’s happened more than once where learning a new technology at work coincided with a hobby project; to prepare for a work project, putting together a home project in Vue.js is infinitely more helpful than just reading articles about it.

The only way to learn a new programming language is by writing programs in it.” - Dennis Ritchie

Recognizing value and separating out fads

With all the new technologies that emerge every day, there’s just as many fads and marketing gimmicks as there are useful or practical tools. Many might remember the blockchain craze, and even companies going up in value for just having “blockchain” in the name.

Even if every project was interesting and useful, it’d be practically impossible to implement more than a handful into a project.

This can make it a challenge to sort out what’s going to help your projects vs what’s going to waste your time. Anyone can wait for the big companies to pick up new tech and catapult it to the spotlight, but for everyone else, we should start by asking is this technology doing something better than a competitor, or is it promising something new?. Also keep in mind, the old adage “if it’s too good to be true, it probably is” comes into play.

If the technology is delivers niche or new functionality, it’s important to ask is this solving an important problem?

If the technology is a new attempt to compete against an existing product

  • How is it better than competing projects?
  • Does it have any caveats or require any additional services compared to competing projects?

And if you have a project that could benefit from the improvements:

  • What’s the cost of refactoring? How much has to be redesigned to take advantage?
  • Could a more established tool be better suited for the project’s needs?

Netflix’s Eureka is an AWS service registry based load balancer with failover protection, first of its kind. My company had a project based on modular and hot-swappable plugins where each plugin could be run on a different container and interacts with the base client using REST. The leader of this project saw Eureka as an opportunity to increase backend functionality to these plugins to add features like canary releases (staged rollouts) of plugin updates or hot-swapping new versions or potentially distributing the base platform to clients and using Eureka to deliver a configurable list of plugins to each. All functionality that sounded great on paper, but there were problems to address. The project itself did not make use of the AWS service registry and the base application was not designed in a way to exactly accommodate the service registry.

Long story short, Eureka was incredibly cool, and the project leader decided to commit to making it work for the project. Months later, the refactor was completed, took more work than expected, pulled time from other engineers to implement, and introduced a slew of new bugs to the system.

The team promised new features, but ran out of time to deliver. The AWS service registry based design was ditched to avoid the new issues, and the project ended up switching to NGINX to provide the new features that were promised, without the need for further backend design changes.

In this case, it would've simply been better to start off with a simple Google search "load balancer for canary releases", and a less trendy but more practical solution could’ve been implemented.

Signs you or your company might be getting stale

The most damaging phrase in the language is… it’s always been done this way” - Grace Hopper

While trying too hard to implement new gimmicks is a problem, never updating processes or tooling in a project or company is a death sentence for the project and its developers. Companies that are lazy or jaded towards change or new technology are generally the same ones reacting to issues instead of preventing them. They also stall out the growth of their own developers.

These projects tend to demonstrate a religious attachment to languages, libraries, or tools, without regard for community consensus about them. Their rigidity can kill new or better ideas from the developers maintaining them. Always ask why a specific language or tool was used, and if you ever hear “that’s just how we do it”, start looking for new work. Never fear feeling stupid when asking these questions.

Places to stay updated on new tech

News aggregators:

A great way to browse a broad set of headlines and find a wide variety of interesting tech news.

HackerNews

HackerNews is a technology news aggregator from tech venture capital firm, ycombinator.

Reddit’s programming subreddit

Reddit’s programming subreddit is a community-managed technology news aggregator including articles and interesting topics of discussion frequently involving experienced members of the community.

Other Sources:

ThoughtWorks TechRadar

TODO

Microsoft’s Channel9

Microsoft’s internal news hub for Microsoft technologies. Recently there’s been a strong focus on Azure services, which is to be expected as it’s their fastest growing offering.

Projects and developers getting recent attention on Github.

Googling “Best tech to do __”

Last but not least, frequently searching for new tech will let you stay somewhat updated on what the trends are for different sectors or common functionality.

Ultimately, passion in the field of software will fuel the desire to always continue growing, but this effort takes commitment as well. If you already love computer science, it’s that much more interesting and rewarding to keep evolving with the field, and it’s never a dull day when you enjoy what you’re doing.