Stack Overflow: Python is on fire, and Kubernetes needs some housecleaning

What has been hot on Stack Overflow over time is not necessarily a good indication of what's cool now.

python.jpg
Image: iStockphoto/DenisKot

Google's Felipe Hoffa just published an excellent analysis of Stack Overflow questions, trying to uncover the direction developers are moving. For example, it's interesting to know that developers have a long-standing affinity for Java (which also shows up in the Redmonk rankings and elsewhere), it's much more interesting (and surprising) to see that Java doesn't crack the top 10 in the most recent quarter, and even vanishes from the top 30 questions (measured by current page views).

Java, a mainstay for decades, struggles to get noticed with developers today, according to Stack Overflow data. Python, by contrast, is on fire, largely due to its flexibility and applicability to the burgeoning world of data science.

So what else can we learn about developer interest?

But first, more Java vs. Python

Before we answer that question, let's dwell for a second on Hoffa's uncovering of all-time vs. current interest. As mentioned, Java has been a general purpose workhorse for many years, which is why it (along with Javascript and HTML) dominate the top 10 and top 30 Stack Overflow charts.

SEE: Hiring kit: Python developer (Tech Pro Research)

But when we look at the most recent quarter (or recent quarters, which you can do by using the interactive tool running Google's BigQuery), Java evaporates from developer consciousness (Figure A).

Figure A

screen-shot-2019-02-20-at-2-52-30-pm.png
Image: Felipe Hoffa

Python, by contrast, rises to claim 40% of the field in the top 10, and more than 50% if we consider the top 30 questions asked. It's an impressive dose of programming language hegemony, born of Python's accessible, extensible nature. Given how easy it is to pick up, many developers add it to their "toolkits" in addition to their primary language (which may well be Java). This could be one reason we see so many questions about it on Stack Overflow: Interest is exploding as developers apply it to applications they're building and may need help getting familiar with it. Those questions are a leading indicator of a heck of a lot of Python adoption to come.

What about Kubernetes?

Like Python, Kubernetes is hot. Like Google's TensorFlow, however, as Hoffa noted, Kubernetes users keep asking the same questions. At the end of 2017, the top question for Kubernetes was "How can I keep container running on Kubernetes?". By the end of 2018, that question retained its top spot. Ditto the second-most asked question: "Pods stuck at terminating status." Across the top 10 Kubernetes questions in 2017, they remain largely unchanged in 2018.

On the one hand, the persistence of these questions could simply indicate a broad-based influx of new users that all have to cut their teeth on Kubernetes in the same way. On the other hand, it could suggest that Kubernetes developers may not be solving the most common teething pains with this open source container darling.

SEE: Riding the DevOps revolution (ZDNet special feature) | Download the free PDF version (TechRepublic)

Thinking beyond Kubernetes, if you're a product owner or core committer to a growing open source project like Kubernetes (or Kafka, MXNet, or more), this Google-spawned dashboard is a great way to gauge the progress you're making. If the same questions refuse to go away, that's a sign that while you may be inventing the future (new functionality!), you may be forfeiting present adoption by leaving newbies out in the cold.

That said, Python has also maintained largely the same top questions without resolving them for its user base, and that clearly hasn't diminished Python interest (which is true for Kubernetes, as well). So let's be clear: The sky isn't falling on these popular projects. Even so, these Stack Overflow questions do suggest that adoption could be facilitated further with more attention on resolving the nettlesome questions that aren't going away.

Also see

By Matt Asay

Matt Asay is a veteran technology columnist who has written for CNET, ReadWrite, and other tech media. Asay has also held a variety of executive roles with leading mobile and big data software companies.