Technologies go out of fashion with hipster hackers pretty quickly, but these same developers can serve as a crystal ball to decipher tomorrow’s mainstream enterprise technologies. Tim O’Reilly has long signaled the importance of “watching the alpha geeks” to gauge where technology is headed, and HackerNews’ whoishiring discussion threads give a glimpse into what these developers are doing.

Web developer Ryan Williams regularly culls this data to find insights into emerging technology leaders, and recently released results for May 2016. In them we see a lot of interest in Postgres, React, and Docker, but will this translate into enterprise adoption?

Building the future

Web developers change their frameworks every other month, it seems. First there was SproutCore, then BackboneJS, then EmberJS (arising from SproutCore), then…you get the picture. As popular as each of these frameworks has been, everything took a backseat to Google’s AngularJS when it arrived.

That is, until Facebook’s React was released, which now tops the charts with the HackerNews elite:

Unfortunately, it’s probably too soon to draw any firm conclusions from this data. After all, for each framework there’s a too-familiar rise-and-fall of popularity as something sexy and new comes along to displace it. So far, React has gone highest, but the slope of its rise roughly mirrors that of AngularJS as it boomed.

SEE Stack Overflow founder Spolsky: The three skills every software developer should learn (TechRepublic)

Plus, if we look beyond the hipster hacker crowd at general hiring trends, AngularJS blows away everything else by an astounding margin. It’s (very) possible that React will eventually eclipse AngularJS with enterprise developers just as it has with startup developers, but it’s too soon to conclude this definitively.

The same is true of containers and associated orchestration technologies. Docker is the clear leader today, but it’s hard to see how Kubernetes, Mesos, and Vagrant will do, longer term.

Hitting rewind on databases

But what about databases? We have more data to work with here because things like MySQL have been around for over a decade. Let’s take a look.

Not surprisingly, MongoDB and Cassandra continue to crush the NoSQL pack, while MySQL and Postgres dominate RDBMSes. What’s interesting, however, is the slide that the two frontrunners–MySQL and MongoDB–have exhibited with cool kid developers, even as Postgres’ popularity has soared:

I wrote about this sustained interest in Postgres a year ago, suggesting a confluence of several fortuitous factors (convenience, innovation, concern over Oracle’s ownership of MySQL, etc.) as the reason for its rise. Even when we step outside the hipster hacker crowd and look at general database popularity trends, Postgres fares well against more stolid peers like IBM’s DB2.

SEE Why every developer is an open source developer now (TechRepublic)

Once we step outside this startup crowd, however, MongoDB and Cassandra outpace all other major databases, at least as measured by relative growth in employer hiring interest from

In general, then, startup developer interest might be a leading indicator of a technology’s popularity, but it’s also divorced from the reality of what large enterprises are willing to use today. In the case of Postgres, which has waxed and waned and now appears to be waxing again, it’s not clear that developer interest has done much to spark similar enterprise interest.

All of which suggests that it’s worth paying attention to startup developers to see what they’re using, but figuring out the timing of when that startup excitement will transition to sustained enterprise adoption is tricky. After all, often the very enterprise popularity of a particular technology makes it uncool to the startup crowd. No one wants to drive the same database as their grandpa and grandma.