Q&A with JNBridge's Wayne Citrin: 15 Years with a Foot in Two Worlds
When Wayne Citrin and his partners founded JNBridge back in 2001, Java was about five years old, .NET was still in beta, and the term "cross-platform interoperability" wasn't exactly rolling off the tongues of software vendors. But what seemed like irreconcilable differences at the start of the 21st century looked to Citrin and his partners like opportunity.
"We never had any doubt about this business," Citrin told me. "We'd go into the forums back when .NET was first introduced, and one of the most common questions we'd see was: Will I be able to use it with my existing Java? Microsoft was saying, Oh no, you'll want to take out that Java and rewrite it, or if you don't want to do that, we have something called J#, or we have these Java conversion tools. But there were all sorts of problems with those approaches. What was needed was a bridge between the two platforms, and that's what we provided."
JNBridge showed up on my radar nearly a decade ago at a JavaOne conference, and I've been talking with Citrin, who serves as CTO of the Boulder, Colo.-based company, regularly ever since. I covered his company's product announcements when they were newsworthy, of course, but I also found myself turning to him for his unique perspective on these important enterprise technologies. He's a thoughtful observer with a foot in two different worlds. (Sometimes downright hostile worlds: I think the year the company was founded Microsoft and Sun were suing each other.)
With anniversaries colliding this year -- both JNBridge (the product) and .NET turned 15, and Visual Studio turned 20 (Java turned 20 in 2015) -- it seemed a propitious moment to check in with the guy who spends his days getting Java and .NET to play nice.
Sixteen years is a respectable stretch for a niche tech company. How do you account for your continued success?
We recognized the value of that niche right away and focused on it exclusively. Then we went beyond the general-purpose bridge with the Adapters. And with our Labs, we're able to focus on the common and urgent scenarios people were tell us about.
Just to clarify, your namesake product, JNBridge, is a general purpose Java/.NET interoperability tool designed to bridge anything Java to .NET, and vice versa. Your Adapters connect the Java Message Service (JMS) with BizTalk Server and other .NET apps. And the Labs are free kits that showcase scenarios that bridge Java and .NET.
The JMS Adapters are a big part of our business, now. There's a lot of legacy JMS infrastructure out there. It's still one of the most common ways to connect components in an enterprise and pass data around.
What's the most significant change you've seen in Java in the past 15 years?
Most people would probably say it was the handoff of Java from Sun to Oracle. But I think the most significant change we've seen lately is the decline of Java EE and its replacement with a variety of open-source frameworks. Spring, the various Apache projects, the various web frameworks; I think this is probably a good thing. Java EE was too complicated and hard to use, there was a lot of ambiguity in the specs, and things were different from one app server to another. On the other hand, what you have now is a fragmentation with a whole lot of frameworks. The good news is, a lot of them are very good frameworks, easy to use, and supported by a committed community of users who actually like them.
We've talked about this before. You see a difference between Java EE and enterprise Java?
They're not the same thing, and they haven't been for a long time. Enterprise Java is not going away any time soon, and I think it's fair to call Java EE a subset that, in many ways, has had its day.
Looking specifically at the next versions of Java, which changes excite you the most?
I really like Jigsaw, the modularization coming in Java 9. A lot of people are worried about, but it's not such a big deal to us. They really did a good job of accommodating legacy code. If you have a non-modularized jar file, you can still use it in Java 9. You could use it in the crosspath or put it in the module path. You could do the same thing with a modularized jar file. You could also use it in an early version of Java; it just wouldn't be recognized it as a module. They put a lot of thought into this, and I applaud them for it.
But there's another great feature in Java 9 that hasn't gotten a lot of press. It's called multi-release jar files. The jar files in Java 9 have these nooks and crannies. There's a nook just for stuff Java 9 will know about; Java 8 doesn't know about this stuff and won't use it. Java 9 will look there first and override the general-purpose stuff in the jar file. I think is a very cool feature and we plan to use it going forward. We're just waiting for the tooling to catch up to make it easy.
While you've been watching the evolution of Java and .NET, you've also had a ringside seat for the evolution of the people using these platforms. How has the developer community changed over the past 15 years?
Fifteen years ago, developers had much more uniform characteristics. But there has been a noticeable bifurcation. With the rise of open source projects, we're seeing a large group of developers who are really interested in getting under the hood and understanding how things work. They're using lots of really cool tools to see what's happening at a much lower level, really getting their hands dirty and using those tools to solve problems. We're interacting with a lot of developer customers who are very sophisticated about this stuff in a way they weren't 15 years ago.
But we're also seeing a smaller group of developers who will go until they see a problem, and then stop and come to use for a solution. I'm not saying they're bad developers; they just have a different mindset. I think this group is the result of the rise of these frameworks I mentioned and a world of low-code development where things just work. Unless you really want to, there's no reason to look under the hood, and it's even kind of hard to do it. Plus, there's a lot of legacy software out there that just gets maintained, which we didn't see 15 years ago. Often people we're working with are not the original developers; they're just tasked with maintenance. And it might not even be their full-time job, just something they have to do on the side, and only when there's a problem or a new feature.
Speaking of mindsets, your company is now, and always has been, headquartered in Colorado, which is not exactly a high-tech hub. Why haven't you set up shop here in Silicon Valley?
I like it here in Colorado and don't see any reason to move. You don't need to be in Silicon Valley to run a successful tech business. And as expensive as office space is out there, it's cheaper just to get on a plane once in a while.
Posted by John K. Waters on 04/12/2017 at 10:08 AM