Swimming with the Poets

Interview – Meet Jim Zemlin

Article from Issue 196/2017

Jim Zemlin has directed the Linux Foundation since 2007, when the foundation began with the merger of the Open Source Development Labs (OSDL) and the Free Standards Group (FSG). Today the Linux Foundation has gone beyond Linux and become a huge umbrella that houses many open source projects that are critical to our economy and society.

Jim Zemlin has directed the Linux Foundation since 2007, when the foundation began with the merger of the Open Source Development Labs (OSDL) and the Free Standards Group (FSG). Today the Linux Foundation has gone beyond Linux and become a huge umbrella that houses many open source projects that are critical to our economy and society.

To celebrate the 25th anniversary of Linux, I sat down with Zemlin for a 40 minute interview. He is very humble and doesn't like to talk much about himself. However, since we have known each other for such a long time, he opened up a bit. Here is the edited version of that exclusive interview with Jim Zemlin.

Linux Magazine: Can you tell us a bit about your childhood and your exposure to computers? What kind of technical background do you have?

Jim Zemlin: I have a very weak technical background.

When we were really young kids, personal computers were just starting to emerge. The computing industry had not yet consolidated. There was Commodore 64, the Ataris, Apple… and there were the Tandy kits. So I grew up with computers.

My grandfather was one of the founders of Cray Research. He had been in the computing industry since World War II. Then my father was a computer programmer at Control Data Corporation, which was headquartered in Minnesota. At that time, Minnesota was the hub of companies like Control Data Corporation, Honeywell, and 3M.

My dad was actually really smart about how to teach us about computers. He said that we could play games for 10 minutes, or we could program for an hour. We just wanted to be with the computer (and those were boring games), so even as a kid I started doing programming work. Just simple things like Basic. My brother and I ran a bulletin board system back in the mid 80s. Later on, my brother also worked as a developer on a game titled Oregon Trail (the game we used to play).

That's about where my technical acumen started and stopped. Just a lifelong fascination with computing systems.

LM: When did open source enter your life?

JZ: After the collapse of the dot-com bubble, I got involved with Covalent Technologies, which was started by early Apache Software Foundation developers. That was my first introduction to open source. I think it was through Apache that I originally got involved in open source.

LM: What led to the creation of the Linux Foundation?

JZ: When Covalent Technologies ended up being acquired by VMware, I kind of started having a midlife crisis. At that time I thought of becoming a chef, a professional rock climber… But I really got intrigued by the open source community. I got involved with the idea of creating a standardized way to run applications across the variety of different Linux distributions. That was sort of a response to fragmentation in Unix. That led me to working on the Linux Foundation by essentially consolidating a lot of different open source organizations into what the modern Linux Foundation is.

LM: The Linux Foundation has come a long way from those early days. It has become a huge umbrella that goes beyond Linux. What are your reflections on those early years?

JZ: You've known me for a very long time. You have known me before I was married, and I have been married for 11 years now. I remember my first date with my wife, and she asked me what did I do for a living, and I said, "I work at this non-profit organization; it's open source; the software is all freely available, and we give everything away."

She went to Harvard Business School. She is a very Type A senior executive at a technology firm in the Valley. When I told her about what I was doing, there was this disappointment on her face; it was just palpable. She started glancing at her watch to get out of there.

Fortunately (chuckles), I was charismatic enough to get her to marry me. And the Linux Foundation turned out to do something that I had hoped and thought it could do. It continues to be our North Star. At that time, we were very focused on Linux; we were creating the greatest shared technology in the history of computing.

I think everyone stumbles across this idea of what is the purpose of a particular organization. Once you understand that, once you grasp it, you can move ahead. In our case, it became clear that the DNA of Linux – the development model, the economy around it – could become a template for other large-scale open source endeavors. We realized that it would lead to the creation of a great collective shared technology investment, beyond Linux.

That continues to be our goal. Everyone at the Linux Foundation knows that we're here to support the creation of these great shared technology resources that are shared in the sense of the code, that are consumed by society – resources that governments and industries depend on.

Linux is a great responsibility today because the world is truly dependent on the software to remain secure and stable. It has become an effective platform for a lot of important computing systems that run our daily lives.

We are building a foundation for every level of the software stack, whether it's web frameworks, network function virtualization, software-defined networking controllers, or container management systems like Kubernetes. These are the projects that define the most important aspects of the technology economy.

LM: I think the Linux Foundation's biggest contribution is the fact that you have made it easier for corporations to work together on these technologies that are building our future.

JZ: As I said in my keynote speech, Linux has proved that you can better yourself while bettering others at the same time. The ability to make your company (and yourself) better and share that with everyone else is the real value of what we are creating.

Commercial adoption and sharing are not mutually exclusive. It is better to have a situation where everyone wins as opposed to someone wins and someone loses.

If we look at the code part of it, at the highest level, it's like writing poetry. People like Linus Torvalds and many other great developers are like poets. They're writing this incredibly important software and enjoying it, because it's an active artistic creation for them. That is then being used in some of the most interesting things in the world. We see people using Linux in unmanned aerial vehicles; they are using it in some interesting aspects of search and rescue; they are using a Raspberry Pi running Linux to teach kids about computers… it's amazing.

It's also important that companies have committed to running Linux on all of their modern systems. Amazon is a good example. Their shopping system, their entire e-commerce platform, devices like Kindle… all of that runs on Linux.

That dependency then begets Amazon improving that code, sharing those changes back with the open source community. Those changes then improve that open source code base, and then that cycle starts over again, but maybe this time with Facebook and many other players.

You get this virtuous cycle, which is complementary in nature and in fact is the whole point.

LM: Linux has played a huge role in making companies comfortable with the open source development model.

JZ: Yes. It's the proof that we can all work together on things that are non-differentiated plumbing, for example, as competitors and still compete effectively. In fact, we can compete more effectively, which is the key insight that Linux has taught. It has essentially redefined how software gets created.

The thing I like to think about is every time I talk to a company, whether it's GoPro making a camera or Toyota making a future in-vehicle navigation system, they are not going to go write their own kernel. Look at Linux. There are tens of thousands of packages, millions of lines of code, changes seven times an hour. Why would someone want to do all of that alone?

LM: What do you think Linux has achieved in these 25 years, and where do you think it is going?

JZ: I think it's going to be around for a long time, mainly because, if you look at history, operating systems tend to be much longer technology waves than applications. There are so many interdependencies upon it, both from a hardware perspective below the OS, and from the application economies built on top of that. But at the end of the day, you need software to interact with all the hardware underneath, and that's essentially what the operating system in Linux does.

What we've seen over the years is you had the mainframe operating systems, then the Unixes came into being, then personal computing operating systems came into existence. But in each of those instances, they kind of peaked, and then they sort of slowly eroded away.

What has always limited many of these platforms is their inability to jump from one form of computing to the next. Unix, at the end of the day, is still pretty much a server operating system. I guess you could include Mac OS as, maybe, a counterexample to that. Personal computing operating systems have had a harder time jumping to mobile. The modern smartphone systems that we know today were new operating systems. There has been one exception: Linux.

Linux has jumped from server to high-performance computing to mainframes to mobile devices to embedded systems to tiny little real-time applications. That has not been achieved before, so I think that it's the malleability of Linux, it's the organic nature of the roadmap (which I think Linus, to his credit, has effectively managed by not having a real permanent roadmap) that will make Linux endure for decades to come.

LM: In previous LinuxCons, Torvalds said that Linux can't be shrunk anymore. If there are people looking for really small devices, they should look elsewhere.

JZ: These have been some interesting technology experiments. What becomes too tiny for Linux, or in other words, how small can Linux be before it just doesn't work and you have to move to a more traditional RTOS [real-time operating system]? Developers have got it really small. You can get it small, but there are real limitations.

Real-time operating systems are an example of an industry where there has been a lot of fragmentation. And the reason for that is when you are building some very small application using a tiny, real-time operating system, you generally buy a chip that comes with an RTOS. You then build your app on top of that (it's all custom development) and ship your product.

Today, the reverse is happening. There has to be an ecosystem, an application and developer ecosystem, around an operating system, even for a tiny RTOS, to enable more rich functionality that's required for today's modern computing environment. That's totally opposite of custom development done on one chip, which leads to fragmentation. There is an opportunity to consolidate a lot of that fragmentation in the real-time operating system market. We have a project called Zephyr, which is a good example of that. There are a lot of other real-time operating systems out there, and I expect that there will be consolidation in that space.

I don't think it comes at the detriment of Linux. I think that people trust Linux as a neutrally owned territory, and they can invest in it. People trust that no single entity can control or monetize Linux to the exclusivity of anybody else. We will see how it pans out; there is a lot of exciting development happening in that space.

LM: Recently SiFive announced open source chips. After conquering software, are you interested in doing hardware?

JZ: I haven't looked into those kind of initiatives, lately. We've looked into some things around quantum computing and other areas where the hardware dependencies are unique. But I tell you, our cup runneth over right now, just on the software side.

LM: In that case, what's your big goal now after conquering the world with software?

JZ: If our goal is to create the greatest shared technology resource in history, there are a few things we want to accomplish to help make that happen. One thing we've already done is that now open source essentially powers every modern computing system. If you want to build anything today, you use open source to create that technology.

The problem we have is that the industry and enterprises outside of the very narrow tech sector aren't capable of managing that external research and development as effectively as they should. They don't have a good procurement process. They don't have a good method for letting their developers participate in open source projects, to pull code into the company, to modify it, and to then release it back to the open source community. They don't have good methods for picking open source projects strategically. And as every company becomes a technology company, they need to be good at managing open source.

To address these challenges, we are developing a set of resources, whether it's attending our events, whether it's training, or whether it's participating in our projects, that will help bring in that next set of participants in open source to teach them essentially how to manage external R&D and how to leverage open source effectively.

The biggest bottleneck that we find is when companies come to us and say that they know they are using open source, they know that they need to understand how to maintain it over time, but their lawyers don't understand how to work within the licensing regimes. They don't understand the procurement process. They need help in all those areas, and we are building resources to help enable that to happen.

We also want to make sure that open source software is written securely. Secure coding workshops, threat modeling efforts, and better testing for all open source projects are things that we really want to do.

Those are big things. They will keep us busy for many years to come. So hopefully in 25 years, I'll still be around.

LM: That's too specific, what's the big picture?

JZ: If we can create this great shared software resource that's secure, that's economically sustainable by organizations taking a real commercial dependency upon it, and then in turn underwriting joint development of it, if we can foster a community that embraces diversity and brings in a new generation of the sort of code poets who write great software, I think we will be on the right track.

LM: After doing all of this amazing work, what satisfaction do you get when you go to bed every night?

JZ: I read my daughter a book, every night before I go to bed. I'm satisfied that she's going to live in a society (for as many perceived problems as the world has today) that hopefully is a little better because people share, and that's certainly satisfying.

Buy this article as PDF

Express-Checkout as PDF
Price $2.95
(incl. VAT)

Buy Linux Magazine

Get it on Google Play

US / Canada

Get it on Google Play

UK / Australia

Related content

comments powered by Disqus

Direct Download

Read full article as PDF:

Price $2.95