Q&A: Open source veteran Brian Behlendorf on leading the Hyperledger Project

"The blockchains, marked contracts, distributed ledgers and databases in general are really about building the next generation stack for the Internet," says Behlendorf. It's "hard to imagine an industry that isn't going to be affected."

Brian Behlendorf is a renowned open source personality. He was one of the primary developers behind the Apache Web Server, a foundation of the Web as we know it. Behlendorf was also a founding member of the organization that later evolved into The Apache Software Foundation. In 2003 MIT Technology Review named him one of the top 100 innovators in the world under the age of 35.

Brian Behlendorf by Joi Ito Joi Ito

Brian Behlendorf

He is now joining the Linux Foundation to lead the Hyperledger Project, a Linux Foundation Collaborative Project whose mission is to advance blockchain technology.

I talked to Behlendorf to learn more about the Hyperledger project and his role there. Here is an edited version of that interview.

How do you think the Hyperledger Project is going to change things for the financial industries and banking? What encouraged you to get involved with the project?

There's a whole lot of visions out there for how this set of technologies — the blockchains, marked contracts, distributed ledgers and databases in general are really about building the next generation stack for the Internet. It's going to build on top of the previous stack: TCP, HTTP, all that. It's going to be about finding a way to truly build cloud applications.

I think to this point, cloud has really just meant other people's servers. But that’s changing. I think this is the basis for building data that truly does live in a cloud: Transaction record that is shared; databases that can still have some parts private and some parts public; applications that run in multiple places at the same time. There's a whole lot of people out there who have big starry visions for how this happens. Don Tapscott wrote a whole book on the Blockchain Revolution.

There's some very optimistic people out there. I felt like if even a fraction of these visions are likely to come to pass, both in terms of currencies and also in the form of simply permission chains, it would have a tremendously positive impact on not just the technology space but on society as well. I felt like I wanted to get involved. I wasn't really all that eager to get involved on behalf of a single company, a big company or a startup company. I really felt like there was a need for an organization to come forward to help build a common technology base. I'd spent some time looking at how technologies like Bitcoin work and then I talked to Jim Zemlin about Hyperledger. I really felt that this is a project that has a whole lot of promise, is targeted at the right layer of that stack, and can be very pragmatic, very practical.

With Apache and the HTTP Web server, it was partly about helping fight for the open Web. It was also about building a printing press for everybody and putting it into everyone's hand. I think that degree of pragmatism and practicality is pervasive through the Hyperledger Project. I really wanted to be part of it.

Can you expand a bit on how projects like Hyperledger benefit from being part of the Linux Foundation?

I joke that we should hire Rasmus Lerdorf (PHP creator) and one of the MySQL guys. Then, we could be the LAMP (Linux, Apache, MySQL, PHP) foundation. I think what Jim and and Linus have done with the OS and with the amazing stuff they've pulled together is pretty remarkable. I think they've been able to find a way to engage the companies that are building upon common technologies, who a generation earlier might have been looking to standard bodies to help them build interoperable systems. Now, everyone's realizing open source is the new path to standardization.

It hearkens back to the viewpoint we had in the earliest days of Apache. The best way to get adoption for a standard and to get everybody on the same baseline is to release high quality, production quality software that implements that standard that everybody can embed inside their product.

The Linux Foundation has had this great experience managing the kernel, managing the Linux standard space, managing the needs of specific sectors like automotive or the telecom industry and others. They made sure that those needs don't corrupt the core open source, the collaborative software development process that is the magic that makes this all work. But still bring those companies in, help them understand how that process works, help them understand how their financial contribution and their developer contribution result in a greater win for everybody involved. I felt very comfortable saying I want to be a part of that, and that's the right direction for Hyperledger.

Let's switch gears and go back to the Hyperledger Project. What industries will clearly benefit from blockchain or the Hyperledger Project?

The first industry to really jump into this, commit real resources in figuring out that it's ready to move into production is the financial industry. There's no surprise about that. There's a lot of situations where the financial industry needs to maintain a consistent database of records between a whole bunch of participants in the ecosystem. It's not just one ledger; it's lots and lots of different ledgers, different chains. Lots of different sub-communities and permission rights and all those sorts of things. The idea of a generic technology for building these chains is incredibly powerful to them.

What's the next industry to really jump in with both feet? It's hard to say.

It's also hard to imagine an industry that isn't going to be affected by this. Just look at the supply-chain industry. There are people working on extending this to figuring out how do we optimize and bring transparency to the supply-chain industry by putting blockchain technologies at the core of that. How might we optimize and address the healthcare industry? Healthcare data sharing is actually something I got involved in while I was in D.C. for 2 years. I worked for the Department of Health and Human Services on a fabric for changing personal health records and open-source software to accelerate that. We were able to accomplish a lot but this is an industry that doesn't quite understand information sharing, yet. They don’t yet understand the importance of the patients being at the center of that health/information exchange.

And there's a whole lot of potential for public ledger technologies, for blockchain technologies, to provide a quantum leap for that industry. It's not unlike the adoption and the spread of mobile networks in emerging market countries where as mobile technology developed they were able to leapfrog a whole lot of deployment of infrastructure that was required in the United States when we first started to spread telephone networks. There was a lot of stringing of wiring and a lot of physical plants that was very expensive. Emerging markets got to skip all that, they got to leapfrog that. I think blockchain technologies in healthcare could help leapfrog a whole lot of information sharing infrastructure that would otherwise have to have been built. And still do it in a way protects patient privacy and empowers patients to move their data portably and that sort of thing.

In any major industry you're going to find a community of business partners who need some shared ledger to keep track of who's got what and who's sending what to whom. I guarantee you, you'd be able to find a use case.

What will be your role or job as executive director of the project?

The project is already underway. It already has contributions and active developments by a large number of developers actually coming from multiple different companies: IBM, Intel, Blockstream. That contribution needs to continue to grow. There's already been face-to-face meetings, hackathons and all those sorts of things.

My job is to really help accelerate the momentum that's there. Look for more open source developers who want to join the project. To try to nudge the project towards using the best practices and the best techniques from the open source communities that I've been a part of from Apache to Mozilla. How do we drive this towards a product that people can deploy and really start using in production circumstances.

My role is also to some degree being a buffer between the developer community, the contributor community, and the funding community. The funds are really important. Funds are what enable full time resources at the Linux Foundation to be used to drive this project forward. We are kind of the Switzerland in the space. We're the ones who provide a neutral ground for the collaboration to take place, where the developers can say, "No matter who your employer is, we're here to build great code, and let's work together collaboratively to do that." That kind of environment needs facilitation. It needs someone like me and the Linux Foundation to act as neutral ground. That's our role. I'm not here to write code. Those days are long behind me. I'm certainly technical enough to keep up. I can call BS when I need to. It's really to empower the developers and foster a culture of "do-ocracy" amongst those who show up to our party.

What are the big challenges you think you will be tackling because it’s a relatively new project?

I think the first big challenge is to ship the code. We already have contributions, we already have a process. We need to put this together as a package and describe what are these cases that it's great for, what are these cases that are still in the future. We have to make it something tangible. Something people can put their hands around and start using in production. That's a high priority.

The second challenge, I think, is that it is still a new technology for people. This is, in some ways, like when NoSQL databases started coming out, and there were some great use cases for it. There were some places where it would not really provide an advantage and might even be a disadvantage. Understanding those differences took some time. It took some time figuring out when you didn't want to use a NoSQL database. I think likewise helping people understand when their needs map to what Hyperledger can deliver. I think there's going to be a need for a lot of experimentation, a lot of sharing of experiences. A need for smart developers out there to dive into the business use cases, understand what are we really getting at at the core of those needs, and how can we write software to meet those needs the best way possible. Those are some of the challenges, but those are the fun kind of challenges.


Copyright © 2016 IDG Communications, Inc.

The CIO Fall digital issue is here! Learn how CIO100 award-winning organizations are reimagining products and services for a new era of customer and employee engagement.