i2C Member Spotlight: Protocol Labs

Protocol Labs Lawyer and Software Developer Ian Darrow Explains How the Internet Could Use an Upgrade

Protocol Labs is a new member of the i2Coalition. Find out about the projects Protocol Labs is engaged in to make the Internet more decentralized, faster and more resilient for everyone.

Most Internet users don’t think about the underlying infrastructure until it fails or connections get cut off. The Internet isn’t as resilient as many of us think. Over a decade ago, Harvard Law School professor Yochai Benkler wrote a seminal paper on the need to build a more “survivable critical infrastructure” for the Internet based on “distributed, peer-based systems for communication, computation, and data storage and retrieval.” Today, many companies have taken up the challenge to create such a survivable Internet infrastructure through open source protocols. Protocol Labs is one such company, with a focus on building distributing data storage leveraging existing investments and creating a marketplace available to all data storage providers.  

Protocol Labs is best known for three protocols in development—IPFS (a distributed technology project for content-addressing of content), libp2p (a peer-to-peer, or P2P sharing library, and Filecoin (a cryptocurrency protocol backed by the world’s top venture capitalists to enable all data storage providers to network in the same marketplace to market their storage solutions).

Protocol Labs is the company behind these technologies, which not only defend the right to information and help enable civic and political action, but it also has the potential to build a more robust, faster and resilient Internet for everyone.

Ian Darrow, Counsel and software developer at Protocol Labs, works on a range of topics. He had a moment to chat with i2Coalition about some of the projects going on to upgrade the Internet.

David Hamilton (i2C): For those unfamiliar with the cool stuff going on at Protocol.ai, please give us a little background.

Iain Darrow: We’re an open-source software company. Our mission is to upgrade the fundamental infrastructure of the Internet. We’re taking that on by building new protocols that we make available to everyone that will help make the internet more decentralized, more private more robust and more efficient. We’re a bunch of true, old-school geeks who know how much better the Internet has made everything but we’re starting to observe the cracks developing in the architecture of the internet over time.

DH: So, what are some of the ways you’re trying to address these cracks as the Internet evolves?

ID: We built a few products. All of them are free, open-source software.

The architecture of the internet makes it really hard for engineers to establish peer-to-peer connections when building projects. So, when I gchat with you, for example, what’s actually happening is that we’re actually both talking with Google servers is relaying messages for us because the client server model works very neatly on the internet.

So our first project, libp2p, makes it very easy to establish P2P connections. There’s a lot that goes on under the hood but the end-result is a library that developers can drop into their software that gives them the superpower to enable P2P connections really really easily.

Then we used libp2p to build IPFS—the Interplanetary File System. The basic principle of IPFS is that it flips the current model of how we deliver content today. Currently, when you make a webpage request it ultimately gets resolved through DNS to a specific physical server that sits somewhere and it serves the specific file you’re looking for and that travels back. Even if you and I are in the same room in Toronto and we request the same file that lives on a server in California, both of our requests will do this 3,000 mile roundtrip to California even though when the second one of us makes that request information is sitting in the same room as us.

IPFS flips this model from a server-addressed model to a content-addressed model. So, whenever you store content on the IPFS network, it gets a unique identifier, then instead of asking for a specific server for that content, you ask the IPFS network for that content and the closest peer that has the content you’re looking for will service you. So, we’re hoping the existing hardware that we use to power the Internet, the internet backbone with large adoption I can see this making the Internet 10 to 100 times more efficient using the same infrastructure because you’d have so much traffic being routed over long distances. There are a lot of challenges before we get to that, but that’s the overall vision.

DH: Why do you think it’s so important to address these weaknesses in the current Internet?

ID: We build things that make the internet more resilient by removing any single points of failure. That’s how Internet censorship is possible today: those single points of failure.

DH: Beyond just being a good global citizen, how do you incentivize participation?

ID: The current problem with IPFS is that ultimately despite it being on a peer-to-peer network it has to live somewhere, which is a hard drive that has the data. There’re not a baked-in way to pay people to reliably store and maintain data that you care about  on the IPFS network, so we’re building a thing called Filecoin, which hasn’t launched yet but it’s the incentive layer for IPFS. It’s a cryptocurrency that creates a two-sided market of storage purchasers and storage providers.

Ten years ago, it would have been difficult to rent out a room in our house and compete with Mariott. Now we have access to these market platforms like AirBnB where you can have lots and lots of small providers that joined together that create a platform that in aggregate compete with big centralized providers.

The difference between AirBnB, though is that AirBnB takes a huge cut from transactions that occurs on their network, and Filecoin is open source software so it’s purely disintermediated transactions between service providers and service purchasers.

DH: Protocol Labs obviously has a lot going on. Is there anything in particular that made joining i2C particularly attractive?

ID: In terms of i2C, we’re this little company and it’s wonderful to be joining an organization that has all these legends of Internet infrastructure—and other little companies. I think almost everyone who’s responsible for the way the web functions today is an i2C member, and that’s really cool. Being able to work with the people who make the pipes work and listening to how we can make the pipes work better is really really exciting, and could lead to upgrading the internet which is what we all really want to do.

Find out more about Protocol Labs at protocol.ai.