“We should accelerate the ideas-to-superpowers pipeline. At the heart of computing progress lies a simple process: the research, development, and deployment cycle. Ideas are conceived and refined, encoded into mathematical rules, programmed into software, and deployed into computers, which grants super-powers to humans world-wide. The better and faster we are at sifting ideas through this pipeline, the better and faster our superpowers will come.“
Science accumulates knowledge idea by idea, across various fields and efforts. Frequent incremental contributions build up until we break through and leap forward. Research is where our “ideas-to-superpowers pipeline” begins.
Today, we are thrilled to share some important updates about Research at Protocol Labs, and an exciting announcement. This post covers the following:
First and foremost, Protocol Labs Research is now an established team within the organization. Our goal is to direct and support all the research efforts across our projects and communities. We seek to ask and answer important questions, organize our work, share our contributions, provide a direct and explicit conduit for potential collaborators, and even to fund external research endeavors.
Protocol Labs began with the IPFS community, steeped in the values of open source. For years we have shared both research and development work to the greater commons, working openly and collaboratively and making our work as reusable as possible.
Research at Protocol Labs is a highly collaborative endeavor. We are pursuing several projects with colleagues at other labs and the broader open source community. We are also building a network of scientific advisors who can assist our efforts. These efforts are led by Research Scientists, who work with our communities to develop skills and ideas. We plan to work on many more, so if you share our research interests please check our job listings.
IPFS began its life through a technical paper. Beyond the codebases of its implementations, we are producing lists of open problems, architecture documentation, design docs, and specs. Both libp2p and Multiformats, as well as the first version of Filecoin have also followed the same path. We have also gathered bibliographies and thoughts on specific topics like CRDTs and pubsub.
Last year, we improved our research and research-writing efforts substantially. We produced a much improved version of the Filecoin Protocol, and wrote it up in the new Filecoin technical paper. We also released two related technical reports: Proof-of-Replication, and Power Fault Tolerance, which generalized results achieved through research on Filecoin. Though not yet ready for academic conferences, these work-in-progress write-ups served the purpose of disseminating the ideas quickly, and inspired several other works. We count this as a great success.
We have also used video recordings of lectures and talks to share our thoughts, research directions, and results at various meetups and conferences for almost four years. We highlight a presentation of IPFS at Stanford’s EE380, and both an overview of the Filecoin protocol and more information on a Proof-of-Replication construction from the BPASE 2018 conference.
We will continue to improve the presentation of our work, and look forward to the next opportunities to share it with you.
We work with hundreds of developers and researchers across many open source projects. We’ve always discussed interesting problems and new directions on Github, through repositories like ipfs/notes). Recently, we’ve begun a new effort to formalize and organize open problems in a new research repo to serve as inspiration to other researchers, ranging from undergraduates to professors. We intend for this repo to be a location for discussion and collaboration on these important open questions, where individuals can propose and discuss their own research problems.
We have also started supporting research conferences, events, and student scholarships to attend them. Our sponsorships to date include Crypto 2018, FC ‘18, BPASE 18, MGS 2018, and the MIT Bitcoin Expo 2018. We hope to host our own conferences in the coming years.
For the future, we are exploring fellowship and internship programs to explore new ideas and support the development of expertise at various levels in the community.
We are thrilled to announce that, in proper internet R&D tradition, we are launching a Request For Proposals (RFP) Program with an initial commitment of $5,000,000 to fund multiple rounds of grants. If these are successful, we hope to grow the program even further. The first round of RFPs and application instructions can be found on Github. Each RFP has an associated award ranging from $5,000 to $200,000. Applications will be judged on quality of the proposed research direction and relevant experience of the individual or team; complete solutions are not necessary to apply. Each RFP references an Open Problem Statement in our public research repo.
To inspire greater contributions to the commons, all IP generated via these grants must be open source and widely usable. Any software produced will be available with a permissive license and should benefit numerous projects. We’re still working out the specifics; we’re considering the public domain, licences like the Mozilla Open Software Patent License, and other arrangements in the same spirit.
The first RFPs solicit improvements to components used in Filecoin and a breakthrough that may dramatically simplify consensus protocol including Filecoin’s. Future RFPs will fund problems across our whole stack of protocols, from Multiformats to Filecoin, and we also welcome open problems and suggestions for work we may be interested in funding.
If you would like to propose a research collaboration, or have any questions or comments for the research team, please email firstname.lastname@example.org. If you might be excited to discover, manage, or work on open problems, we are hiring for full time positions. Questions about the RFP program and proposals should be sent to email@example.com.
 Due to the open-ended nature of these problems, it’s worth noting that we may end up choosing more than one proposal per open problem. In this case, we would fund multiple applicants to pursue their respective proposed solutions. Award amounts will not be affected by the number of accepted proposals.