Distributed Ledger Technology for Open Scientists

What is it and how can it move open science forward

Jonathan Starr, Matthew Bagazinski, Carla Ostmann, Erin Magennis

Introduction

The open science community has made great strides over many decades. Successes such as governmental mandates on open access, the establishment of access to knowledge as a human right, and the creation of various open science tools give more people the ability to produce, publish, and consume knowledge than ever before. Organizations such as GoFAIR, Just One Giant Lab, and the Center for Open Science lead to the adoption of open science policies across the research community. All in all, the open science movement inspires future scientists and the public alike to advocate for a system where the search for knowledge and understanding over-rules, but does not preclude, any capital incentive or corporate mission.

Despite these advances, however, challenges to the adoption of an open system of science persist. Incentives remain misaligned. Funding continues to be bureaucratic, poorly allocated, and time intensive. The scientific record is under constant threat of content drift, censorship, and malpractice while publication continues to rely on single points of failure and a struggling peer-review process. One hypothesized reason for the struggle to adopt open science is that the movement is pushing open science principles and practices into a system that is not designed for them. The movement is trying to give permission to do open science in a system designed for something else.

Distributed ledger technology, or DLT, presents the opportunity to develop a new tech-stack for the system of science, one that guarantees a diverse array of systems that are locked open by default instead of held open by permission.

This new tech-stack of the system of science enables countless novel possibilities for funding, producing, and incentivizing scientific outcomes rooted in open practices. Examples include permanent and immutable identifiers, as well as new, interoperable, transposable metrics for credit, citation, and impact. Further, DLT and the resulting stack greatly increases systemic efficiency, provides professional freedom, reduces costs, democratizes power, and reconnects the public with the scientific process.

The greatest risk to those of us combining open science with DLT is the youth of and within the movement. Full of energy and excitement, many projects are building solutions without a full understanding of the history of the problems. Such unguided progress may lead development away from a positive revolution in science toward the dangerous, hyper-financialized and corporate captured realms of DLT.

It is also important to recognize that while DLT is a tool with which we manipulate the foundational infrastructure of systems, it does not solve all problems plaguing the traditional system of science. Metadata, ontologies, and the handling of human data are examples of problems that need to be solved outside of the distributed ledger space.

Building an open system of science with DLT will require the experience, knowledge, and leadership of people who have dedicated their careers to making the system better for future generations. We hope that this brief introduction to distributed ledger technology will spur thoughts, questions, skepticism, and ideas, and we invite you to share these with our growing sub-movement that seeks to utilize DLT to advance open science.

Introduction to DLT

There are three core aspects of DLT that make it an interesting technology for scientific enterprise.

First, distributed ledger technology solves the Byzantine generals problem of decentralized systems. In short, this solution enables permissionless systems of trust that rely on cryptography instead of human intermediaries. A user can look at a ledger, the scientific record for example, and trust that the publishing time and publishing source is accurate. That user can also trust that the content they are viewing is immutable, public, and tamperproof. With DLT these qualities exist by default instead of through a guarantee by a human authority susceptible to bureaucratic bloat, error, corruption, or malpractice.

Second, while the various mechanisms within DLT are diverse and rapidly evolving, “programmable money” persists as a core structure consistent with all distributed ledgers. This money is created by the ledger protocol, defined by rules agreed on by all participants of the ledger’s network, enforced by code, and maintained through open source practices. The money can be divisible, non-divisible, transferable, non-transferable, and can imbue its holder with specific rights and privileges. The money structure’s emission mechanism defines incentivization processes for system participants. Changes to these aspects of the protocol therefore change behavior of system participants. For example, money can be distributed directly to system participants openly publishing their data, micropublishing, peer-reviewing, or otherwise performing tasks critical to a system of open science. 

The third core innovation of DLT is its open-source nature. This means that anyone can copy a system, change some structures, and make their own system of open science. Anyone can then choose to participate in any of the diverse set of systems, or multiple systems at the same time. While systems with many participants might be considered “standard” at a given time, smaller, niche systems continue to exist and innovate, drawing in new participants as they become more useful and exporting key innovations to systems that are more widely adopted. DLT ensures that the scientific community has the power to choose in which system and structures they wish to operate, instead of being locked into a suboptimal system, as is the case today.

These core qualities of DLT mean we now have the technology to build not only one, but a multitude of infrastructures for the system of science that guarantee openness by design instead of by permission or authoritative instruction.

A New Infrastructure

The ultimate capability of DLT to transform most structures and mechanisms within the system of science is vast. We introduce several significant possibilities below, but we encourage deeper diving into the potential of the new tech-stack for science. We encourage questions, skepticism, doubt, and good-natured debate. We invite you to contact us or visit one of the resources provided at the end of this section for direction into the distributed ledger environment.

The Immutable Ledger

At the core of any DLT is a public, permissionless, immutable ledger of data. While the most popular distributed ledgers, such as Bitcoin, store financial data, the data within the ledger can be anything. Anyone can publish to the ledger, no one can manipulate entries recorded on the ledger, and so long as the network supporting the ledger persists, ledger entries cannot disappear. 

Imagine such a ledger designed to contain the scientific record. Permanent IDs immune to drift and rot, efficiently maintained and preserved with zero risk of human failure, and accessible to anyone. Imagine inexpensive permanent storage of data and publications that is not dependent on a single point of failure for custodianship. Imagine verifiable and permanent timestamps credited to the publisher at the time of entry. These are only a few possibilities enabled by open, permissionless, distributed, immutable ledgers.

Publication vs. Curation

It might be easiest to think of the immutable ledger as a lake of data and raw knowledge accessible to anyone. Everyone can tap into the same lake of knowledge whenever they desire. If I wish to make a journal, I pull knowledge directly from the lake instead of first needing to convince people to submit their manuscripts to me;

Instead of scientific journals determining what gets reviewed and published, publication is effectively free to all while everyone, including existing journals, has access to every publication and is free to design a curated editorial. 

Nature, for example, can highlight the manuscripts that fit their specific criteria and display them in their magazine. Such an accolade would carry the prestige associated with being published in Nature. At the same time, that same manuscript can be highlighted by any other editorial company, institution, scientific society, or individual that desires. 

The scientist publishes; the editorials curate.

Attribute Verification

If anyone can publish to the scientific record, how are we to determine effective vs. ineffective manuscripts? There are several technical solutions within DLT that various groups are experimenting with, however these solutions are out of the scope of this paper. In theory, however, the open nature of DLT guarantees that anyone can assign an attribute to a publication. These attributions, such as “contains an open data set”, are verified by other network participants. If a reputable organization, GoFAIR for example, assigns or verifies an attribution, it is reasonable for other system participants to trust that the claimed attribute is accurate. The key to DLT is that any system participant that desires can ignore all reputation, organizations, and institutions, and verify the claim themselves. They do not need to trust, they can verify.

If a system participant finds a falsely claimed attribute, they mark it as such and the reputation of everyone connected to that claim is affected. If a system participant, let’s say a group of physicists, accurately assigns, verifies, and rebukes a number of attributes, they can be said to have built a reputation on par with that of established organizations. Maybe that reputation only applies to their review of physics manuscripts. Maybe it applies to all natural sciences. Maybe it applies to all sciences. The reality is that they will have seamlessly entered the marketplace of attribute verification without capital requirements. They will have built an organization that contributes to the advancement of knowledge and understanding simply based on their passion and capabilities. 

A second possible verification process relies on monetary compensation. Any network participant can request that a claimed attribute be verified. They can also attach a payment to that request. Verification organizations and societies can fulfill these requests and establish a transparent reputation behind their ability to accurately verify attributes while also getting paid. These societies can be comprised of field specialists, patient groups, engaged citizen scientists, institutions, accredited experts, or any combination of interested parties. 

It is also possible to reward attribute verification directly from the emissions mechanism of a distributed ledger instead of requiring people to pay for attribution verification. Such an implementation would be highly complex and produce countless downstream economic effects well out of the scope of this paper. We invite you to the discussion on the topic.

Peer-Review

Peer review on a distributed ledger is similar to attribute verification. Many editorials will likely have their own internal processes that include peer-review. Peer-review can also be requested by any network participant, including the publisher. Payment can be attached to peer-review requests. Various participants, groups, scientific societies, organizations, and institutions can  participate in the open peer-review marketplace, vie for reputation and income, and leverage that reputation and capital in their other endeavors. The end result is a more effective, efficient, transparent, and fluid process of peer-review that provides the reader of a manuscript with additional information and nuances around the work. 

Review can also be incentivized directly by the emissions mechanism of the immutable ledger. 

Metrics

With DLT, every measure of “good science” or “good practitioner” is now open to experimentation. Countless metrics can be developed and tap into the lake of knowledge. Various institutions, curation editorials, funders, or whomever can choose to rely on any metric or combination of metrics they desire. 

One could imagine a certain set of metrics being the “gold standard” within one field of science, while a completely different set of metrics are of highest relevance for another field. Maybe one metric is important to institutions while another is used to help funders. With open metrics, a continuous debate on the intricacies and agreed upon frameworks within scientific communities, informed by practical implementation, forces convergence on metrics measuring, and thus incentivizing high-quality research practices and output.

Organizations and Funding

DLT revolutionizes how and why organizations exist. If the scientific record is maintained through cryptography, if publication is free, if access to the lake of data and knowledge is guaranteed, and if anyone can enter the system freely, access these public goods, and prove their worth through positive contributions, monoliths will struggle while open science thrives. Scientific societies will freely organize and build. Patient groups will more easily advocate and support research that might be considered too fringe under a commercial system. Funding groups will thrive on novel processes such as crowd-sourced IP ownership, retroactive funding hyper-certificates, or the fundamental incentive structures of DLT, including those of programmable money.

If the scientific record is maintained through cryptography, if publication is free, if access to the lake of data and knowledge is guaranteed, and if anyone can enter the system freely, access these common resources, and prove their worth through positive contributions, monoliths will struggle while open science thrives.

Open and Permissionless

With DLT, everyone is free to construct their own vision of the perfect system of science. It is not necessary to first be granted permission from a degree, a career path, an institution, or an authority in order to test a novel idea on how to improve the scientific process. Systems can openly integrate various aspects of other systems while participants can seamlessly shift among any system they desire. There are no gatekeepers reserving entry for those deemed worthy. 

The foundational infrastructure is locked open and prospective scientists across the globe and spanning the range of socioeconomic status can research, publish, review, and curate without much more than an internet connection and an eager mind. The hope is that never again will process and value-determination become siloed and gated behind monolithic entities. Never again will a system become mired and held back from progression.

Contact and Resources

contact@thesciencecommons.org 

https://twitter.com/TheSCInitiative 

https://github.com/DeSciWorldDAO/awesome-desci 

Conclusion

Many questions remain on which implementation of DLT will most benefit the system of science. Furthermore, DLT does not solve every problem that plagues the traditional system of science. It will take decades of experimentation, engineering, and dedication added to the decades already spent establishing open science principles and practices.

Ultimately, however, DLT guarantees countless possible implementations of the system of science. The open and permissionless nature of immutable ledgers means that all implementations can exist simultaneously while scientists, scientific communities, and the public can choose the structures by which they operate; Everyone can access the same lake of data and knowledge, build novel means of interaction, and let people choose which methods best fulfills their needs.

At the very least we must recognize that distributed ledger technology is here, it is changing how the world operates, and we need to work together to ensure the changes it makes to science are positively directly toward open practices.