DeSci is a broad term used to describe a movement to overcome existing difficulties and bottlenecks in scientific research with the help of web3 tools and existing ideas and paradigms in the blockchain space.
The Ethereum Foundation gives a good summary of the DeSci space in a dedicated page on their website [1]. At the core, the benefits come down to open publishing, checks on reproducibility, decentralized funding, improved ownership of IP, and open-access data storage. As Sarah Hamburg points out in her “Call to join the DeSci movement” [2] there are unfortunately few established academic communities taking part in the debate, and hopefully this will change in the near future. There are however existing initiatives, often centred around particular scientific disciplines, such as longevity research or around a particular technical challenge, such as high quality peer review. Science is multi-facetted and encompasses so many dimensions, we here merely seek to touch upon a few particular themes to highlights some contrasts between traditional scientific systems and their decentralized counterparts.
Much of today’s research funding is managed and distributed by large governments & institutions like the National Science Foundation in the USA, which for instance had a budget of 8 billion in 2020. However, receiving funding from conventional funding bodies can be difficult, particularly for novel and disruptive research ideas as two recent investigations have shown [3,4].
This was already noted by Thomas Kuhn, who, in the Structure of Scientific Revolution claims that “novelty emerges only with difficulty, manifested by resistance, against a background provided by expectation” [5]. Nevertheless, a reason for funding unestablished and experimental research areas is given by the dispositional potential for long-term impact, both for the advancement of science itself and for the development of wide-ranging uses. For example, David Deutsch first started investigating quantum algorithms in 1985 [6], which was long before a quantum computer could be built and before quantum algorithms were practically relevant. Doing the foundational research, however, laid the foundations for the decades of research that followed and played a role in motivating the pursuit of fault tolerant quantum computing and the development of many more advanced quantum algorithms.
Moreover, smaller scientific endeavours, which include educational or early stage exploratory work may not fit into the canonical funding frameworks. Often these projects don’t require large scale grants which are usually given long in advance and actually operate on much shorter time scales. Funding smaller scale projects internationally may also incur high transactional costs or lead to legal difficulties. These challenges can lead to an abandonment of the research project in the worst case.
How does blockchain technology, or the concept of decentralization play into the funding of scientific research? On the one hand there is the possibility of improving the efficacy and transparency of fund allocation. One could for example consider particular grants which are either crowd-funded, funded by a philanthropist, a foundation or a different organization. These funds can then be allocated quadratically [7] across various research projects by conducting a Snapshot vote or a similar smart contract architecture. Disbursement of the funds could then be tied to the outcome of the vote. The infrastructure required for this is relatively lean and it can be accessed from all over the world. There would be no issues of currency exchange or lengthy cross-border bank transfers. Domain experts would be whitelisted based on their addresses and funds would be transparently allocated to the most promising research projects, as votes could be viewed on a secure data storage platform like IPFS. One would however, require a clever decentralized identifier and Sybil resistance mechanism to ensure voting occurs correctly. Moreover, one would need to be able to uniquely associate addresses with particular individuals who have the relevant domain expertise and publication records.
The nature of such grants is open-ended, they could be like standard research grants awarded to various research proposals as a proactive, upfront grant, or they could be awarded retroactively, after the quality of a particular project has already been assessed. The retroactive funding paradigm is of particular interest to the DeSci space and has generated much excitement, as this generally allows expert fund allocators to make particularly informed decisions. Funding decisions could also be tied to more quantitative metrics and be based on an objective criterion of merit, for example paper citation network metrics or GitHub repository engagement, which are recorded in an open-source manner.
It is also possible to monetize existing research results and intellectual property by creating non-fungible tokens. The monetization could lead to the creation of an entire market where research results, and IP are traded with regards to their impact [9]. This could therefore funnel funding to the most impactful projects. In the case of tying research impact to monetary value, negative impact would need to be disincentivized, for instance by the creation of a debt, but the details and difficulties in realizing this type of market are discussed at length elsewhere [8].
Another interesting demonstration of this nature was performed by p1anck [10]. It was hypothesized by a Berkeley Professor in 2004 that flavourless calories reduce appetite. Even though this hypothesis went viral, no conclusive studies were ever performed during his lifetime. Eventually, p1anck took matters into their own hands and realized this study by analyzing and aggregating data from volunteers and selling the results in the form of NFT’s for a total of $24000 which funded the study. An interesting observation was that the motivations behind buying the NFT were varied, some valued the scarcity and novelty of the token associated with the first test of Seth’s appetite theory, but some simply wanted to directly fund the study for the sake of seeing the results.
The peer review process provides essential quality control for published scientific work, but in its current form, the journal publishing industry does not optimally incentivize reviewers for providing high quality peer reviews. Although academic journals charge fees for publishing papers, reviewers are not paid for their services. Thus, it can be challenging to source highly qualified reviewers. This can be explained by noting that reviews are unpaid and conducted on a voluntary basis and reviewers are rewarded implicitly with opaque reputational accreditation rather than a standardized form of credit or financial compensation. It has been shown in a 2015 survey with thousands of participants that reviewers strongly believed that reviewing is inadequately acknowledged [14]. These structural problems can lead to two particularly undesirable outcomes; potentially good research is severely impeded because no qualified reviewers are found who are willing to review the work and publication is thus delayed, or poorly qualified reviewers are found and potentially bad research is published.
Ants Review [12] has addressed some of these issues and proposed a privacy-oriented smart contract protocol. It allows authors to issue a bounty for open anonymous peer reviews on Ethereum. A standardized process ensues, and reviews can be submitted by an anonymous third party, these reviews are then accepted or rejected by an external approver and rewards are issued in proportion to the review quality. A further extension is then to implement community-wide assessment of the reviews which are visible on IPFS.
Some scientific research is locked behind paywalls and this proves challenging, particularly for those who don’t have access to institutional subscriptions. There are various blockchain data storage solutions for openly and decentrally publishing research results, including entire datasets, IPFS, Arweave and Filecoin are particularly suited for such endeavours. dClimate, for example, provides open and free universal access to climate and weather data. Publishing entire datasets is also essential for replication which also lacks sufficient incentivization mechanisms. In similar fashion to the AntsReview protocol, bounties & grants could be issued for replicating an existing scientific result. In particular, this is of great importance in the social sciences, where according to recent surveys [13], researcher agree that they are facing somewhat of a reproducibility crisis. Financial rewards could be an important tool to increase the amount of reproducibility studies, particularly when the non-financial rewards for replicating an existing results are simply not as grand. Reproducing an existing result rarely leads to the same reputational gain as the publication of a new, more exciting result. A pilot project of this nature is pursued by the DeSci Foundation.
Molecule is a particular example of an existing decentralized scientific funding platform focussed on biotech research. It is a research funding platform with an associated marketplace accessible to both researchers looking for funding and funders of projects. Intellectual property and its development is monetized through IP-NFT’s and this makes the platform attractive for funders. Researchers can list and advertise their projects and currently Molecule has over 250 listed research projects listed and over $10M of funding in their network. According to Molecule, “The future of life science research will be driven by open, liquid markets for intellectual property powered by web3 technology” [15]. Three particular DAOs in their ecosystem are VitaDAO, PsyDAO and LabDAO. VitaDAO’s focus lies in the funding of early stage longevity research. PsyDAO is a decentralized organization furthering research into psychedelics and LabDAO is a community-run network providing access to dry and wet laboratory services to advance general research progress in the life sciences. This is particularly important if the upfront costs for setting up a laboratory environment are high, but a particular smaller scale service is required by a researcher.
Nevertheless, there are some issues with regards to many of the technological innovations here proposed. When it comes to funding scientific research by monetizing intellectual property in the form of NFTs, it is not clear whether non-scientists, or even non-domain experts can distinguish high quality projects from bad ones, or even from scams. We require robust quality control and the opinions of domain experts who can properly assess the merits of a particular project. For example, projects to be funded could be screened by verified scientists. Voting also requires Sybil resistance tools and robust governance mechanisms, such that for instance, a DeSci DAO could not be infiltrated by a group of creatonists or flat earthers, who direct all treasury funds in this particular direction as noted in [10].
Moreover, assigning funds to projects based on their impact is notoriously complicated. No practical metrics have been designed or tested for a large range of research works. Another challenge is that one does not want to incentivize the development of potentially harmful work or research with dual use potential, which could foreseeably be even easier with open grants that are paid out in crypto and voted on by a community of domain experts. Some mitigation strategies, are however to narrow the scope of particular grants, such that they focus on particular areas with little potential for harm and one could go even further and focus on funding risk reducing as opposed to risk enhancing work [16].
Furthermore, without reliable implementations of Know Your Customer (KYC) checks, it is difficult to disaggregate honest sources of funding from corrupt, nefarious, or otherwise illegal sources. This can also be an issue with regards to correct fund disbursement. In general, the legal environment surrounding blockchain-based funding is foggy, and it’s unclear where NFT and token-based funding schemes stand in court. Sourcing funds decentrally can reduce the financial burden of taxpayers (who usually foot most of the bill which funds academic research), as well as provide innovative funding sources for novel research ideas, but it also introduces significant compliance challenges. Such challenges include how those contributions are considered with respect to taxes, whether IP held in an NFT is defensible in court (and in which jurisdictions), as well as how to prevent and detect undesired activities & illegal activites like money laundering.
We’d also love to see more engagement from the traditional scientific community in improving and expanding the traditional scientific ecosystem to encompass some of the ideas that have been explored in this article [2]. It would be a shame if decentralized science ran in isolation to the flourishing traditional and centralized research endeavours that already exist and we would love to see more collaboration between accredited researchers and blockchain technology experts.
Image taken from: “Networks.” Nature News, Nature Publishing Group, 2022, https://www.nature.com/collections/adajhgjece.
“Decentralized Science (DeSci).” Ethereum.org, Ethereum Foundation, 2022, https://ethereum.org/de/desci/.
Hamburg, Sarah. “Call to Join the Decentralized Science Movement.” Nature News, Nature Publishing Group, 7 Dec. 2021, https://www.nature.com/articles/d41586-021-03642-9.
Barnard, Jeffrey. “Funding Agency's Reviewers Were Biased against Scientists with Novel Ideas.” Science, https://www.science.org/content/article/funding-agency-s-reviewers-were-biased-against-scientists-novel-ideas.
Veugelers, Reinhilde and Wang, Jian and Stephan, Paula E., Do Funding Agencies Select and Enable Risky Research: Evidence from Erc Using Novelty as a Proxy of Risk Taking (August 2022). NBER Working Paper No. w30320, Available at SSRN: https://ssrn.com/abstract=4184306
Kuhn, Thomas S. The Structure of Scientific Revolution. University Of Chicago Press, 2012.
Deutsch David. 1985, Quantum theory, the Church–Turing principle and the universal quantum computer. Proc. R. Soc. Lond. A400 97–117.
Lalley, Steven, and E. Glen Weyl. “Quadratic Voting: How Mechanism Design Can Radicalize Democracy.” SSRN, 13 Feb. 2012, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2003531.
“Toward Impact Markets.” EA Forum, https://forum.effectivealtruism.org/posts/7kqL4G5badqjskYQs/toward-impact-markets-1.
“Hypercerts: A New Primitive for Public Goods Funding.” Protocol Labs, 29 Sept. 2022, https://protocol.ai/blog/hypercert-new-primitive/.
Stencel, Adrian. “Science and WEB3: Do Non-Fungible Tokens (Nfts) Provide a Way to Democratise and Increase...” Medium, Medium, 10 Oct. 2021, https://medium.com/@AdrianoStencel/science-and-web3-do-non-fungible-tokens-nfts-provide-a-way-to-democratise-and-increase-b57b5caa246b.
“P1anck Learn More.” Planck Manuscripts, https://www.p1anck.com/learnmore.html#NFTs.
Trovò, B., Massari, N. (2021). Ants-Review: A Privacy-Oriented Protocol for Incentivized Open Peer Reviews on Ethereum. In: , et al. Euro-Par 2020: Parallel Processing Workshops. Euro-Par 2020. Lecture Notes in Computer Science(), vol 12480. Springer, Cham. https://doi.org/10.1007/978-3-030-71593-9_2
Baker, Monya. “1,500 Scientists Lift the Lid on Reproducibility.” Nature News, Nature Publishing Group, 25 May 2016, https://www.nature.com/articles/533452a.
Walker, Richard, and Pascal Rocha da Silva. “Emerging Trends in Peer Review-A Survey.” Frontiers, Frontiers, 1 Jan. 1AD, https://www.frontiersin.org/articles/10.3389/fnins.2015.00169/full.
“Molecule Protocol Home: A New Era of Drug Development.” Molecule Protocol Home | A New Era of Drug Development, https://www.molecule.to/.
Sandbrink, Jonas and Hobbs, Hamish and Swett, Jacob and Dafoe, Allan and Sandberg, Anders, Differential Technology Development: A Responsible Innovation Principle for Navigating Technology Risks (September 8, 2022). Available at SSRN: https://ssrn.com/abstract=
Exponential View by Azeem Azhar. “🔮 Sunday Commentary: Blockchain Use Cases for Science.” 🔮 Sunday Commentary: Blockchain Use Cases for Science, Exponential View by Azeem Azhar, 28 Aug. 2022, https://www.exponentialview.co/p/jocelynn-pearl-desci?s=r&utm_campaign=post&utm_medium=email.
F. Wang, et al.,"The DAO to DeSci: AI for Free, Fair, and Responsibility Sensitive Sciences" in IEEE Intelligent Systems, vol. 37, no. 02, pp. 16-22, 2022.