DeSci, or ‘Decentralized Science’ is a community using crypto and web3 to address some of the deep-seated problems in science. In January, I had the good fortune to attend the DeSci London conference to find out more. Ever since then, I’ve been reflecting on what I’ve learned, the challenges and opportunities facing DeSci, and how I think the space could develop.

My conclusions are somewhat heretical: DeSci is right about the problems facing science, but web3 is only a small part of the solution. If we care about impact, we should loosen the blockchains.

I’d like to explain how I got here, why I don’t think it’s oxymoronic for DeSci to dabble in other technologies, and suggest practical ways science can embrace the values of crypto, without getting so tangled in web3.

For an intro and other perspectives on DeSci, checkout:

• Ethereum.org’s definition
• A Guide to DeSci, the Latest Web3 Movement
• Decentralised Science — a cure for the Science sector’s woes?

Bright Sparks: DeSci London

In-spite of the headwinds of a global economic slowdown and (another) cryptocurrency crash, the decentralised science community appears to be growing, and the DeSci London conference is a testament to the the movement’s vigour. Hosted in the prestigious Crick institute, the organisers put together an impressive range of speakers and backers, pulling hundreds of guests across two days.

My highlights from the event included a keynote presentation by Vibe’s Alok Tayi; a great demo from lateral.io; an investor panel hosted by Local Globe’s Julia Hawkins; and a disease workshop featuring the pioneering Ethan Perlstein. The conference also featured a DeSci Podcast premier with Vincent Weisser and Vitalik Buterin.

The quality of guests was also striking, with many accomplished scientists, engineers, entrepreneurs, and investors. It might seem odd to laud the seriousness of a community, but the tone and quality of conversations in DeSci feels very different from those I witnessed during the crypto manias of 2017/18 and 2020/22.

In the end, although I remain skeptical about many of the ideas falling under the DeSci umbrella, the community’s talent is undeniable. But if these sparks are to kindle a flame, and then a revolution in science, passion and expertise must rapidly translate into impact.

The tail wags the Dog

Listening to the presentations, talking with guests, and reading around DeSci, two themes leap out: (i) Science is broken, and (ii) Crypto is the answer.

The first claim is compelling, and is by no means exclusive to the movement. Whether it’s the allocation of funding, the accessibility of data, the management of IP, or the undue influence of institutions, there are a web of problems hobbling our ability to make scientific progress.

But is blockchain always the answer? It often felt jarring when a conversation or presentation would turn to ledgers, smart contracts, DAOs, or decentralised compute. Suddenly the technology would seem more important than the problem itself, and the analysis would retreat to shibboleths like ‘centralisation is bad’. It is as if the tail were wagging the dog.

There are good reasons to think crypto can play a role in improving science, and you might puzzle at my complaining about the prominence of web3 in a community dedicated to decentralisation… But before going all-in, we need to be realistic about what the technology can and can’t do.

Here is none other than Vitalik Buterin, in 2017:

In the thread he goes on:

How many dapps have we created that have substantial usage? …How much value is stored in smart contracts that actually do anything interesting?…

…The answer to all of these questions is definitely not zero, and in some cases it’s quite significant. But not enough to say it’s $0.5T levels of significant. Not enough.

5 years later the market cap of crypto is over 1 trillion, and while the technology has advanced, how much has really changed? Not enough.

Observing DeSci, I’m left with the impression of a community comprised of talented (and frustrated) people who have coalesced around the ideal of decentralisation, but who risk being constrained, and ultimately short-changed by the reality of web3.

Blockchains haven’t scaled, decentralised software development remains difficult and niche, and most importantly of all: wallets, smart contracts and DAOs are near-impenetrable for the average user.

The Uncanny Valley of Progress

In other sectors, crypto maximalists have argued that our present institutions are not just flawed, but irredeemable, and that we need new technology to overturn a corrupt order. Speaking with members of the DeSci community, the view is less militant: web3, they argue, is critical as an accelerator and enabler. As one presenter put it:

“The world is too complex for legacy systems and coordination mechanisms”

To my mind, this is a misapprehension. We don’t need web3 to build ‘Decentralised Autonomous Organisations’ (DAOs), create transparency, or develop innovative reward systems.

We have many examples of independent, decentralised organisations that thrive without cryptographic governance — just look at open source projects like the linux kernel. Equally, Wikipedia, PubMed and the NIH Genome library show that open access to information is possible without the need for convoluted decentralised compute or storage. And finally, we have innumerable examples of companies, individuals and institutions working together using contracts and the legal system to create value for all.

Indeed, what’s instructive about these examples, is that although they are non-web3, they have made extensive use of cryptography to solve many of the problems crypto advocates care about: hashes, public keys, and signatures to ensure the integrity of files, and of course SSL to validate the identity of websites. They use these tools because they add real value to their users and mission.

The stated goals of DeSci include accelerating scientific collaboration, increasing access to research data, and improving economic incentives. I only argue that as we make technology choices, we should be optimising for these goals, first. Whether it’s choosing between Postgres vs MongoDB, or OAuth vs Web3, the benefits must justify the cost.

✅ IP-NFTs

Let’s say we’re running a university lab and have identified a potential treatment for an important disease. We want to hand-off the asset to industry so that it can be developed, while retaining a slice of the IP, so that we can share in any potential upside.

This is precisely the challenge DeSci company Molecule is tackling, and they have done so using web3 — but not only web3. Their service seeks to pair a traditional legal contract with an IP-NFT primitive. This hybrid approach answers the obvious question: “Would Pfizer care that the university holds a virtual token entitling them to 5% of revenue?” — Not unless they also hold a corresponding legal contract!

Discord > DAOs 🤷

Let’s apply the same reasoning to DAOs (in the true, web3 sense). They are at-once ultra-secure, ultra-transparent, ultra-cool …and so hard to understand that most people cannot join or engage with them.

…And the same can be said of decentralised compute and storage: These tools do not add enough value to scientists or scientific research. Moreover, their byzantine complexity is a barrier to the wider understanding and adoption of DeSci’s principles. An unfortunate irony, since one of the rallying cries of DeSci is to remove obstacles. I experienced this many times speaking with non-technical attendees in London. Several, including patient advocates, were clearly baffled by the jargon and just wanted to know more about the outcomes.

These deep UX and DX challenges are seen across the crypto ecosystem, but is science the best sector in which to pioneer all the solutions?

Instead, I want to argue that the discord servers of DeSci DAOs are far more interesting and important than their smart contracts and tokens. It is in these chatrooms that like-minded researchers will connect, collaborate and move science forward. Engagement is everything.

Tokens & Fiat

Economic opportunity sits alongside idealism as the other engine of DeSci. 2020–22 saw a boom in web3 investment from venture capital firms, and many of the players in the space receive funds from crypto orgs such as Protocol Labs.

The arguments against crypto tokens and especially ‘utility-tokens’ are well-rehearsed (e.g. here and here), and I won’t repeat them. Instead I’ll simply observe that you can’t mint a token or NFT and declare success — where is the impact? Creating communities and platforms is hard graft, it requires people, and can’t be accomplished by fiat.

Summary

Technical purity is a false economy when it is purchased at the cost of impact. I see a risk that many DeSci outfits will get trapped in an uncanny valley of progress: Nobody will care that a research org has a smart-contract constitution if it only has 10 members. Likewise, nobody will care that your data is hosted on IPFS, if the experience is indistinguishable from S3. The opportunity costs we pay for these optimisations is vast and unnecessary. We should leverage web3 like any other technology — when the benefits justify the costs.

So, how can web3 benefit Science?

With caveats, and with emphasis on mainstream adoption.

1. New Funding Mechanisms

As described eloquently elsewhere, there are interesting opportunities to change how scientific research is funded using crypto principles. These opportunities are not dependent on the web3 stack, but since web3 has formalised many of the concepts, such as quadratic funding, it may play a helpful role. VibeBio is a company exploring this space.

But we must remember that the objective is not to allocate funding via blockchain, it is to make science better. This means outcomes such as:

• Increasing the total amount of funding available.
• Widening the pool of funders and recipients.
• Shortening allocation time.
• Enabling more useful research to be done (e.g. replication studies).

Success must be judged on these terms.

Organisations who allocate funds could do so with the help of web3 scaffolding, but these new funding systems must also be backed up by robust legal contracts if they are to gain traction. Without engagement from existing institutions, DeSci’s impact will be very limited indeed.

2. Intellectual Property

As with funding, this opportunity has been well-explained by others. There is every reason to think that IP-NFTs (exemplified by the work of Molecule), could change the landscape of scientific innovation by providing paved roads for the allocation and management of IP. This kind of approach might even be extended to patient data.

And again, it must be understood that these opportunities are only possible with a hybrid approach: Off-chain legal contracts supported and augmented by on-chain smart contracts. Web3 must be a tool, not an end.

KPIs for DeSci

Given the above, why don’t we create some KPIs for DeSci’s adoption and impact? Here are some potential examples, assuming a bio-tech lens:

• Funding allocated per quarter ($M)
• Institutional funding via DeSci ($M)
• Phase IV > III > II > I clinical trials funded per quarter
• Questions Answered > Papers Published
• Replication Studies as a % of all studies
• Number & Value of IP-NFTs purchased by Pharma
• Discord Monthly Active Users
• On-chain transactions per quarter, by DAO

What about web3 data storage? Peer Review? Reputation?

At best, these web3 solutions would be ‘nice to have’. But when we step back and look at the bottlenecks and challenges facing scientific research as a whole, none of them come close to high-priority. And so even if the technology could be made to work, and even if they gained adoption, none would be high-impact or high-value.

The future of Science: beyond web3

The ideology of decentralisation has so many exciting applications that it’s easy to get lost down a crypto rabbit hole. We can get carried away talking about new structures and rulesets, that we don’t pay enough attention to the work of science itself: being in the lab, running experiments, testing new techniques and technologies, processing data, coordinating with others, and writing up results.

Yet, these are all things DeSci advocates have every right to care about. Web3 can help support alignment, but it doesn’t deliver collaboration, nor can it increase the yield of a plasmid, or boost the speed of a data pipeline. Whether it’s in pure software, AI, or robotics, there are huge opportunities for technology to accelerate scientific research beyond web3. Transforming science depends on us seizing both types.