This is the second part of a two-part series on reputation systems in web3. Read part one here.
Web3 rests on the idea that every internet user will possess a distinct internet identifier, such as an email address, that can be natively tied to any software and kept on a blockchain. As part of a person’s “decentralized identity,” some of that person’s online activity would then be “on chain,” meaning public and easily searchable through their individual crypto wallet.
With decentralized identity — a legible record unique to each person — one’s crypto wallet would act like a profile, akin to Facebook or LinkedIn. But unlike web2 profiles, decentralized identities are supported by hard proof: a lasting, timestamped log of a person’s accomplishments, contributions, interests, and activities to date.
If decentralized identity became broadly adopted, people could bring their whole selves with them as they move through cyberspace: their tastes and experience mirrored by what they’ve made, contributed to, earned, and owned online, regardless of the platform. This would move us nearer to how things operate in the physical world, where our belongings and reputations are tied to us, not to the places we happen to be; we can carry them along and use them however we want.
Although it may seem unrealistic now, a genuine opening here lies in using on-chain reputation for off-chain purposes. In the creator economy, for example, we could see a web3 version of YouTube where the videos belong not to the platform, but to the creator. Then the creator could move those videos to any online environment and capture any value attached to those assets. Likewise, on-chain data could help measure engagement or interest in particular creators or brands. A music artist, for instance, could quickly spot their biggest fans from platform-agnostic, on-chain interactions and give them exclusive access or other benefits. In finance, decentralized identity could make possible an internet-native credit score, which might be useful for making loans, running tenant screenings, setting credit, and more.
Web2 to web3: Moving from fleeting “likes” to durable reputation building
Unlike most web2 profiles, decentralized identity does not disappear. That means an NFT of a diploma in your crypto wallet, for example, would become a permanent academic credential. In the same way, every piece of content you post online would stay permanently tied to you, unless you decide to remove it. More broadly, public histories would make it possible to show that you were early to a trend or involved in a project before it became successful — like, say, liking Taylor Swift before she was famous or reading this article while web3 is still in its beginnings.
This durability creates new incentives for building reputation: rather than making short-lived profiles as we do in web2, web3 encourages long-range thinking. If people were enabled to build and preserve permanent identities online, we believe on-chain systems could push people to curate more carefully the reputation markers they carry into the future. In this sense, curating a permanent NFT library is higher stakes than, say, curating a stream of social media posts, because it is a reputation signal you take with you across cyberspace.
The infrastructure decentralized identity requires
If decentralized identity is the web’s future, then the contributions you gather in your crypto wallet — the articles you write, the content you curate, or the things you purchase — will matter a great deal. They will become the signals that carry who you are online and the foundation of your reputation.
But to bring decentralized identity into the mainstream, we first need systems that map people’s relevant off-chain experiences and affiliations onto the blockchain. Next we need mechanisms to standardize, process, and prioritize the stream of data that will be added on-chain. Along the way, we’ll have to address endemic problems in decentralized identity, including the absence of context around on-chain records and challenges with accessing the decentralized web.
From there, we can begin exploring the many use cases unlocked by incorporating on-chain reputation into offline activities, given the openly queryable nature of decentralized identities.
Mapping off-chain experiences and affiliations on-chain
Right now, blockchain technologies are often opaque to the average consumer, and are mostly built for financial transactions. To make decentralized identity useful and broadly adopted, it is important to create on-ramps for people to record their experiences and affinities on the blockchain, such as their educational achievements, professional successes, or fandoms. Some companies are already working on this problem. Violet seeks to authenticate off-chain, personally identifiable information on-chain. Rabbithole, Polywork and 0xStation let users record professional achievements on-chain. And Koodos helps map varied affinities on-chain by letting people turn any internet content into a customizable NFT, building a collection that meaningfully reflects who they are online. (Disclosure: Esber cofounded Koodos and Kominers provides market design advice for the company.)
There are two ways to record people’s activities and affinities on-chain: active and passive. Proof of Attendance Protocols (POAPs), which allow event organizers to create tokens that attendees can collect, are one example of active recording. When people claim these tokens, they are intentionally recording a “life event” on-chain, turning it into part of their digital identity in the process. Likewise, people may actively attest to other people’s skills, character traits, or accomplishments by sending them tokens; these endorsements can be verified through peer review or by evaluating on-chain activities.
Alternatively, when a person’s wallet connects with apps or sites, details about those interactions may be recorded passively. For example, if your wallet were linked to Discord and you earned server badges, those badges could be added to your on-chain collection.
Processing reputation data — and data standards
Once all this identity data has been stored on-chain, we’ll need systems in place to map and interpret it so it makes sense to the wallet holder and the people they interact with. Because wallets are also used for ordinary content storage and transactions, this is especially difficult: wallets collect a lot of noisy data.
For the sake of logistical simplicity (and because different aspects of a person’s identity naturally overlap and interact), it makes sense to keep one’s relevant activities and contributions in a single wallet. But no person is only a creator, builder, investor, or collector; people are many things at once. So we need some way to take pieces of their decentralized identity and present them in different digital spaces to different audiences. Imagine a design where the wallet’s full contents stayed private, but reaching a particular aspect of a person’s reputation required a specific access key.
It would also help to have more advanced versions of Etherscan — a navigator for transactions on the ethereum blockchain — that would let users see all of a wallet’s interactions within a particular category of activity, for example, activities tied to community service.
And beyond organizing the raw data, it’s important to have systems that aggregate and simplify the information inside a person’s wallet so that it becomes more easily interpretable and comparable across people. One recent example is a DeGen score, an interpretable measure of how “crypto-obsessed” someone is based on an aggregation of wallet transactions. Meanwhile, Ceramic helps applications track large amounts of mutable data and encode relevant information on-chain, letting viewers compare people’s contributions to different projects at a high level.
Managing reputation information also requires publicly accessible data standards — agreed-upon formats for, say, “what an attestation of a contribution looks like” or “how to formulate an on-chain record.” Organizing these inputs to reputation promotes interoperability, both by (1) allowing people to meaningfully combine pieces of their decentralized identity from different services and (2) reducing friction for platforms to interact with different facets of a person’s on-chain reputation.
Challenges to mainstreaming decentralized identity
Of course, there are inherent challenges to building the decentralized identity system described above. The current technology that records inputs tends to be very basic — mostly limited to transaction histories — but something as subjective and complex as identity will require much more context. Putting information in context explains why you’ve earned or bought an artefact. Earning an NFT for a contribution to a project, for example, should supply context about why you earned it and what skills you showed. Buying digital art should include context on why you bought the art, what it made you feel, or what you thought about it.
A system organized around permanence also creates challenges for escaping earlier versions of one’s identity, as credit scores and other long-horizon reputation systems today show. It can mean you never completely escape a mistake from then on.
In addition, crypto wallets offer little protection against data loss or compromise. Before crypto can serve as a mainstream store of identity and reputation, it will need the security protections we are accustomed to from other services in practice and at scale.
More generally, it is vital to broaden access to sources of on-chain reputation. At present, crypto tends to skew affluent, male, and white. If the field for building on-chain identity is not leveled, this new source of reputation will accrue mainly to the already privileged, worsening existing social and socioeconomic divides even more.
Decentralized identity will create downstream opportunities overall
Despite these challenges, we believe on-chain reputation will unlock a wide range of use cases. Some of these uses are explored below, but many are still beyond our ability to imagine right now. By moving identities on-chain, we could upend outdated financial processes and provide more equitable access to capital, rework how talent is matched with projects, and create entirely new ways to recognize and compensate labor online.
Traditional financial systems make it very hard to shape — or even access — one’s own credit history. The portability and accessibility of decentralized identities, on the other hand, could allow us to significantly disrupt how our reputations are used within our current financial systems. For example, on-chain reputations could enable an internet-native portable credit score that would unlock a variety of everyday financial use cases, including issuing loans, performing tenant checks, or establishing credit too.
Moreover, on-chain systems are easily portable across borders and applications, unlike current credit scoring systems today. Here especially, however, there needs to be strong thought given to how users might add context and curate their permanent public histories to avoid unfair prejudice. It would be important to bypass the digital equivalent of being evicted, which can hinder your ability to obtain future housing.
In the realm of talent sourcing, decentralized identities promise “open CVs” based on detailed records of professional activities and accomplishments. This could give a fuller picture of someone’s past work and provide more precise insight into their skills and overall contributions.
Although recording and storing professional activity at this level would be complex, aggregating such activity on-chain would be much faster to digest and use than static user-generated professional profiles, such as those on LinkedIn. This opens use cases around talent matching. Imagine a version of Behance in which a designer’s portfolio is filled automatically from tracked and verified projects, enabling direct, blockchain-mediated recruiting for teams seeking verified creative experience across remote markets today.
Additionally, there is an entire class of applications built around using reputation to encourage new forms of creation or contribution. Today, it is hard to tell who is a subject-matter expert and to acknowledge and reward them. With decentralized identities that record contributions across the internet, we will be better able to gauge how early someone was to a topic or trend and how much value someone is adding to a subject area. Startupy, for example, encourages experts to share startup knowledge and earn an internet-wide reputation over time in specific domains.
The wealth of information stored on-chain could even reshape markets as large as search and discovery. A range of matching algorithms would be enabled by the openly queryable nature of decentralized identities; before now, matching algorithms could only be built by whoever had access to our data. Additionally, until now we’ve only been able to judge how much value people associate with content based on basic engagement metrics, such as how long they’ve spent on an app or whether they clicked “like.” If we had a better sense of people’s willingness to pay for distinct pieces of content, we would have a better read on the scale of engagement or interest overall.
Furthermore, if every piece of content on the internet were truly composable or remixable, it could become possible to gauge which content inspires more creation, or is the most meme-able. This would offer a new way to surface content, similar to TikTok’s system. Finally, and perhaps the most disruptive of all, people themselves could become the new platforms for discovery — especially when incentivized through token models.
On-chain reputations derived from identity will create massive opportunities for business and society. But as we move through this new terrain, we will be forced to consider what permanence implies. Inevitably, we’ll need to strike a balance between what is ephemeral and what should last forever.
If we move toward an internet in which our identities are attached to us rather than scattered across siloed spaces on the web, we’ll get closer to how things work in the physical world: our online possessions will belong to us and we’ll be able to carry our full, authentic selves fully across cyberspace.
This is the second installment of a two-part series on reputation systems in web3 today. Read part one here.
Acknowledgments: Thanks to Mike Bodge, Chase Chapman, Joey DeBruin, Abigail Fradkin, Default Friend, Jonathan Glick, Andrew Hong, Jihad Esmail, Nir Kabessa, Metadreamer, David Phelps, Brenner Spear, Danny Zuckerman, Curtis Roach, and Andrew Wang for their input and insight into this topic.
Disclosures: Jad Esber is an investor in a number of NFT and DAO projects. Scott Kominers provides market design advice to a number of marketplace businesses and crypto projects, including Novi Financial, Inc., the Diem Association, Koodos, and Quora.