Staking Truth: An Immodest Proposal for Elon Musk, TWTR, and Chainlink
Musk should pivot Twitter towards a Play-to-Earn marketplace of decentralized truth, vetted by staked oracles and a staked downvote mechanism
Twitter was a failure as a public company for several reasons.
Financially, TWTR never developed an advertising model commensurate with its users’ engagement.
Reputationally, TWTR became infamous for politicized censorship while miserably failing to ship actual new product that addressed its business challenges, and morphed into a union of overpaid woke upper-middle-management combined with very weak senior management.
Technologically, TWTR failed to evolve. It spent billions of dollars on “R&D” that yielded pathetic ARPU gains, and didn’t broaden the Twitter use case beyond Tweets and Tweet “cards” visible on third-party websites, like Substack.
The resulting stasis was toxic for top management, shareholders, advertisers, and pretty much everyone outside Twitter’s censorship apparatus and its favored parties, i.e. the Biden 2020 presidential campaign.
While centralized content monitoring is clearly not good, there is a strong commercial incentive to filter content on social networks: Discussions on Twitter (and probably Instagram) revolve around creating a satisfactory experience for their top content contributors.
The first key perspective is that most people who are users only look at social platforms from one lens, namely:
1: what can people say?
— Yishan (@yishan)
Apr 27, 2022
However, anyone who has ever OWNED or RUN a social platform looks at it through two lenses:
1: what can people say?
2: the platform functioning
— Yishan (@yishan)
Apr 27, 2022
Besides pleasing their major content creators, digital platforms are required by law to police content to a certain degree (eg, child pornography hosting carries massive platform liability).
In today’s Twitter, the ecosystem could be divided into several different kinds of over-compensated vs. under-compensated users. In my opinion, the overcompensated users include:
Super-commentators: The top 5000 or so Twitter mega-celebrity and commentator accounts whose posts and replies probably account for 98% of Twitter engagement.
Reporters: the recipients of the super-commentators’ favor, who bring new verified facts around highly engaged/controversial topics.
Fact-checkers: Users who validate specific claims of reporters. Often filled by reply guys, never systematically rewarded.
Parasites / Flamers / Spammers: the haters whose activities quantifiably discourage further participation from others, especially the super-commentators, who are the golden geese of Twitter’s model today.
The “overcompensation” today comes mostly from the spamming side, which is a huge problem for people like Elon Musk who want to productively engage with productive content. However, super-commentators are overcompensated in other ways, so it’s not a systemic problem unless another platform solves this much better than Twitter does today.
Twitter, by its focus around hot-button issues and personalities, controversies du jour, and extremely rapid feedback loops, had a tendency to create excessive negative engagement (defined as explicit threats of violence to commentators) in its earlier years. My guess is that Twitter central controls have mostly mitigated this issue today.
Reply guys: Kind of like the kid at the front of the class who raises his hand every time the teacher asks a question, these people are annoyingly active in Twitter discourse, but usually well-informed, net-net constructive, and derive some sort of digital status or celebrity-engagement satisfaction from constantly riding the coattails of experts with big Twitter followings.
There are several other distinct agent roles on the Twitter network, which are arguably undercompensated or uncompensated:
Connectors: Users whose content catalyzes lots of new connections between previously disparate tribes of users would be boosted, for creating new connections (new future platform engagement) that arguably wouldn’t otherwise exist
The “middle class”: aspiring commentators who are consistently ahead of the quality-engagement-with-verified-facts curve, or the quality-engagement-with-memes curve. They tend to be unusually passionate users starting to figure out the Twitter engagement game, who’ve honed in on a niche where their commentary is rewarded.
Generic, passionate users: people who are very passionately engaged about a small set of issues, and aren’t in it for any form of status, engagement farming, etc. They spend time very disproportionate to what they get in return, and never click on ads. This type of user probably has very episodic engagement as they become hyper-engaged for a while, decide that Twitter stresses them out or wastes their time, and then leaves, or the issues that drove their participation, like March Madness or Covid lockdowns, fade away.
Finally, there’s a piece of metadata that’s missing from Twitter tweets today, which I’ll call “truthiness.”
Truthiness is objectively verified truth which receives strong engagement.
Truthiness would be determined by some objective machine — not a human.
Truthiness would have to be interoperable: just like tweets are highly interoperable across other platforms (like Substack, via embedding), so would verified data points be interoperable.
Web3 has a primitive for truthiness — oracles — which doesn’t exist in web2.
As a media company, Twitter is optimizing one equation above all others:
Quality user growth = Quality social media content = quality engagement * number of downstream connections made * “truthiness”.
Truthiness is a word I’m making up, to stand for “getting closer to the truth” or “upvoted, incremental truth.”
These contradictory goals are not served by an advertising model, especially in the type of media that Twitter specializes in: hyper-current, hyper-controversies of the day. How can Twitter rebalance its internal incentives around a functioning platform for quality, independently verified information?
Twitter’s internal economy can satisfy all these unmet needs and competing priorities via shifting towards a staked engagement token, because quality engagement is the lifeblood of Twitter. Twitter could implement this as follows:
A user slowly accrues Twitter Engagement Tokens (TWITs) via non-negative platform engagement.
You (user) can stake your TWITs to your profile to receive an overall modest boost in algorithmic ranking, or a larger boost by staking it to a specific piece of content that means more to you.
TWTR allows power-users to buy and stake as many TWITs as they want. TWITs’ boosting power would be as a function of the % of circulating TWIT supply, so over time, a single TWIT would lose value at a regular rate.
If you think you’re bringing a specific fact to the table, you could stake your TWIT on a specific fact to receive a specific boost.
You could also stake your TWIT against a specific statement, as a skin-in-the-game type of downvote.
Now, you might argue that Twitter already has a Twitter Engagement Token: your follower count. But lots of people contribute lots of activity that adds virtuous engagement without accruing followers. These people generally aren’t compensated in the current construction of Twitter’s “play to earn engagement” game.
By creating supply and demand around a native token, Twitter could bring both its financial and content economies into better balance, more optimally for content participants than Twitter does today (under a super-topheavy likes/ dislikes/ followers/ status model) and more optimally for shareholders under today’s advertisement-for-engagement model. The economy would revolve around quality content and quality engagement, not raw impressions or raw engagement. Most other platforms have found ways to balance this without incorporating web3, but for Twitter, balancing quality engagement around advertising is extremely hard, and probably very suboptimal even after expected significant optimization from its present state.
However, Twitter, as the go-to digital town square for breaking news, has an additional, unique problem: How is truth defined in the absence of a central agent?
Decentralizing truth: Bringing web3 oracles to Twitter
In crypto, smart contracts require a decentralized source of truth to be trusted. If you invest in a DAO that has a treasury of $100 million, you can’t blindly trust the DAO for stating that it has $100 million of assets. You also shouldn’t trust them if they say “auditor XYZ says we have $100 million of assets.” You need a third-party referee to validate it by pinging the wallet that the DAO calls its Treasury, and then pinging all the prices of those holdings against the reference currency that you’re using, and doing so via provably random validators on a network, to ensure that the process can’t reasonably be gamed. This function — provably impartial validation of quantitative truth — is what Decentralized Oracle Networks, or DONs, do for web3.
Chainlink probably has 70% or more market share in DONs. A piece of software, not controlled by any party, relays to a smart contract a piece of information — like the amount of funds in a wallet that’s been designated as the treasury of the DAO you invested in — across a decentralized computer network, in a way that’s extremely difficult to corrupt.
LINK has extended its model to impartial validation of off-chain data as well. For example, Frax Finance recently launched a CPI-indexed stablecoin (FPI) whose value self-adjusts after running a monthly job relaying the latest monthly CPI data point from the Bureau of Labor Statistics website to FPI. If you hope that billions of dollars will ride on that token’s value someday, you need absolute proof that its underlying data driver is being updated as impartially as possible. For this service, FRAX pays Chainlink 1 LINK per validated query, according to the job description.
Staking off-chain truth: Twitter’s chance to set the market for decentralized truth
In Twitter’s case, if a user staked TWIT around a specific fact, s/he would have to a) create an objective feed to that fact on his/her own (probably laborious, and only done by highly engaged, savvy users — call these agents “reporters”) or tweet a reference to someone else who did the heavy lifting to create that oracular truth (in which they’d get some much smaller reward).
This oracular truth would consist of a Chainlink (or other oracle, if appropriate) feed to a specific reference. In the background, TWIT would be exchanged for LINK to sustain these data feeds. After the user made the connection between the source of truth and his own statement of truth, a second, randomized layer of validators would agree or disagree with whether the designated statement of truth matched up with the endpoint data feed source. Subsequent validators could stake their TWIT on rejecting erroneous oracle<>statement links. Ultimately, validators of erroneous truth would be slashed by later counter-stakers.
A second, more ambiguous degree of “truthiness” would be created by staking TWIT for or against particular tweets (appropriate for qualitative statements). Staking in favor of a given tweet wouldn’t be any different from clicking the “like” button, but staking against would function as a skin-in-the-game downvote (if staked against). Losers of the downvote would get slightly slashed, and proceeds would go to the winners, if the losing margin were large enough to reflect a likely consensus among the TWIT voters.
Is gamifying truthiness and staked oracles actually practical, or additive to signal value?
Many users might find it extremely annoying; in which case, they’d have complete freedom to opt out of the game, and the TWIT they’d naturally receive for engaging could be reallocated to a smaller pool of higher-conviction, higher-information users who enjoy playing the game to fulfill underserved roles within Twitter’s socially-networked news network. People who opt out of the game would not accrue TWIT for any activity, or see its impact on individual posts. TWIT holders, over time, would evolve into a decentralized fact-checking organization within Twitter.
If building a decentralized fact-checking architecture runs too far against the grain for Musk, or is impractical for other reasons (spawning lots of lame, gotcha fact-checking, like many MSM fact-checkers do today) TWIT could be allocated to reward other types of behaviors instead. Does your commentary bring lots of disparate tribes together to engagement deemed constructive by Twitter HQ, resulting in novel engagements? Great, here’s some TWIT. Did you write a tweet that generated tons of embedded references off of Twitter, causing significant traffic back to Twitter’s platform? Great job, here’s some TWIT proportional to the value of that traffic.
By creating composable, verified statements within tweets, Twitter would open its platform up to new, algorithmic use cases, as well as setting a new standard in high-content news. Not too far into the future, software could react to alerts from Twitter’s oracle-verified statements in the same way that trading algorithms react to crypto prices (determined by blockchain oracles pinging multiple exchanges) today. Twitter would no longer just be “the pulse of the planet,” it would also be the collective CPU of the digitally engaged masses, a repository of oracle-verified statements and TWIT-weighted upvoting or downvoting of specific statements or sentences.
The initial complexity and cost of linking oracles to real world data feeds would probably mean that only corporate accounts who want to make emphatic claims, or dispel harmful narratives, would use it. However, adoption would be sped up dramatically if users could stake their TWITs to upvote or downvote specific oracle-validated facts, or weigh in on the truth or falsehood of specific controversies which they’re heavily invested in. Tweets in controversial topics that contain oracle proof would be significantly promoted: they’d show high user conviction (oracles aren’t exactly cheap) and would be injecting falsifiable evidence into charged public debate.
For Twitter’s “middle class” of hyper-engaged reply-guys and middle-class commentators, their engagement would be incentivized into a rewarding activity for them (verifying bigger accounts’ claims) as well as for the platform. Every time a user staked oracular truth, or de-staked oracular falsehood, they’d gain a piece of “prize pool” TWIT emissions proportional to how early they were in accurately upvoting a specific story or staking a particular oracle. Users who consistently staked oracular falsehood would get slashed, or (if slashing weren’t used) their TWIT would be inflated or taxed away in the absence of getting upvoted for staking.
With a slashing mechanism, Twitter would finally have a working downvote that would require user skin in the game to put to use. That’d be a new, potent weapon against viral disinformation.
(If you are thinking this would take all the humor out of Twitter because all humor would be downvoted, once again, you could opt out of the whole game. The goal of this game wouldn’t be to alter the old Twitter experience for people who liked the old Twitter, it would be to build a database of crowdsourced truthiness and oracle-verified truth, for users who really care about truth.)
Practically speaking, staking or de-staking truthiness would have to feel as frictionless as other expressions of approval or disapproval (the like button). Setting up an oracle to an external data point would have to be not much more complex than writing a single tweet.
Lots of Twitter content wouldn’t be eligible for this type of objective truth, which would be heavily (maybe completely) skewed to quantitative statements: “X number of people died of Covid in Mt. Sinai Hospital last week, per data-feed XYZ.” I have no idea if it would be practical (most people would say it’s impossible), but there’s an obvious need for oracle-driven fact-checking in news articles, and fact-checking independent of a particular news organization. While people don’t trust reporters or organizations to be impartial anymore, they also crave impartially-vetted truth wherever it can be found, especially within narratives around the hot-button, rapidly-evolving controversies which are Twitter’s specialty.
The objective truth validation mechanism could also be broadened to prediction markets, a micro-market for which Twitter is ideally suited (since every Twitter pundit fancies himself as a master prognosticator). I never understood why Twitter didn’t pursue prediction markets via a token; performing well in prediction markets on popular issues of the day is a much better credibility score for a Twitter pundit than a follower count. And by tokenizing rewards instead of mediating rewards via USD exchange, Twitter would avoid the financial regulations that would otherwise strangle prediction markets’ innovative capability.
This, too, is ideally suited for Elon Musk and his zest for skirting the boundaries of the law in pursuit of something cool.
Staking at a Glance
Net asset inflow into staking (separate from fluctuations in asset values of existing stakers) dropped by $436m in the past week. This figure was distorted by the Terra Protocol’s unstaking of $800m LUNA to sell for BTC. Ethereum, Solana and Avalanche led the leaderboards on the positive flow side.
Unusually, despite the positive (ex-Terra) global staking flow, the population of total global stakers also declined slightly week-over-week.
Global Staking Research
USN Stablecoin Goes Live on Near Protocol
Near Protocol has launched a decentralized stablecoin called USN, a scalable Layer 1 blockchain, according to a statement shared by Near’s DeFi arm, Proximity Labs. The stablecoin can be checked on Near's official blockchain explorer at this account address.
TRON’s Sun Vows to Rock Stablecoin Market With his Own $10B Play
On 21 April, Justin Sun released a blog post stating that TRON DAO will be taking out $10B in collateral to launch an algorithmic stablecoin and offer a 30% annual percentage yield to investors. It’s set to launch on 5 May and it’s called USDD.
Stripe to Let Clients Make Payments in USDC Stablecoin via Polygon
Payments processor Stripe will use Ethereum scaling platform Polygon to allow its customers to pay sellers, freelancers, content creators, and service providers in crypto.
Ethereum Scaling Project Scroll Raises $30 Million
Scroll, an Ethereum scaling project that uses ZK-rollup technology, has raised $30 million in a Series A funding round as it plans to launch a testnet. Scroll’s testnet is expected to launch in the second half of the year, according to co-founder, Sandy Peng.
Polygon Introduces Supernets
Polygon Supernets are dedicated, scalable blockchains that aim to ignite mass adoption of Polygon and Web3. To support this ambitious product and goal, the Polygon team have announced a $100M support fund.
Derivatives Exchange dYdX to Become '100% Decentralized by EOY'
Ethereum Layer 2-based crypto derivatives trading platform dYdX has vowed to become ‘100% decentralized by EOY’ by way of the protocol's V4 update. dYdX primarily offers perpetual contracts, which are derivatives products that borrow elements from both spot margin trading and futures trading but do not have an expiry date.
Bitcoin Suisse Launches Decentralised Finance Offering
Bitcoin Suisse has begun offering DeFi services to its clients with the addition of the Liquity protocol to its product lineup. It allows customers to post Ethereum collateral in the protocol to mint and borrow the Liquity Dollar (LUSD) stablecoin.
Liquid Staking Protocol Lido Allocates $6 Million to Ethereum Development
Lido Finance's governance forum has passed a vote to support Ethereum protocol contributors via the Protocol Guild to the tune of about $6 million.
🔔 New Staking Mondays Episode Featuring Henrik Gebbing, Co-Founder & Co-CEO of Finoa
🔔 New Staking Mondays Episode Featuring Henrik Gebbing, Co-Founder & Co-CEO of Finoa
Watch it here!
📢 New Integration for DAOMaker is LIVE
To learn more, check out the DAOMaker asset profile.
🔔 We Have Partnered Up With Stakin to Offer Stakin Merch to Our HEARTS Holders
You can now claim a Stakin T-Shirt Pack and/or a Stakin Summer Pack right here.
Learn more about the partnership → https://www.stakingrewards.com/journal/claim-stakin-branded-merch-for-hearts/
Staking Rewards is Hiring Researchers, Engineers and a Head of Marketing
We are looking for proactive, intelligent, and determined people to join our team across all functions. Positions are fully remote.
We are looking to hire 2-3 more Researchers to join our current team of 3; on a mission to become the unprecedented experts on staking data and economics. Apply here.
We are looking to hire Software Engineers with Web3 experience to develop the data aggregation pipeline for Staking Rewards. Apply here.
Furthermore, we are looking for a Head of Marketing to lead our marketing efforts. Apply here.
If you know anyone who would be a great fit, please ping us. We offer a generous referral bonus.