TCRParty Wants to Pay You for Finding the Smartest People on Twitter

Early last month, a deceptively simple novelty called TCRParty emerged from the bowels of Joseph Lubin’s ConsenSys. In a nutshell, TCRParty is a popularity contest—a list of 99 top Crypto Twitter accounts that are regularly retweeted by the core account @TCRPartyBot. But the list isn’t compiled by an individual or a committee: Instead, literally anyone can start a TCRParty account, then use cryptocurrency to attempt to “bid” to add or remove Twitter accounts from the list.

Though it launched on the Ropsten test network, the Ethereum-based Twitter game quickly amassed more users than some of the top live dapps, or distributed applications, on Ethereum. More than a month later, the experiment is still going strong, and its creators are working to refine it into a working demonstration of the big ideas behind it.

Those creators are part of Alpine, a ConsenSys unit that provides token design and related services to outside customers. Gregory Rocco is Alpine’s strategic lead, while Steve Gattuso oversees the tech side of the operation. TCRParty is effectively a sideline to their main gig, though the attention it has gotten can only be good for business.

When we met for coffee in mid-February, Rocco and Gattuso told me that TCRParty had amassed about 400 regular users since its launch earlier in the month. By comparison, relatively high-profile Ethereum dapps like Augur only have around 30 or 40 users per day. TCRParty, of course, has the huge advantage of its frictionless Twitter interface—but that’s kind of the point.

“The one thing we do not want to compromise on,” says Rocco, “is the user experience and user interface. We don’t want users to set up a Metamask, remember their seed phrase, buy ether, send the token with ether, figure out how to use it in a foreign interface. Screw that. We wanted the whole thing to be abstracted behind Twitter completely.”

Get the BREAKERMAG newsletter, a weekly roundup of blockchain business and culture.

Users can send commands like “balance” to the @TCRPartyVIP account, and they’ll be told how much of the game’s token—TCRP—they have in their wallet. They can also “nominate” new Twitter accounts to be added to the list, “Challenge” existing list members for removal, and vote for or against nominations and challenges submitted by others. Voting costs 50 TCRP, while nominating and challenging both require “staking” 500 tokens. And that’s where things get interesting: Those who successfully nominate or challenge get their stakes back, plus more tokens, meaning that good curators can actually earn money (albeit, for now, play money) by being active users of the system.

This is, Rocco and Gattuso told me, was all drawn from an idea championed by another ConsenSys staffer, Mike Goldin. Starting with a Medium post in late 2017, Goldin has outlined the concept of a Token Curate Registry (hence “TCRParty”), and drilled into hypothetical mechanics. Goldin theorized, broadly, that TCRs’ voting and staking mechanics would create economic incentives to collate collective insight.

“Mainly the TCR pitch,” says Rocco, “is getting rid of central arbiters of information controlling registries that have meaningful value. Like, why is Princeton the body that controls what the top hundred colleges are? How do you evaluate lists independently? How do you have democratic curation by a wider body?” Collective intelligence could be pulled together into something smarter than a single central arbiter could ever be, as well as possibly tamping down bias and corruption.

Putting it all on the blockchain, meanwhile, guarantees that the system is honest. “You don’t have to worry about censorship,” says Gattuso. “If I decide to make the [Twitter] bot malicious and start censoring people, people can very easily circumvent me … I literally can’t cheat on this right now, even if I wanted to.”

The TCR concept is arguably a cousin of prediction markets like Augur, which invite betting on future outcomes. In a TCR, you’re doing something akin to “betting” on whether other members of a community will agree with your support for a person or entity’s inclusion on a list—but you can also lobby to make success more likely, which is generally harder in a prediction market.

Similar betting/voting systems are being used in at least two other high-profile blockchain projects, TruStory and Civil. But where TCRParty has started with a simple curated Twitter list, Civil tried to leap directly into the stratosphere by both establishing a network of news operations, and linking them together by a tokenized voting system that would let users punish outlets for publishing inaccurate or otherwise bad content. The project’s complex, grandiose vision almost certainly contributed to the failure of its high-profile token sale last October.

To Rocco and Gattuso, Civil—and by extension TruStory—also represent a different kind of overreach: their focus on establishing an absolute truth. That impulse, seemingly in response to online “fake news” in the wake of the 2016 election, may be noble. But TCRParty instead accepts that human knowledge is always imperfect. “Personal biases will always reign true in new situations,” says Rocco. Gattuso points out that the early TCR list was “pretty Ethereum-focused,” simply because users from that ecosystem were among the first to try it out. Asking the crowd to rank things, instead of rule on the binary truth or falsehood of facts, lessens the impact of that inevitable imperfection.

In fact, Rocco and Gattuso aren’t even presuming that their core concept is viable. “We’re definitely skeptics of TCRs,” says Rocco. “And that’s kind of why TCRParty came to be. We want to test these assumptions,” whether social or technical. “Let’s do this on a small scale and slowly build it up, and [find] the faults.”

There’s one particular element missing: TCRP tokens on testnet aren’t worth real money. “The one thing we can’t test right now are economic assumptions,” says Rocco, “because it is a testnet launch.” The economic incentives that theoretically motivate people to participate in a TCR or similar system are a double-edged sword: Once TCRP can be purchased on the open market, a bad actor willing to spend real money could exert influence not derived from their own insight.

There was a partial preview of what that might look like, thanks to user Colin Platt of the Blockchain Insider podcast. “He broke the [TCRP] faucet,” says Gattuso. “He got a ton of tokens,” giving him huge sway over the rankings. “I told him he needs to be a benevolent Whale God, and so far he has been, except for kicking me a bunch of times.”

On top of exposing technical bugs, those types of shenanigans help give Gattuso and Rocco a preview of the worst behaviors to account for while tweaking the overall incentive system. To help prevent pay-to-win scenarios, they won’t be selling tokens, but are still thinking through another way to distribute them. A deeper guard against a big player throwing their weight around, they say, is economic. “If they want to troll the list,” says Rocco, “The value of the list will then technically go down, which would devalue the tokens.” Of course, that assumes a rational Whale God—and as someone once said, some men just want to watch the world burn.

Rocco and Gattuso are still thinking through how to migrate TCRParty to the Ethereum mainnet, while avoiding an even bigger threat than a malicious Whale God: transaction fees. “The idea we’re flying around now is essentially a sidechain implementation,” says Rocco. “We’d have an Ethereum sidechain where we’d mint all the tokens as ERC-20s, then peg them down to the sidechain. The sidechain is a lot cheaper to interface with, the validators are a lot cheaper to pay. We’d have TCRP [on mainnet] as an ERC-20, and on the sidechain it would be, like, xTCRP, the sidechain token that represents one TCRP.” Rocco says users who wanted to “cash out” their TCR tokens would be able to move their holdings to mainnet, but couldn’t use them to interface with the TCRP contract after that.

There are other key elements missing from TCRParty in its current preliminary form, above all a waiting list function for new nominees. Currently, to add someone to the list, a user has to time their nomination meticulously, sliding in just after a list member has been kicked, but before anyone else has been nominated. That quirk is likely keeping the list from being as good as it could be: It features some heavy-hitters in the space (Andreas Antonoupolos), but is missing other obvious names (Binance CEO Changpeng Zhao) in favor of more obscure figures.

But at the very least, though, TCRParty does seem to be made up of people interested in crypto—despite having no active, central curator, the list hasn’t been successfully “broken.” Gattuso and Rocco even tested this by nominating Donald Trump and Barack Obama for the list, and both were quickly voted off.

That’s a solid foundation to start from, and the TCRParty team acknowledge they have a lot of building left to do. “There’s a lot of different avenues and different levers we can pull to make this more interesting,” says Gattuso. “But we haven’t even scratched the surface yet, because we just want to get this out there.”