Halting Quicksilver and Rewilding Solana

Lucas Griffin -

Let's be real: With all the momentum behind large, centralized companies co-opting the "Web3" narrative for growth, last year felt, in many ways, like a step back in the movement to decentralize the Web.

Yet, as we look to 2023, we remain optimistic. We are, after all, an infrastructure company, and, despite how hollow (or outright criminal) many "Web3 products" were in 2022, we're still inspired by the progress made last year by blockchain protocols and the communities behind them.

Here are some of our personal highlights:

We Went Live on Three New Protocols

Last year, we started operating in the active set of validators on three new mainnets:

  • Agoric: A new home in the Cosmos for JavaScript smart contracts
  • Quicksilver: Liquid staking across the Cosmos (more on this one below)
  • Stargaze: NFTs in the Cosmos

Three new chains may not sound like much, but, in a year as dynamic as 2022, less was more. We take a principled approach to supporting new networks, and, while these three chains are all quite early, each shows some promise in helping to democratize the future of the Web.

Notably, they're all part of the Cosmos. As we've talked about before, the Cosmos has a history of innovation, and last year kept up the trend. New IBC-enabled protocols launched and existing protocols (like the Cosmos Hub, Osmosis, and more) took exciting steps forward. With Interchain Security (ICS) coming soon, 2023 is shaping up to be another banner year.

As always, we'll be on the lookout for other promising ecosystems to support as well.

We Partnered with Friends to Create More Opportunities for Indie Teams

In May, we invited a group of our most inspiring peers to re-launch Chainflow's Staking Defense initiative as the Staking Defense League, a coalition of independent infrastructure operators working together to onboard and support new blockchain infra teams. You can learn more and get in on the action yourself in our Discord.

Then, in June, we joined friends ranging from academics to node operators in co-launching the Validator Commons, an initiative to make blockchain governance more inclusive, accessible, and effective. You can learn more and get in touch on Twitter or the Web.

We Connected with Our Communities around the World

In 2022, we racked up more miles than ever, as we joined events across four continents:

Opportunities like these reflect just how much is possible for small, independent, and values-driven teams like ours. And that's only a handful of our highlights from 2022. As always, we feel grateful for the chance to do what we do, alongside the people we do it with.

It's a new year, but our mission remains the same: build the infrastructure that powers a more inclusive, equitable, and fair digital economy. Some parts of that infrastructure will be technical, others will be social; we're looking forward to all of it in this next year.

Let's jump right in with updates from two of our most dynamic protocols: Quicksilver and Solana.

Halting Quicksilver

Quicksilver Comes under Attack

Last month, we joined dozens of other validator operators in launching Quicksilver, a new protocol for liquid staking in the Cosmos. In our previous piece touching on the launch, we noted:

Like some other Liquid Staking protocols, Quicksilver uses an algorithmic system to decide where staked tokens go. One of the considerations in that algorithm is the decentralization of stake, meaning that Quicksilver should—in theory, at least—distribute stake more evenly across all active validators over time, helping to improve decentralization in The Cosmos. While we're supporting the project out of optimism for that impact, we'll keep you posted on how things play out.

In the short time since we published that, things have played out rather surprisingly.

Quicksilver is designed to facilitate the liquid staking of tokens from other Cosmos networks—such as STARs from Stargaze or ATOMs from the Cosmos Hub. In order to activate liquid staking functionality for a given token, the Quicksilver protocol must first go through an "onboarding process" to formally integrate whatever network that token lives on. This onboarding process relies on Interchain Accounts (ICA), a Cosmos feature which allows for an account on one Cosmos chain to natively control an account on another.

Upon launching, the first network that Quicksilver planned to onboard was the Cosmos Hub. And the plan was straightforward: On December 23rd, Quicksilver would activate ICA to create a Cosmos Hub account, and the protocol would control that account to enable liquid staking functionality for the Hub's native token, ATOM. There was just one problem: The process was so straightforward that an attacker was able to exploit some outdated software on the Cosmos Hub to anticipate, and take control of, the Hub account that Quicksilver created. By the time the community caught on, thousands of ATOMs had already been sent to the compromised Hub address.

That's the short version of a long story, and you can get the full details in the Quicksilver protocol team's official report. To help shed some light on how infrastructure teams like Chainflow played a role in resolving the incident, we'd like to focus on one detail in particular: how the Quicksilver community stopped the attack.

Pumping the Brakes on a Blockchain

When the Quicksilver protocol team recognized that the attack was taking place, they moved to initiate something called a "chain halt," an event that would stop the Quicksilver blockchain from adding new blocks—effectively pausing any activity on the network. Halting the chain would both prevent users from sending ATOMs to the compromised address, and allow the Quicksilver protocol team to update the network's code without causing collateral damage to any active users. But actually getting a blockchain to stop isn't easy—in fact, the whole point of a blockchain is to be a system that's really hard to shut down. Unlike on a centralized network, the team building the Quicksilver protocol didn't have the ability to unilaterally pump the brakes. Instead, they needed to convince the decentralized system running the Quicksilver chain to pump the breaks together.

That's where we, and the other Quicksilver validators operators, came in.

On a Proof-of-Stake network like Quicksilver, the decentralized system running the chain is made up of "validators," machines that run the network by doing things like, for example, verifying and adding new blocks to the blockchain. To carry out a chain halt on a Proof-of-Stake network, you need to convince a critical mass of the teams operating those validators to agree to simultaneously pause them. But, what defines a "critical mass" of validator operators?

A given validator's influence on a Proof-of-Stake network is proportional to the amount of tokens that have been staked to that validator. Put roughly: The more stake that a validator has, the more influence it has on the network. The most acute way to measure that influence is by block production; the more stake that a validator controls, the more blocks it gets to add to the chain. Critical mass for a chain halt, then, isn't so much a critical mass of validators as it is a critical mass of the stake controlled by validators.

The magic number for that stake on Quicksilver, and many leading Proof-of-Stake networks, is 33.3: The consensus rules for the network are set up in such way that, if validators that control collectively more than 33.3% of all the stake on the network decide to halt the chain, then the chain will halt. This is simply a threshold built into the network's code. It's also a social convention that's been adopted by Ethereum, Solana, the Cosmos ecosystem, and other Proof-of-Stake networks.

So, when Quicksilver moved to initiate a chain halt, they needed to enlist the operators behind validators controlling at least 33.3% of the network's stake. On more mature chains, we can see a kind of absenteeism, whereby validator operators with more stake (like large, centralized exchanges) are less engaged with the goings-on of the network over time. As you'd imagine, coordinating a critical mass of checked-out operators can be hard. Fortunately for Quicksilver, its chain had only just launched, and the network's validator operators were still very much plugged into the day-to-day events of the protocol.

Like we've noted before, much of crypto is powered by group chats and Discords of independent infrastructure teams. Quicksilver's chain halt was no different. Alongside the Quicksilver protocol team, we joined the other Quicksilver validator operators in the Quicksilver Discord to get the collective infrastructure community up to speed on the attack, and coordinate, with surprising speed, shut-down procedures for validators controlling stake well above the 33.3% threshold. On December 24th, the chain shut down, ending the attack within a day and allowing the Quicksilver community to fix the issues that led to it.

Then, on January 3rd, we joined the other validator operators in relaunching Quicksilver. Plans to reattempt the Cosmos Hub onboarding process are currently in the works.

Decentralized Infrastructure, Strong Chains

Ideally, Quicksilver would have avoided this attack altogether. Nonetheless, the network's response feels hopeful, with the infrastructure community demonstrating that we can coordinate quickly and effectively in the event of a catastrophe. Looking forward, however, we want to highlight a couple of points about blockchain resiliency—for Quicksilver, and for the Proof-of-Stake ecosystem at large.

The first is the critical importance of the people who operate your blockchain infrastructure. Whether keeping up-to-date with the latest patches and upgrades or responding quickly to emeergency situations like a network attack, these operators are responsible for keeping your networks running. It seems that users can take for granted that blockchains are "immutable." But, they're only as immutable as their infrastructure. And, especially on rapidly-evolving networks like Quicksilver, that infrastructure requires a lot of hands-on attention. When you're learning about a new network, make sure to investigate who its validator operators are. Block explorers like Mintscan make this easy by hosting dedicated lists of networks' active validators.

The second point we want to highlight is the importance of decentralizing the infrastructure behind your networks. In the Quicksilver incident, the community coordinated a chain halt to protect the network. But, you can imagine a scenario where a bad actor moves to halt a chain for more nefarious purposes, like holding the network's assets hostage. To prevent these kinds of attacks, a network needs to make it difficult for bad-intentioned parties to control more than 33.3% of its stake. The simplest way to do so is by ensuring that stake is well-distributed across validators operated by many different good-intentioned parties.

There's a special name for the smallest number of independent parties that can halt a chain: "The Nakamoto Coefficient." As a general rule-of-thumb, the higher a blockchain's Nakamoto Coefficient, the more immutable and resilient the network.

We talk about this a lot, but one straightforward way that you can help to raise the Nakamoto Coefficient on your networks is by staking with smaller, independent validators. Consider it the next time you stake, on Quicksilver and beyond.

Rewilding Solana

Moving Past Old Narratives

Solana has had a strange go of it. The network rose rapidly from "alt-L1" obscurity to become one of the highest-valued, most buzzed-about blockchains within months of its "mainnet-beta" public launch in 2021. And, for as long as the network has been live, it's been dogged by accusations of being a centralized wolf in decentralized sheep's clothing. But, what, exactly, have critics meant by that?

In our framework "The Layers of Decentralization," we lay out the two key dimensions that define the decentralization of a Proof-of-Stake blockchain, like Solana:

  1. Capital: The money required to make the network run
  2. Infrastructure: Where and how the network runs

In terms of capital, Solana has frequently been crticized as a "VC-chain," one that distributed an outsized amount of the network's native token, SOL, to Venture Capitalists and other investors. The most famous of these investors was the FTX-Alameda chimera, which bought a lot of cheap SOL early on, and touted the investment rather publicly. Critics argue that these kinds of high-profile investments artificially pumped the value of SOL through its previous highs.

Now, though, the fiat value of SOL has fallen, alongside the declining overall market and the ongoing implosion of FTX-Alameda, more than 90% from its peak. Obviously, you would not like to see the value of your decentralized system be so closely tied to the fortunes of so few participants.

But, despite the loss in paper value, there's a more optimistic take on the FTX-Alameda fallout. As our founder, Chris Remus, wrote back in November:

Ethereum founder Vitalilk Buterin later shared a similar sentiment:

From the perspective of decentralizing capital, it feels pretty hopeful to see Solana lose some of its biggest whales.

In terms of infrastructure, Solana has been called a centralized database masquerading as a blockchain. One popular meme rebrands the network as "Sqlana," in a nod to centralized SQL databases. But, assessing the decentralization of blockchain infrastructure is as much about narratives as it is about numbers. And, as we've tracked in our own reporting, Solana has quite a compelling narrative for decentralization in some areas, even measuring among the most decentralized blockchains.

One area wherein that narrative is less compelling, however, is governance. Many Proof-of-Stake protocols have adopted some version of a community governance process, wherein token-holders can propose, and vote on, initatives to steer the future of the network. Solana, meanwhile, has kept its future largely in the hands of its core protocol team, Solana Labs, and its primary community organization, the Solana Foundation. In response, some observers have called to mind the adage, "If you want to go fast, go alone. If you want to go far, go together."

Solana has indeed gone fast, embracing a kind of Hot Rod spirit to push the technical envelope for L1 blockchains. And in a new piece detailing its outlook for 2023, the Solana Foundation highlights the following technical areas as top-level focuses for the coming year: Mobile, Reliability and Resiliency, Programmability, Performance, and Security.

Even though "governance" failed to make the list, there are signs that the Solana community is ready to embrace a more inclusive spirit in 2023.

The Solana Community Looks to the Future

The first sign comes from the development of the Solana protocol itself. Historically, few developers have been allowed to propose or implement changes to the software that runs the network. But now, we're beginning to see new opportunities for more participants to contribute.

At November's Breakpoint event, we saw the start of informal discussions around opening up governance on the protocol. And today, we're beginning to see activity increase around the Solana Improvement Document (SIMD) process, currently the most formal avenue for Solana developers to propose and discuss (if not exactly vote on) changes to the protocol.

The second sign that Solana is ready to "go together" comes from the social initiatives that elevate the network beyond the protocol. Solana boosters have often touted a vibrant community of "builders" as one of the network's competitive advantages over other L1s. Much of the most visible work to foster this community has come from the Solana Foundation directly, in the form of its global "Solana Hacker House" program. But now, we're seeing that start to change, with independent community members developing their own social programming.

For example, this week, Chainflow is joining dozens of other teams from across the Solana ecosystem in hosting the Sandstorm Hackathon, a fully-remote, independent event to foster community-driven innovation on the network. Led by two Solana teams, LamportDAO and Helius Labs, Sandstorm is an ambitious effort to re-energize developers on the protocol after the dramatic end to 2022. The Solana Foundation has joined as a sponsor, but the event seems to be a refreshingly grassroots initiative.

We'll share more about Sandstorm as the hackathon develops. In the meantime, check out our "Decentralized Infra" track, and consider joining to help decentralize Solana while competing for our prizes: $2,000 in USDC and six months of free enterprise-grade RPC access. You can learn more and sign up at the Sandstorm Discord. The event ends on January 23rd.

BONK, Briefly

And then there's BONK.

"Memecoins" are cryptocurrencies with no inherent utility beyond speculation. "Dog Coins" are a subcategory of memecoins that are, well, named after dogs. With that in mind: "Bonk Inu" (identified by the ticker "BONK") is a new Dog Coin on Solana that has, in recent weeks, drawn a lot of attention from the network. (Some people have been calling it "Solana's first Dog Coin," but that feels like Samoyedcoin erasure.)

The Decentralized Exchange (DEX) Orca, for example, has seen quite a bit of recent action from BONK. And while it's not unheard-of for a memecoin to catch fire for a time, it feels unsual for one to espouse a kind of righteous, populist indignation:

What does BONK mean in the face of Solana's movement toward a more inclusive and democratic future? Who knows. Some signs suggest that the winds may already be changing for the token.

Regardless of what happens with BONK, or any single project on Solana, it's clear that a shift is taking place on the network, toward a more community-led spiri  . As Solana supporters since before the network's public launch, we're excited to do our part to center decentralization in these important next steps.

Eyes Peeled

Here are a few threads we’ll be keeping our eyes on over the coming weeks:

Livepeer Integrates with Bundlr for Decentralized Video Storage and Streaming 📺

We've previously highlighted Livepeer (on which we've operated since 2018; find us here) as a potential player in the future overlap between Crypto and AI. In the present, meanwhile, Livepeer's primary application concerns video transcoding, the process by which streaming video is made viewable in different formats. Given that streaming video is the largest single driver of Web traffic, there's a lot of transcoding going on around the world at any given moment. This is a big opportunity space to work within.

Last week, Livepeer shared some details about their new integration with Bundlr, a decentralized storage platform. Esentially, the integration will leverage Bundlr for the storage of video files, and Livepeer for their transcoding.

This creates an interesting synergy that opens up more of the streaming video stack to blockchain-backed decentralization. We're curious to see how partnerships like this catalyze (or fail to spark) the adoption of more decentralized infrastructure for the Web.

Check out Livepeer's recap of a recent chat with the Bundlr team, here:

Flow Considers the Onboarding Experience 👩‍👧

Flow is an interestingly consumer-focused blockchain (on which we've operated since genesis in 2020; find us here) that powers, among other things, the trading card-like "NBA Top Shot." The network is notable, from a decentralization standpoint, for a taking a "decentralize over time" approach.

This, then, feels noteworthy: In a recent piece, the Chief Architect of Flow ponders a "hyrbid" custody model that would allow users to start with a custodial account, then transition to self-custody. You can read the full post, here.

While, in many cases, the "decentralize over time" approach is a bit of "Decentralization Theater", Flow has a track record of actually doing it. Last year, for example, the network finally launched permissionless smart contracts.

This latest custody proposal is, of course, speculative, but we'll be following along closely. For our part, we're leading our own research into onboarding and wallets, and we love to see protocols openly question parts of the crypto user experience that have begun to crystallize around user-unfriendly patterns.

The Cosmos's Interchain Security Timeline Comes into Focus ⚛️

In our last update, we wrote about the Cosmos's forthcoming Interchain Security (ICS) feature:

ICS solves [the challenge for new chains of spinning up their own validator sets] in a clever way by simply allowing new chains to run using the validator set of an existing chain. In the official ICS vocabulary, the new network becomes known as a "consumer chain," because it's "consuming" (or, "leasing") security from the existing network, which in turn becomes known as the "provider chain" because it's "providing" security to the new network.

Cosmos commentator "Robb Stack" sums up the upcoming timeline of events for ICS's launch nicely:

And researcher "Thyborg" breaks down next steps for the Hub:

The first major step to watch will be the passage the next major upgrade for the Cosmos Hub, the "Rho" upgrade. (Chain updates often have fancy code names like this). This upgrade hasn't been put up for a formal vote yet, but the details are being hashed out on the Cosmos Forum now.

Nakaflow Maintenance Update

Nakaflow.io is our tool for comparing the Nakamoto Coefficient for stake distribution on more than a dozen of the leading Proof-of-Stake blockchains (like Ethereum, Solana, the Cosmos Hub, and more). The Nakaflow site is currently down as we make some upgrades to the backend and user interface.

We'll share an update on Twitter when the tool returns. We appreciate your patience in the meantime.