January 1, 2022
With Bitcoin as “proof of decentralization”, Web3, a jungle of experimental networks, has sprouted. Web3 networks are mediated by ledgers not maintained centrally but by the nodes themselves. Accordingly, the value created by Web3 networks accrues not to a gated siren server but to the network itself via tokens. At risk of sounding pollyanna, it is those who make the network valuable that are rightly rewarded for those contributions.
Within Web3, a hardware-focused niche has emerged. Not only do such networks seek to disrupt centralized competitors but also to launch infrastructure that was previously impossible to build. These are not decentralized moneys, but decentralized services networks. These are hives.
The Networked Industrial Age: Introducing the Hives of Web3
Hives are an evolution of the gig economy: they seek to mobilize slack resources and direct them to paying consumers. Similar to Uber and AirBnB, the underlying software contains the instructions necessary to rearrange existing resources into a more meaningful, efficient form. But dissimilarly, it is within a distributed, as opposed to a centralized database, that resource suppliers and resource consumers convene to exchange value.
“These digital public goods are operated through the use of Minimally Extractive Coordinator (MEC) protocols — self-running systems of logic that connect buyers and sellers of a particular asset or service together, with the goal of allowing those buyer/sellers to retain as much value as possible during their transaction by minimizing excessive rent extraction. In many ways, MECs are similar to companies like Amazon and Uber, except the company is replaced with a decentralized computation network that automatically matches supply with demand based on preset parameters that all parties can verify, yet no one can tamper with.” ChainlinkGod Medium
The quality of a supplier’s work can be checked by the network. If a supplier node delivers, it will receive in-kind token payments from the network . If a supplier node fails to deliver, it will be punished via the slashing of its token holdings. As in any token-based Web3 network, the token wears multiple hats. In hive services, the token represents ownership, a medium of exchange, and a demonstration of skin in the game. When all participants share skin, they are incentivized to move in the same direction over the long-run. That a network has such a versatile monetary unit is completely novel—well-designed incentives drive good behavior.
To me, the clearest example of a thriving hive service is Arweave. Arweave is a decentralized storage platform. Arweave’s pitch is perma-storage. This is different from the contract-based model of its centralized competitors, AWS and Azure, or decentralized competitors, Storj, Sia, and Filecoin. (See a Vulpine overview of the decentralized storage industry here). The Arweave hive has begun to make honey: the network has generated ~$500K in fees over the past 90 days.
Source: Web3 Index
But Arweave is merely the tip of the decentralized gig economy:
Chainlink, Band Protocol, and API3 are decentralized oracle networks, which connect off-chain data providers with on-chain decentralized applications. (See a Vulpine overview of the oracle space here)
Render and Golem protocols connect nodes with spare GPU compute to creators in need of graphical rendering.
Livepeer likewise sources GPU compute but towards a different end—the transcoding of video for consumption across devices.
The Helium and NKN protocols unlock idle bandwidth, the former for usage by IoT sensors and the latter for broader consumption.
Akash is the self-proclaimed “AirBnB for cloud compute”.
Aragon Court connects an ad-hoc, anonymous, and fully digital jury to disputes in need of timely resolution.
Most of these protocols operate on the tacit assumption that their decentralized versions can and will outcompete the existing, centralized alternatives.
Which Hives will Survive the Darwinian Machine?
In my view, contemporary Web3 hive services fall along a continuum of feasibility. Depending on what, if anything, decentralization and tokens add to their core value proposition, hive services will either thrive or wither.
Blue Sky: When Hives Unlock a New Market
Arweave’s product is effectively the same as Bitcoin’s: permanence. Permanent storage is a product the market has yet been unable to deliver. The protocol has healthy demand and no real competitors to speak of—a blue sky. Data-dense public ledgers like Solana need an affordable and immutable way to store history. NFT owners want a secure place to store their assets. The appeal for political dissidents to immutably archive their contributions is obvious. The Arweave team recognized that a centralized version of the network was not only a downgrade, but untenable. A single point of failure undermines the fundamental strengths that make the core product valuable to users. Like with BTC, the decentralized architecture unlocks permanence, a product that heretofore has been impossible to provide.
Helium provides a mesh network for IoT devices and seeks to provide 5G soon. The IoT revolution has been “a few years out” for ten years. Equipping devices with sensors makes sense. But without a network to transmit data through, an IoT device is inert. Large telecoms are perfectly capable of constructing the connectivity infrastructure. But without demand from IoT devices, the incentives to do so are low. And without connectivity, the incentives to equip devices with sensors is low. A paralyzing catch-22.
Solving the chicken-and-egg deadlock, the protocol bootstraps the supply side of the network (connectivity) through token issuance. Suppliers who demonstrate Proof of Coverage (PoC) are rewarded with issuance of the native token, $HNT. These proofs are conducted through radio waves transmitted by proximate nodes. Quality of coverage and location are validated and the proof is posted to the Helium blockchain. As the number of nodes supplying connectivity increases, token rewards decrease. The premise is that at some point nodes will be remunerated through data credits denominated in $HNT, or actual demand from IoT customers (as opposed to protocol rewards). Early adopters are rewarded for the risk they incur for accepting payment for services in a currency that is non-established (network’s success is unlikely).
The Arweave and Helium models are product-first, crypto-second. For Arweave, permanent storage is supported by redundancy of storage and token incentive design which are supported by decentralization. For Helium, dense network coverage is supported by incentives and quality-of-service proofs which are supported by tokens and decentralization, respectively. In both cases, by building the products on crypto rails, a completely new product and (hopefully) market are unlocked.
Red Sky: When Hives Compete with Centralization
In the decentralized computing realm, you have Livepeer, Render, Golem, Ankr, Flux, iExec, and Akash. Each of these networks is predicated upon the fact that most processors/computers in the world are not being fully utilized. An idle or under-worked GPU or CPU can contribute to a networked cloud and get compensated for any compute consumed by users of the network. Payment is delivered in the network’s native tokens. Similarly, nodes typically access the right to supply compute to the network by bonding a certain number of native tokens.
NKN is a peer-to-peer bandwidth protocol. It rewards nodes for providing internet connectivity to the network of paying users. With enough nodes providing connectivity, NKN claims it will be able to provide faster routing than via traditional alternatives.
BitTorrent is a peer-to-peer file transfer protocol. The more nodes transmitting files on the network the harder it is for a party to block or censor information transfer.
These aforementioned hives compete in far more crowded airspace: cloud computing, internet bandwidth, and file transfer are commoditized and competitive. Incumbents like Amazon, Microsoft, and Google benefit from economies of scale and decades of knowhow, development, and infrastructure. Is there any reason to think that decentralized versions of existing services will succeed? What needs to happen for this to be the case?
The Next Five Years for Hives
“To Satoshi, decentralization was valuable insofar as it mitigated some other fundamental risk: censorship, platform security, corruption etc. It’s the properties decentralization gives us that we care about, not decentralization itself.” - Haseeb Qureshi, Dragonfly Capital
We’ve seen plenty of examples of dependency risk inside and outside crypto. AWS worked fine for dydx, a derivatives exchange, until an east coast outage caused it to go down. Twitter and Youtube worked well for Trump until he was de-platformed. Meme stock traders reaped the benefits of low cost trading on Robinhood until trades for certain tickers were blocked.
As single-point-of-failure events continue to happen, companies and protocols will increasingly look for something with redundancy, or modularity. If a viable and mission-ready decentralized alternative presents itself (e.g. Akash), principals will certainly consider the option to hedge against uncommon, but impactful events. Hives that provide anti-fragile infrastructure may continue to accrue customers as censorship (witting and unwitting) and platform risk continue undermining business continuity. In my view, adoption will surge with each crisis of centralization and then simmer as people forget. But over time, more and more applications will realize that to be truly insulated, their entire stack needs to be decentralized. (This full-stack decentralization will be particularly important in countries where the government controls corporations)
More prosaically, hives may also provide gains in efficiency. Companies and protocols will continue to outsource non-core activities to specialists. And these groups want the best service at the cheapest price. To the extent that hives provide a more dynamic on-demand service that is efficiently priced, they may accrue customers who are sensitive to wasteful pricing schemes. Efficiency also benefits the supply sides of hives: nodes supplying resources to the network would’ve otherwise generated no revenue on their assets. For example, as Ethereum moves to Proof-of-Stake, the GPUs/CPUs that previously secured the network (and were rewarded for doing so) become non-productive. Such dynamics will drive significant supply to hives, driving competition and pushing costs for end-users down. By embracing a networked market where many players can compete, protocols tend towards being minimally extractive.
As many others have opined, I believe 2022 is the year of the DAO. DAOs are grassroots communities who self-organize around a common mission and treasury. Usually these communities live on (centralized) messaging services such as Discord and seek to eventually deliver a product or service. (See Index Co-Op as an example). DAOs are not recognized as legitimate entities outside of Wyoming. These communities thus cannot execute service agreements with infrastructural service providers . For political and regulatory reasons, such communities seek to remain sufficiently decentralized. If an individual were to sign service agreements and execute the payment for such services, such an event would serve as obvious evidence of centralization of power. So many DAOs will opt for hives to provide permissionless infrastructure on demand. Instead of relying on customers to switch from existing infrastructural providers, the boom in DAO formation would serve as an organic demand driver for hive services.
A Hive’s-Eye View
Hives are a productive experiment in combining infrastructure hardware, the software that rearranges that hardware, and tokens that incentivize the whole system to move in the right direction. Many iterations will not find product-market fit. They’ll be too early, too expensive, or unnecessarily decentralized. But this grand experiment will bear completely new business models. That a network can reward an anonymous constellation of suppliers with internet-native equity for quality contributions is transformative. With the genesis of the hives comes a new frontier of the networked economy.
Disclosures: Lupine Capital I, LP/TJ Ragsdale may hold some of the tokens mentioned in this piece