Original author: Mason Nystrom, investment partner of iant
Editor: Zen, PANews
Decentralized infrastructure networks, especially crypto networks that leverage token incentives to generate liquidity to fund physical infrastructure operations, are proliferating rapidly. The value of these networks is clear: they enable better solutions for consuming resources, from computation to energy to data. Otherwise, these resources are consumed directly by the company or, more commonly, are used for the company’s own products and services.
For example, decentralized networks such as Web3 mapping company Hivemapper sell data directly to transportation companies such as Uber, which in turn use imagery data to improve their products. Likewise, the decentralized streaming delivery protocol Livepeer has enabled live streaming apps to enter the market for its video transcoding service, but has also been integrated by companies like Bonfire, which makes it easy for creators to launch their own live streams.
To better assess the potential of these networks, we need a better way to classify them. The current popular industry name is DePIN, but iant proposes to introduce a further division into two categories of decentralized infrastructure networks:
Decentralized Physical Infrastructure Network (DePIN): An encrypted network with consumable and non-substitutable resources, using incentives to deploy location-dependent hardware devices
Decentralized Resource Network (DeREN): A cryptographic network that leverages incentives to build markets and increase the supply of existing or idle consumable, fungible resources that rely on location-agnostic hardware
DePIN differs from DeREN in three core dimensions:
Resource substitutability
Hardware location deployment
Resource creation
Resource Substitutability
The most meaningful of the above differences is the fungibility of expendable resources.
In a resource network, consumable resources are interchangeable because the network’s hardware assets are generally interchangeable. For example, computing resources provided by networks like Akash or Render are highly fungible, with one GPU having the same processing power as any other GPU of the same specification and capacity. Outside of highly specialized activities such as high-frequency trading, users generally don’t care where their hardware is deployed geographically, as long as network latency is acceptable compared to centralized architectures.
Instead, DePIN utilizes non-fungible or semi-fungible resources. In this case, expendable assets are not easily interchangeable, and the hardware is often unique to a particular network. For example, Hivemapper’s dash cam can map a specific location, generating data unique to that location and place in time. Also, an imaging network like Spexigon cannot contribute its aerial imagery data to a Hivemapper network; each network’s asset is the image map data, and it is unique to that network.
Of course, there are also assets in the middle. For example, energy is semi-fungible because it can be used for many purposes, but its utility is limited by the distance it can be transmitted.
Hardware location and resource creation
Hardware location and resource creation are closely related; the deployment of application-specific and location-dependent hardware often occurs concurrently with building proprietary resources.
At this point, DePIN faces more challenges in structuring the market supply and demand sides. The supply side requires a location-dependent hardware setup, and demand generation depends on the supply side having sufficient scale to make the network valuable to consumers.
Resource networks are easier to bootstrap supply because idle supply can come from anywhere, and often there is no need to create new hardware and infrastructure. But resource networks with fungible assets also face greater competition because of the lower cost of switching from one resource network to another.
Build moats in DePIN and DeREN
Encryption-based resource networks still have to compete with web2 peers like AWS and Google. While DePIN and DeREN can utilize tokens to subsidize initial resource costs, the most successful networks will not compete solely on price, but will unlock new demand or expand markets in unique ways.
For example, Arweave doesn’t compete on file storage pricing. It offers new functionality and convenience through permanent storage, and has finally found appeal in storing NFT metadata. In the DePIN category, mobile networks like DIMO aggregate previously siled data to power a new wave of applications, from battery intelligence and energy management to better vehicle commerce.
Another successful strategy is to vertically integrate and generate demand by building an initial product that leverages an infrastructure or resource network. Render combines its GPU rendering capabilities with its Octane software, driving the use of the underlying network of computing resources.
Related reading: The whole picture of DePIN track: disruptive innovation or “castle in the air”?
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
New Proposals for Taxonomy of Decentralized Infrastructure Networks: DePIN and DeREN
Original author: Mason Nystrom, investment partner of iant
Editor: Zen, PANews
Decentralized infrastructure networks, especially crypto networks that leverage token incentives to generate liquidity to fund physical infrastructure operations, are proliferating rapidly. The value of these networks is clear: they enable better solutions for consuming resources, from computation to energy to data. Otherwise, these resources are consumed directly by the company or, more commonly, are used for the company’s own products and services.
For example, decentralized networks such as Web3 mapping company Hivemapper sell data directly to transportation companies such as Uber, which in turn use imagery data to improve their products. Likewise, the decentralized streaming delivery protocol Livepeer has enabled live streaming apps to enter the market for its video transcoding service, but has also been integrated by companies like Bonfire, which makes it easy for creators to launch their own live streams.
To better assess the potential of these networks, we need a better way to classify them. The current popular industry name is DePIN, but iant proposes to introduce a further division into two categories of decentralized infrastructure networks:
DePIN differs from DeREN in three core dimensions:
Resource Substitutability
The most meaningful of the above differences is the fungibility of expendable resources.
In a resource network, consumable resources are interchangeable because the network’s hardware assets are generally interchangeable. For example, computing resources provided by networks like Akash or Render are highly fungible, with one GPU having the same processing power as any other GPU of the same specification and capacity. Outside of highly specialized activities such as high-frequency trading, users generally don’t care where their hardware is deployed geographically, as long as network latency is acceptable compared to centralized architectures.
Instead, DePIN utilizes non-fungible or semi-fungible resources. In this case, expendable assets are not easily interchangeable, and the hardware is often unique to a particular network. For example, Hivemapper’s dash cam can map a specific location, generating data unique to that location and place in time. Also, an imaging network like Spexigon cannot contribute its aerial imagery data to a Hivemapper network; each network’s asset is the image map data, and it is unique to that network.
Of course, there are also assets in the middle. For example, energy is semi-fungible because it can be used for many purposes, but its utility is limited by the distance it can be transmitted.
Hardware location and resource creation
Hardware location and resource creation are closely related; the deployment of application-specific and location-dependent hardware often occurs concurrently with building proprietary resources.
At this point, DePIN faces more challenges in structuring the market supply and demand sides. The supply side requires a location-dependent hardware setup, and demand generation depends on the supply side having sufficient scale to make the network valuable to consumers.
Resource networks are easier to bootstrap supply because idle supply can come from anywhere, and often there is no need to create new hardware and infrastructure. But resource networks with fungible assets also face greater competition because of the lower cost of switching from one resource network to another.
Build moats in DePIN and DeREN
Encryption-based resource networks still have to compete with web2 peers like AWS and Google. While DePIN and DeREN can utilize tokens to subsidize initial resource costs, the most successful networks will not compete solely on price, but will unlock new demand or expand markets in unique ways.
For example, Arweave doesn’t compete on file storage pricing. It offers new functionality and convenience through permanent storage, and has finally found appeal in storing NFT metadata. In the DePIN category, mobile networks like DIMO aggregate previously siled data to power a new wave of applications, from battery intelligence and energy management to better vehicle commerce.
Another successful strategy is to vertically integrate and generate demand by building an initial product that leverages an infrastructure or resource network. Render combines its GPU rendering capabilities with its Octane software, driving the use of the underlying network of computing resources.
Related reading: The whole picture of DePIN track: disruptive innovation or “castle in the air”?