What is DePIN (Distributed Physical Infrastructure Network)? What is the key to building a sustainable DePIN? | CoinDesk JAPAN

Some of the most progressive players in the crypto industry seem to believe that Decentralized Physical Infrastructure Network (DePIN) will be the first use case to bring crypto into the mainstream. They predict huge growth similar to NFTs and stablecoins.

The reason is that DePIN is capable of amazing feats. By leveraging crypto incentives and aligning millions of participants, the DePIN project is able to develop new products that were not possible before.

In December 2023, Helium Mobile will roll out a $20-a-month unlimited cell phone plan across the U.S. using community-owned 5G hotspots. Render Network built a robust permissionless machine learning cloud in less than six months amid a global GPU shortage.

And my company, Hivemapper, mapped 10% of the world's roads in less than a year. These were not possible before DePIN appeared.

DePIN is an exciting future, but we must learn from the pioneers in this field to ensure it is not just a passing fad.

The original pioneers of the DePIN project experimented with a ton of different models, made a lot of mistakes, and learned from them. However, the learning seems to have faded into memory. Without learning from the past, the new DePIN risks creating a boom-bust cycle that will leave a stain on the entire sector.

To help DePIN chart a more sustainable path, we would like to share some of the most important lessons we have learned from our services. More broadly, for those looking to join DePIN, these lessons are also important to understand when assessing the long-term success of the DePIN project.

The static reward trap

Token rewards will be gamified to encourage contributors to collectively develop products at scale. On the other hand, tokens are consumed by using the product.

For the DePIN project to be sustainable, there needs to be a sound strategy to generate supply and, perhaps more importantly, real demand for the product.

Without that, demand for tokens would simply become speculative, more like meme coins than useful infrastructure projects that could impact the lives of billions of people.

Creating supply is easy if you have the right tools. All you have to do is provide tokens for your labor. The difficult part is figuring out how much to offer. There should be not so many that they disappear in the later stages of growth, but not so small that no one will contribute.

New projects also need to pay special attention to the “cold start” problem. New networks cannot provide much value with one contributor, or even 100 contributors. Before we can think about meeting demand, we first need to reach a certain threshold.

Creating real demand is more difficult. Product-market fit must be found quickly. In our experience, it helps to focus on very real pain points that a decentralized approach can uniquely solve.

It's tempting to take shortcuts, and the most common shortcut we've seen is to employ static rather than dynamic rewards. We call this the “static reward trap.”

It's easier to attract contributors in the beginning if you allow them to get paid just for being involved in the project, regardless of how much value they add. However, static rewards can be fatal in the long run because they break the fundamental incentive structure of the network.

What if helium gave all 5G hotspots the same static reward, whether they're places where people live and work like Manhattan or sparsely populated places like Death Valley? Deaf?

There is no incentive to increase coverage where people need it most, hindering the mission to democratize access to communications and defeat centralized big carriers.

What if the hive mapper issued the same static reward to all dash cams? It would be a disaster. People can get paid to install dashcams in any car, even cars that are rarely driven or parked.

It is possible to install 5 or 10 drive recorders in the same car and capture the same map data over and over again in order to earn additional rewards.

If the DePIN project treated all contributors equally, i.e. rewarded everywhere, it would not be able to achieve critical mass everywhere. Productivity needs to be a protocol imperative. They must establish base rewards for non-useful contributions, insist on quality, and reject duplicative work and low-quality data.

Aligning incentives using utility tokens is a powerful power of DePIN. However, if a project falls into the trap of static rewards, contributors have little incentive to make the product good in the long run. Don't trade lasting power for short-term growth.

Dynamic reward design

In most DePIN projects, dynamic rewards have four aspects:

• Geography (infrastructure is more valuable in some places than others)
• Contributor productivity (more productive contributors deserve more compensation)
• Contributor quality (high-quality contributions deserve more rewards)
• Network advancement (a more useful network should generate more rewards overall)

We didn't get all of this right when designing the Hivemapper network. The Hivemapper Foundation was the first to admit its mistakes, including issuing excessive bounties to incentivize mapping certain parts of the world. However, we are making steady progress and fostering one of the most productive communities at DePIN.

To put it into practice, let me show you how we used this framework in our network.

geography

We designed Hive Mapper to ensure that customer demand in a given region leads to better rewards. 100% of the tokens burned through the use of the map will be reminted to the contributors who provided the data.

We made similar changes to minted rewards, adding reward multipliers for regions that collect more useful map data, even before customer usage increases.

In an excellent essay on the design of the DePIN network, investment firm Multicoin Capital calls this a “land and expand” approach.

Contributor productivity

As a Hivemapper, your compensation as a contributor is determined by how much you cover and the uniqueness (or “freshness”) of your tracks. It would be easier to bring in new contributors otherwise. Some people don't drive much. Some people commute along almost the same route. Such people don't get as much reward from hive mappers, but they should.

That's why Hivemapper is focusing on hiring gig drivers, truck drivers, and commercial truckers. These drivers are highly productive, get the job done while passively collecting map data, and earn the highest pay.

quality of contributors

Creating maps requires high-quality data with accurate location information. For this reason, Hive Mapper sets evaluation scores for each type of contributor. The higher the contributor's rating, the better the reward they receive. If this system works well, it can align incentives correctly.

Hive Mapper contributors must correctly install dash cams to get the best image quality and best GPS signal for precise location mapping. They cater to the needs of their network in order to maximize their own rewards.

Network advances

The size of Hivemapper's weekly reward pool is based on a dynamic metric called map progress, which measures how much mapping is done across the network.

Many projects have a fixed mint schedule, which results in early adopters being offered large rewards upfront.

It's true that early adopters deserve bigger rewards because they take more risks and deal with more technical problems.

However, a fixed minting schedule results in rewards that continue to decline as more contributors come online and the network becomes more useful. It can also create a toxic relationship between early adopters and later contributors, creating unhealthy confusion.

We are open sourcing this framework because it has helped us build one of the fastest growing communities at DePIN.

A big reason why we are able to do more with fewer people is because we build on the achievements of our predecessors, and I hope to be able to repay that kindness.

Incentive design is not easy. But if DePIN developers stay focused and stakeholders hold them accountable, we will live up to DePIN's potential and realize that much of humanity's critical infrastructure is owned by the people who use it. We can build a world in which we operate.

｜Translation and editing: Akiko Yamaguchi, Takayuki Masuda
｜Image: Greg Rosenke/Unsplash (processed by CoinDesk)
｜Original text: The Key to Building Sustainable DePIN