- Author: @ferebee, @jmfayal, @rawrmaan, @gateholder
- Start Date: 2023-03-29
- Category: Economic, Technical
- Original HIP PR:
- Tracking Issue:
HIP-51 specifies the DAO Utility Score, which determines the distribution of HNT among all Helium subDAOs.
Its purpose is to incentivize current and future subDAOs to grow the value of their services provided through the Helium DAO, while also protecting the investment of the LoRa Hotspots that participated from the start to launch the Helium network of networks.
As the implementation of HIP-70 has progressed, multiple theoretical and practical drawbacks of the DAO Utility Score as specified in HIP-51 have become apparent, threatening the interests of the existing IOT and MOBILE subDAOs, possible future subDAOs, and the Helium DAO as a whole.
To solve these problems, HIP-80 proposes a new, simplified DAO Utility Score, which is easier to understand and protects the interests of all participants.
The new Score no longer considers the number of Hotspots and their onboarding fees, instead using only the square root of the product of DC Burn and delegated veHNT.
The IOT subDAO receives an explicit guarding factor to ensure its continued funding.
As the Hotspot onboarding fee is no longer relevant to the new Score, subDAOs may find it advantageous to lower their onboarding fee, and may do so by subDAO governance. The minimum onboarding fee is $5 and may be changed by Helium DAO governance. A Hotspot must burn an onboarding fee towards each subDAO in which it participates.
The existing definition of DAO Utility Score as specified in HIP-51 is ambiguous and difficult to understand. Simplifying it will make it more approachable and reduce the probability of unintended consequences.
In addition, certain intents of HIP-51 are not fully realized by the existing Score definition.
On the one hand, due to an oversight in the implementation of HIP-53, no onboarding fees have ever been burned towards the MOBILE subDAO. Therefore, at launch, the DAO Utility Score of the MOBILE subDAO as defined in HIP-51 would be eclipsed by the IOT subDAO, so that much less HNT than intended would initially be distributed to the MOBILE subDAO. This could limit the growth of MOBILE, which is important to the further development of the Helium Flywheel, and thus to IOT as well.
On the other hand, calculations show that the IOT subDAO risks being marginalized by the MOBILE subDAO in the medium term under the existing Score, if current trends continue. This could cause nearly all HNT to be distributed to the subDAO treasury of the MOBILE subDAO, which would run counter to the intent of HIP-51 and could limit the further development of IOT. This threatens the health of the Helium DAO as a whole, as IOT is expected to deliver a significant contribution to the Helium DAO over the longer term.
With its V factor, the existing DAO Utility Score gives far more weight to the amount of veHNT delegated to a subDAO than to any other factor. This would enable a single large investor to favor a single subDAO disproportionately with a large veHNT delegation, and would prioritize economic incentives for delegation while marginalizing the importance of network growth and utility.
Finally, if a new DAO Utility Score can be defined that does not rely on the number of Hotspots in a subDAO and their onboarding fees, the onboarding fee itself loses most of its importance. New types of low-cost Hotspots have been proposed for both IOT and MOBILE. The traditional onboarding fee of $40 would make up a large portion of the retail cost of these Hotspots. Reducing the onboarding fee can significantly reduce their cost and encourage the adoption of these new devices, contributing to the success of both networks.
This proposal is relevant to all participants in the Helium DAO.
- Existing and future IOT and MOBILE Hotspot owners are affected by the reduction in fees and/or the changes to DAO Utility Scores.
- The cost to Manufacturers for onboarding fees will be reduced.
- New subDAOs will receive a different share of HNT emissions.
HIP-51 introduces the DAO Utility Score, which determines the proportion of HNT emissions distributed to each subDAO, and defines it as:
where
HIP-53 describes a $40 onboarding fee and $10 location assertion fee, analogous to the traditional fees assessed for IOT Hotspots. However, the gateways developed by FreedomFi and used to launch the MOBILE network, as well as models introduced by other manufacturers, include both IOT and MOBILE Hotspot functions, which operate independently. In the existing implementation, the IOT Hotspot component is onboarded in the same way as all other Helium IOT Hotspots, with a $40 onboarding fee and a $10 location assert. However, the MOBILE Hotspot component itself currently performs no onboarding operation of its own, but it is treated as if it were onboarded nonetheless.
As a result, the A component of the Score is currently 1 for the MOBILE subDAO, as no onboarding fees (a. k. a. Activation Fees) have yet been burned to the MOBILE subDAO.
Due to this implementation oversight, the DAO Utility Score of MOBILE as defined by HIP-51 is slated to launch far below its intended value, which would impair the incentive for participants to deploy new MOBILE Hotspots for Helium 5G. The buildout of Helium 5G is essential for the upcoming launch of Helium Mobile and therefore the growth of the Helium DAO as a whole.
Furthermore, no firm launch date has yet been set for the Helium Mobile MVNO, the first Service Provider of the MOBILE subDAO. As a result, the subDAO is unable to burn DC for Data Transfer, even though hardware has already been successfully deployed. This further limits the Score of the MOBILE subDAO at launch.
On the other hand, if the MOBILE subDAO is successful, it is expected to burn a large volume of DC per Hotspot in the medium term, as it is providing network coverage for a device class (smartphones) that is already ubiquitous and consumes large amounts of bandwidth. The IOT subDAO is expected to follow a slower growth path, as it is providing coverage for entirely new applications that are still being developed and are expected to be rolled out in the next several years as technology matures.
Therefore, under the existing model, the DAO Utility Score of the IOT subDAO could be eclipsed by MOBILE in the medium term, such that almost no HNT is distributed to the IOT treasury. This would endanger the health of the IOT network and prevent it from achieving and maintaining the necessary coverage until the demand for the network reaches its potential.
Independent of the installed Hotspots and the Data Transfer and DC burn they contribute, the V component of the DAO Utility Score as defined in HIP-51 considers the veHNT delegated to the various subDAOs. This is intended as a way for network participants to signal their preference for a particular subDAO. By encouraging participants to lock their HNT for veHNT, it also contributes to the stability of the Helium DAO and its governance.
While the A component, which counts Hotspots and their onboarding fees, is counted by its fourth root, and the D component, which counts Data Transfer, is counted by its square root, the V component is counted linearly.
This gives inordinate weight to the delegation of veHNT. For example, 10 times the amount of delegated veHNT would have the same weight as 100 times the DC burn from Data Transfer, or 10,000 times the number of Hotspots.
Modeling has shown that this favors a winner-takes-all scenario. Once one subDAO has received a large majority of delegated veHNT, the V component of the DAO Utility Score dominates the Score to such an extent that it becomes economically unattractive to delegate to any other subDAO, considering the delegation rewards that are available in the different subDAOs, as valued through the treasury exchange rate to HNT. Another subDAO, even if it provides network coverage and DC burn that are valuable to the Helium DAO as a whole, could be unable to attract sufficient veHNT delegation to reward its participants appropriately.
In the Helium network as originally designed, prior to HIP-25, blockchain consensus ran on the LoRa Hotspots themselves. This was a main reason for the $40 Hotspot onboarding fee, which was designed to provide resistance against Sybil attacks, in which illegitimate Hotspots could endanger the security of the blockchain itself. Ever since HIP-25, blockchain security has been provided by Helium validators, and with the implementation of HIP-70, blockchain security will be provided by Solana. Therefore, onboarding fees are no longer needed for Sybil resistance.
Meanwhile, given recent advances in hardware, the traditional fee of $40 per Hotspot has become a significant component of total Hotspot deployment cost. Manufacturers are ready to offer LoRa Hotspots for the IOT subDAO below $100, and low-cost short-range Wi-Fi Hotspots have been proposed for the MOBILE subDAO, complementing longer-range LTE and 5G deployments, which could potentially also reach prices below $100.
For these devices, the traditional $40 onboarding fee is a significant barrier to wide-scale deployment, and hampers the growth of both IOT and MOBILE networks and thereby the success of the Helium DAO as a whole.
The remaining justification for an onboarding fee is to prevent nuisance attacks, such as a bad actor spamming the network with large numbers of invalid Hotspots, which could happen if a Maker key were compromised. For that, a nominal fee would be sufficient.
In response the issues outlined above, HIP-80 proposes a new definition of the DAO Utility Score, replacing the definition from HIP-51 with the following:
where
and Floor is 7 for all subDAOs except for the IOT subDAO.
Floor shall be 40 for the IOT subDAO until the fourth halvening of HNT emissions, which is currently scheduled to occur on 1 August 2027. Then, it shall revert to 7.
HIP-51 recognizes that the participants of the IOT subDAO jumpstarted the Helium DAO with a large investment in hardware despite challenges in the supply chain, and have decided to invite the MOBILE subDAO and future subDAOs into the Helium DAO while anticipating that they would be able to share in the joint success of all future DAO networks.
The A factor of the DAO Utility Score as proposed in HIP-51, which counts the number of active Hotspots multiplied by their onboarding fees, was intended to respect this contribution by giving an initial advantage to the IOT subDAO.
To respect this intent while simplifying the calculation, HIP-80 proposes to give the IOT subDAO an explicit—but limited—advantage by setting its Floor to 40.
The effect of this is to treat the IOT subDAO as if
This special treatment of IOT shall last until the fourth halvening of HNT emissions, which is scheduled for 1 August 2027, when Floor for IOT shall revert to 7.
In other words, as soon as the IOT subDAO achieves a monthly DC burn of $48,000, its founder’s bonus will become irrelevant. If it cannot achieve this burn by 1 August 2027, it will have to succeed or fail on its own merits.
As was proposed in the discussions leading up to HIP-51, this will ensure that the IOT subDAO has the opportunity to receive a majority share of total HNT emissions as long as other subDAOs are not contributing significant DC burn.
Similarly to the calculations above, the effect of the Floor of 7 for all other subDAOs besides IOT is to treat all subDAOs as if they were to burn at least $1,470 of DC per month. This ensures that the MOBILE treasury will get some share of HNT emissions from launch, and provides for a minimum level of HNT emissions (subject to the V factor) to any future subDAOs which may be approved to join the Helium DAO.
On the other hand, once Helium Mobile and possibly other Service Providers begin moving data over Helium 5G, the MOBILE subDAO may quickly begin burning significantly more DC than the notional $48,000/month floor attributed to the IOT subDAO, so the MOBILE subDAO may then be able to capture a larger share of total HNT emissions, while also contributing the greater value to the Helium DAO and thus to HNT itself by virtue of its greater rate of DC burn. This in turn will benefit the IOT subDAO as well.
As discussed above, HIP-80 eliminates the A factor of the DAO Utility Score entirely, replacing the implicit advantage at launch it would grant the IOT subDAO with an explicit guaranteed minimum DC burn factor.
By eliminating the dependence on the number of active Hotspots and the onboarding fees charged by subDAOs, the new Utility Score introduced by HIP-80 greatly simplifies calculations that would otherwise need to be performed by Oracles and/or smart contracts of the Helium DAO.
Furthermore, it eliminates the incentive for subDAOs to artificially inflate their Hotspot count or their onboarding fees.
The Score proposed in HIP-51 gives such a high weight to veHNT delegation, which is considered as a linear factor, that an unequal distribution of veHNT between subDAOs could fully dominate all other components of the Score. Thus, either the IOT subDAO (despite its network size) or the MOBILE subDAO (despite its expected DC burn) could be completely marginalized by a single large delegation of veHNT. This could allow economic interests of investors to override all other contributions of DAO participants in deciding the distribution of HNT.
Therefore, the new DAO Utility Score proposed by HIP-80 considers veHNT delegation and DC burn equally, by multiplying together the square root of each.
The Utility Score proposed in HIP-80 eliminates the incentive for subDAOs to compete for HNT emissions through the onboarding fees charged to their Hotspots, as fees are no longer relevant to the Score and thus the share of HNT emissions to each subDAO.
The sole remaining purpose of onboarding fees is to protect the Helium DAO from nuisance attacks caused by onboarding an arbitrary number of invalid hotspots.
Therefore, the Helium DAO, with HIP-80, sets a minimum onboarding fee for all subDAOs of $5, to be modified in the future as needed through Helium DAO governance. A Hotspot participating in multiple subDAOs shall burn a separate fee to each subDAO independently. Each subDAO may set its onboarding fee as desired by subDAO governance, subject to the minimum.
As a one-time exception, all MOBILE Hotspots onboarded prior to the implementation of HIP-70 through the legacy onboarding procedure, which was introduced in the implementation of HIP-53, shall be considered to have been onboarded to both the MOBILE and the IOT subDAOs.
This proposal may require some DAO participants to revisit their evaluations of veHNT delegation strategies.
- The simplified DAO Utility Score introduced by this proposal is easier to evaluate and model. By giving equal weight to economic investment and network deployment and usage, it is anticipated to deliver stabler economic incentives for all participants.
- The existing MOBILE Hotspots can be properly onboarded without the need to secure additional funding for the missing onboarding fees.
- The MOBILE subDAO is guaranteed to receive a significant share of HNT emissions from launch, consistent with the intent of HIP-51.
- The IOT subDAO is guaranteed a predictable minimum incentive over the longer term to continue network buildout and maintenance, protecting its opportunity to realize its potential.
The calculation of Utility Score will need to be adjusted. Based on feedback from the HIP-70 implementation team, this is possible within the existing development framework. It would be desirable to finalize all changes to the Utility Score before the Solana transition on 18 April 2023.
- subDAOs may continue to assess location assertion fees, which are determined by subDAO governance.
- The Helium DAO HNT emissions contract distributes HNT to HST holders as specified in HIP-20.
- In clarification of HIP-51, all remaining HNT is distributed between all subDAOs in proportion to their relative DAO Utility Scores.
- If the fourth halvening of HNT emissions falls on a different date than 1 August 2027 for any reason, the founder’s bonus IOT Floor of 40 shall remain in effect until then and end on that date.
- This HIP is succesful if simplifying the DAO Utility Score supports the growth and utility of all subDAOs.