In order to keep the application threads clear of discussions (to make it easier for committee members to read and score them), please use this thread for any and all questions and discussions of September 1-9 2023 period of grant, bounty, and retrospective award applications.
Hello, happy to answer any questions folks might have about the OpenUX submitted grant application September 1-9 2023 GMC Call for Grant Applications - Deadline is September 9 - #2 by GeorgiaOpenUX
I’d like to support @calurduran’s RA. I would ask that the GMC consider paying more than requested.
Calur was the motive force that got the GMC started when it did - it was on my list (and I’m sure other lists), but not at the top. Calur was the only person that was willing to step up to herd cats hard (I specifically asked GMC Membership & pDAO Treasurer Selection - #2 by Valdorff). I believe “I volunteered without any expectation of payment” is impressive, and not a good reason to slice pay. Calur worked on a bunch of process that was used later on. Calur helped transition after it made sense for him to step down for non-RP reasons.
I’d suggest $16,875 matching the GMC admin position.
I wanted to note a couple of things. First, I appreciate the vote of confidence. I think a lot of us struggle with feeling like self-promoters when trying to price RAs/grants (but especially RAs). I went with the number I’d be very happy receiving, but I’d be even happier to receive more if the GMC decides I was underselling myself.
The other thing I wanted to note, just because I am aware of the optics around the timing here. I just nominated Val for an RA for the oDAO/pDAO Charter work. That RA has been in the works for several days, prior to my knowing Val was going to post his supportive comments, and my decision to write one was not influenced in any way by Val’s comment. Similarly, I am very confident Val’s decision to write that supportive comment was not influenced by knowing I was working on an RA for him.
I do hope/wish more people would do RA nominations for others. It’s a great feeling to know you’re advocating for others in the community!
I appreciate @calurduran treading very carefully around optics.
I participate heavily in these Community Discussion threads and don’t think it would make sense for me to hold back support for what I see as good work. Things would perhaps be different if I could trust a ton of folks to chime in, but that has simply not been the case historically.
I second calur’s call for more folks to write RAs, and would add a call for more folks to participate in discussion/support in the community threads.
Can I apply for funding for my discord bot?
@underwood if you think it’s valuable to the DAO, ofc. It’s up to you to convince the GMC of the value, and up to them to judge.
I support @uisce’s application. I think the mint fee and Lido’s program (though a bit more involved) are both useful references. This holding is fairly stable, and important both directly and as an exemplar. 0.1% feels about right to me.
@FeelingoodFeelingrt Could you please move your application from the grants thread to the retrospective awards?
Retrospective Award: A proposal submitted by an individual or group that proposes a payment for a previously-achieved result. The retrospective award SHOULD detail the work that was completed and the positive impacts for the protocol that would merit such an award.
I want to opine that @yokem’s work is quite valuable. I’m working with ramana on mev theft. This data would allow us to better analyze what is a vanilla block (right now it’s just RP approved relay or not). A “second opinion” would also be inherently valuable. The archiving function is also important - bloxroute has stopped providing bids on their relays back far enough, and their ethical relay no longer has either bids or actual rewards.
@yokem - I would suggest getting rid of the “rocket” in the name and applying on gitcoin as well - this is an RP good but also an Ethereum good. I could imagine Lido chipping in too.
As I commented last time, I think RP should not pay for this unless we believe it presents a significant value to RP. Last time I asked, the response was essentially mathematical beauty:https://dao.rocketpool.net/t/july-2023-gmc-community-discussion-of-submitted-applications/1935/15?u=valdorff. If there are new reasons, would love to have those highlighted.
Have you conducted user research specifically on staking experiences and pain points before?
If so, could you provide examples of how your past user research insights have driven impact and change?
@Slevin_IPOR_Labs I see little benefit to RP justifying an $80k expense. For example, Pendle have rate exchanges already. Being able to exchange shorter time period rates (below a single update) provides negligible value imo, potentially even negative (I routinely try to get folks to look at 30+ day time periods that have more meaning instead of single update or 7-day). At a high level, it’d be nice to be included in the protocol if IPOR can find an active market – in such a case, the clients are the end users, not us, and where I would expect revenue to come from.
Wanted to comment on the DendrETH grant proposal. I think contributing in some form would be value additive to RP for the following reasons, both practically and optically:
As far as direct usefulness for RP, the oDAO exists today because solutions like this aren’t available yet. This project represents a method for decentralizing one of the most important oDAO duties: rETH balance updates.
As DendrETH expands functionality, there may be further benefits to RP. This is speculative and would require more analysis from the dev team, but we can think of several ways this project could directly impact RP down the line. E.g. improving rETH APR by requiring 2 or 4 LEB NOs to maintain certain performance requirements – currently something that only permissioned protocols can do.
NodeSet is seeking a trustless oracle solution for xrETH price reporting. For our v1 launch, we’re leveraging an optimistic fraud proof solution with an existing service provider, but would like to transition to verifiable computation solution like DendrETH as soon as is practically feasible. It’s simply more robust and extensible in the long run.
We were thrilled when we discovered DendrETH, as it would solve all of our design goals for an oracle system. If NodeSet is value-additive to the RP ecosystem, supporting DendrETH is indirectly supporting Rocket Pool proper.
We strongly believe in the value of public goods funding. The fact that numerous entities (including the EF and other staking protocols) are funding this project speaks to its legitimacy and ecosystem impact. Stepping up to support DendrETH seems like a great opportunity to burnish RP’s brand and strengthen ecosystem relations.
This is work that the Etheruem staking ecosystem needs, and obviously Zahary and his team have the right technical expertise and experience required to develop this project. Therefore, we think DendrETH would be a fantastic project to support.
@NickS Other than funding a public good, do you see any significant value over an optimistic fraud proof method?
You mentioned robust/extensible - what do you mean? In either case, adding a new thing to verify requires new logic (either to enable showing an optimistic proof fails, or new zk circuits). Both are robust, though optimistic does require a delay and incentives to work (zk also requires incentives to run the heavyweight hardware).
User research can impact orgs in multiple ways: commercial, increasing user adoption or retention, improving usability and satisfaction, improving marketing and positioning, and improving engagement from community/developers.
Below are some examples of the work we’ve done and the impact this has had (unfortunately, a lot of the research we’ve done in the past has been competitively sensitive and under NDA, and we’re unable to disclose the outcomes - something OpenUX aims to change by open sourcing our future user research findings!). You’ve asked for staking-specific impacts, and am waiting on approval from the folks at Pocket Network to be able to talk about the work we did for them, but in the meantime, the following (not staking per se), will hopefully give you an idea:
Liquity: Georgia is currently supporting the team with user research into the experiences of rETH and stETH holders to help them develop a new defi protocol. Having worked with the team a year ago to user test their Chicken Bonds product, they learned tons about potential users that would have been better to know much earlier on in the product development journey. The team decided to re-engineer parts of the chicken bonds protocol, in line with the recommendations, but this took time and resources. This time around, the Liquity team has engaged Georgia earlier in the process, before they write all their smart contracts. Not only has the research helped them fine-tune their offerings and products, but has helped them shift their mindset away from ‘code first, ask questions later’ and gather data as early as possible. When working with Liquity, we conduct multiple rounds of iterative research, so that findings build upon one another.
Impact: Shorter development time (faster to market). Improved positioning, usability and adoption.
For Glif, a Filecoin DeFi protocol, we supported the team with user research about FIL holders and Storage Providers (miners), and one of the most important outcomes of this research was the identification of the fact that capital inefficiencies on the Filecoin network can only be improved by breaking the current 1–1 relationship of miner and loan-provider. The assumption was that swapping out centralized entities with smart contracts would make this better for miners looking to borrow, but the research uncovered that a particular concern of miners would not be alleviated just by a smart contract, because there would still be a 1-1 relationship. Understanding this concern from Storage Providers led to a specific design choice for the Glif protocol. Glif allows Storage Providers to lease from multiple pools without being tied to only one. This gives them more flexibility and safety. Read more: Designing a Filecoin staking protocol that meets real user needs | by GLIF | Aug, 2023 | Medium
Impact: Shorter development time (faster to market). Delivery of unique value to users compared to competitors. Better positioning. High commercial impact.
One of our members conducted discovery research for an asteroid mining project incubated at ConsenSys. The research resulted in a total pivot, because the findings uncovered there was no real value to the intended user group, and there was not sufficient product market fit. This de-risked a major investment by ConsenSys and allowed resources to be allocated elsewhere without wasting any further effort and money on the initiative.
Impact: Risk-reduction, better resource allocation.
For MetaMask, I conducted 4 rounds of iterative user research to support the team with the launch of their swaps feature. “Georgia’s research involvement throughout the design and product development process resulted in one of MetaMask’s most successful feature launches - aligning core user needs with a delightful and intuitive user experience. Starting with generative research, Georgia helped us to better understand users’ mental models and motivations. These insights informed initial prototypes which then went through multiple rounds of usability testing to refine the user experience - resulting in an intuitive, user-centric, and widely-adopted product feature launch.” (Rachel Cope, head of design at MM). After launch of the swaps feature, performance was measured against the earlier testing. The changes incorporated before launch resulted in reduced errors, greater comprehension, and higher conversion.
Impact: High commercial impact, better usability.
Happy to answer any other questions, and will share more here about staking with Pocket Network IF they are happy for me to share.
@Valdorff Thanks for your feedback.
Although yield stripping and interest rate swaps (IRS) (or staking rate swaps – SRS) both fall into the category of interest rate derivatives, they have significant differences.
To start, let’s consider how IRS in TradFi typically work. In TradFi, two parties reach a mutual agreement to swap their interest rate payments – one receives fixed-rate interest payments and the other receives floating-rate payments. Hence, the only thing that gets swapped are the interest payments, meaning that the risk is isolated to interest rate fluctuations only.
Person A who is currently receiving a fixed interest rate of 5% believes that interest rates are likely to increase and they would like to have more exposure to a floating interest rate return.
Person B has been receiving a floating rate of LIBOR (4%) + 1% and is pessimistic on the future outlook of interest rates, believing that it will reduce their interest rate returns. Hence, they are motivated to lock in their returns at a fixed rate.
Both parties agree to enter an interest rate swap contract where A will pay B 5%, and B will pay A LIBOR + 1% upon the maturity of the contract, whatever the LIBOR is by the end of the agreement.
You can already see the difference with yield stripping protocols.
- Someone who wants to lock in a fixed rate because they believe rates will go down can purchase PTs.
- Someone who wants increased exposure to the floating rate because they believe rates will increase can purchase YTs.
Therefore, PTs work like zero-coupon bonds where the instrument trades at a discount and is expected to appreciate towards par value by the end of the maturity. Similarly, YTs work like coupon payments – by holding YTs you continuously receive the yield from the underlying. And because the price of YT depreciates to 0 upon maturity, the price of PT goes up. As a result, the prices of YT and PT are meant to be inversely correlated: when YT drops in price, PT increases, and vice versa.
This setup already presents multiple differences with how vanilla IRS work in Ipor, where you enter a contract against the liquidity pool (peer-to-pool) model. In yield stripping, the fixed yield can go negative if the underlying tanks, since the price of PT moves in parallel with the underlying. This is very different from vanilla IRS, where risk exposure is isolated to fluctuations in interest rates only.
I can recommend this report from Delphi Digital:
For example, it says:
Capital Efficiency – With the yield stripping method, the entire value of the principal is required up front. This is far less efficient than an actual IRS, where little to no collateral is required up front and only the discrepancy in cash flows changes hands at predetermined intervals.
This also means that rETH holders can hedge their staking rates without having to forgo the usability of rETH, for example as collateral in a lending market, if they use an SRS.
While both approaches could be used to implement hedging, speculation and arbitrage trades, the IRS approach offers more flexibility and granularity (rolling dates/tenors, essential for arbitrage trades and effective hedging) as well as significantly superior capital efficiency (IPOR is offering up to 1000x leverage for some markets).
Therefore, SRS are also better suited for speculation on rising or falling staking rates.
Another advantage of SRS is the deep liquidity, which enables hedging of very large positions. In yield stripping AMMs, one must take into account the second order effects of the underlying formula. In the past, PTs and YTs had separate liquidity pools, which led to liquidity fragmentation. While this issue has been fixed now, modern implementations use a single pool of SY-PT liquidity, where SY stands for Standard Yield, such that:
P(PT) + P(YT) = P(Underlying)
This leads to illiquid conditions for YT, which is only a single % of the pool’s TVL. In addition to that, the price impact also affects the leveraged yield that you can earn from YTs. Because of that, it only takes a few buys for the yield to turn negative, even if the APY of the underlying stays constant until maturity. If the future yield of the underlying is greater than the cost of acquiring the equivalent in YTs you make a profit, otherwise you incur a loss.
By design, a yield stripping AMM creates an imbalance in the exposure between PTs and YTs, meaning that the risk exposure is asymmetric as well. As an example, someone holding 10 $ETH and who would like to get a fixed rate on their holdings would need to have more capital and accept the risk of exposing their principal amount of 10 $ETH to the volatility of the market – which is completely different from a swap between a spot fixed rate and a spot floating rate.
We are in close contact with numerous institutional investors and regularly hear from them that a key consideration for investing in LST is the possibility of hedging via an SRS. There’s a reason the traditional IRS market is ~$400T large.
There is also a difference in the terms of the derivatives. Yield stripping is usually designed for long periods of a year or longer. The reason for that is because they build on the premise that the relationship between yield and price is inversely correlated (same relationship in DeFi as in TradFi). However, you can tell from experience that, looking at the APY of the underlying rETH, the APY drops when TVL increases. As deposits increase, the APY drops, and in a yield stripping AMM this translates into a drop in YT liquidity and a drop in the fixed rate you can get from holding PTs. At some point, when the discount on PT becomes insignificant, one should wonder whether it makes sense to favor a fixed rate over a floating rate. In fact, the discount on PTs comes from yield and the yield on YTs comes from the yield of the underlying. Therefore, the APY gets diluted by either TVL or price and, because of that, the longevity of each pool offered in yield stripping is inherently tied to the longevity of the yield that comes from the underlying. In other words, the pool is only useful while the yield lasts. This is something worth noting, especially when it comes to liquid staking derivatives now that the Merge has been completed and withdrawals have been enabled, leading to increased participation over time.
SRS are offered at IPOR with shorter terms, currently 28 days, and in a few weeks with the release of v2 also with 60 and 90 days. Longer tenors should follow in the long term.
Finally, SRS are also very suitable as components of a structured product.
I can understand your concerns about the usefulness of an index based on epoch level (6.4 minutes), but I see it a little differently: The index would be updated at very short intervals. However, it would show a 7D MA. The swaps would also be based on this 7D MA. It is also possible to show longer moving averages for informational purposes.
I hope I was able to clear your doubts. If you have any further questions, I’m happy to help at any time.