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 round 7 period of grant, bounty, and retrospective award applications.
1 Like
I’ve just submitted 4 RAs for the random things I’ve worked on since finishing up my Rescue Node stuff.
It’s unclear to me what types of work the GMC wants to remunerate so I figured I’d throw spaghetti at the wall and see what sticks.
I won’t be upset if it doesn’t stick.
Edited to add: The amounts requested are based on what I’ve historically been paid at small startups where I have high equity upside, pro-rated by day (excluding weekends and holidays), and discounted 25% to 50% depending on the type of work (less technical, bigger discount). I have trouble putting numbers on things and this formula felt like a reasonable starting place.
1 Like
I submitted 2 more RAs for the remaining bits of work I did that may or may not be things the GMC wants to fund. In this case, they’re both things that I consider to be the responsibility of the maintainers of smartnode. Whether the GMC wants to encourage community contributions on that front, or leave it to the RP Pty Ltd team, I leave to them to decide. I have observed a growing backlog of smartnode features and hate to see it languish from headcount issues. I am excited for the new dev(s) to make headway here, and if the GMC sees funding these RAs as a stopgap until the new dev(s) come up to speed, please let me know, so I can refocus my efforts.
All right… let’s pump out some opinions 
Retros
/node_tvl_vs_collateral RocketWatch command
Not much of an opinion any direction.
I agree that this data hasn’t been visualized like this. Closest I’m familiar with is https://harpocryptes.github.io/, which is able to answer the question of “what amount is typical” that’s used as an example, however it would not have the “per user” resolution or show how it’s impacted by “user scale”.
RP book
This is one of those things that lives and dies by how much it’s used.
Off-hand, it’s missing the rp_community dune dashboards (primarily lst comparision), treasury spreadsheets, the defi opportunities spreadsheet. This isn’t a knock on patches or the work - it’s intentionally meant to be community-maintained. There were updates on the first 3 days it existed and there hasn’t been any since then. I worry about the bars of “knowing it exists” and “needing to use git to update”. Standing here today it’s hard to judge whether this will be critical infra, or a black hole that won’t get used.
RPIP-30 preview trees
I’m much too involved in RPIP-30 to have an unbiased opinion here. Pass.
Treegen v8 Thoughts / Spec Proposal
I’m an awardee on this one, so feel free to ignore me. I do think this will save time when Joe gets here, which is beneficial.
Committee Membership Record (RPIP-36)
Strong resource we can point to when ppl have questions. Like the contracts integration page, but for committees n such.
RPL Defi - Rocket Pool DeFi Market Opportunities
Not much commentary from me. I would suggest a quick look at view/engagement metrics for the videos, blog posts, and threads vs RF or other things we pay out to see if the levels seem reasonable.
Smartnode Rescue Node Addon
Feedback about usability has already been positive. The real benefit here is for the NOs and #support folk that help them if they mess some configuration up – value in terms of minutes saved in aggregate is quite significant. The quoted price seems low, if anything.
Fix truncation of user-settings.yml by smartnode when ~/.rocketpool device is full
I support patches getting issues out of the way for Joenax (or in this case preventing it from ever getting in their way). If at some point, they are less on the critical path, I’ll change my tune. Contract rate seems kind (to Rocket Pool - thanks patches!).
IPFS CID calculator without web3storage dependencies
I support patches getting issues out of the way for Joenax. If at some point, they are less on the critical path, I’ll change my tune. Contract rate seems kind (to Rocket Pool - thanks patches!).
Grants
RocketPool Chat
As I’ve opined on similar thoughts before, I worry that “often correct” is the best we’re likely to get and an unacceptable bar. Real humans are better at knowing when they don’t know and calling for help. I would hate to see someone get dissolved, mess up wallet hygiene, etc, etc based on a chatbot. The small set of curated data sources, which is a reasonable approach, also severely limits utility. Eg, there have been times when #trading is well aware we’re going to slip a quarter or more on a plan, but that info hasn’t trickled into a biweekly update.
Bounties
Good golly we need to do more bounties. This is getting silly.
Rapid Research Incubator
I can’t support a bounty more than this. It’s great. We need to brainstorm, we should incentivize.
Bounties
Rapid Research Incubator
Fund it. $10k seems cheap if it produces a good idea that can save the protocol.
Grants
RocketPool Chat
Don’t fund it.
Sample questions seem weak. The questions are something that a good FAQ page could answer.
$10k for proprietary work that uses a paid API. Their open source work will be all weekly updates and docs compiled in one document that you feed to the model.
I’d rather fund someone who uses an open source model, fine tunes it, open sources everything (so anyone can run and host it), the community uses and tests it and then we give a retro award for that work if it’s good. This is something that should be a passion project that comes from the heart, not a paid deliverable.
Retros
Patches
Fund everything. Pay him double so he’s motivated do even more for Rocket Pool.
Committee Membership Record (RPIP-36)
Fund it. $800 seems high for one page. However, the costs are broken down and make sense. Time is money. Good things cost money. Plus we want them to do stuff for us.
RPL Defi - Rocket Pool DeFi Market Opportunities
Fund it.
I think the production quality is poor, the content quality is mediocre and the impact is non-existent (two videos I watched had 14 and 24 views).
But everyone successful started with crappy videos and we want people make content for Rocket Pool. $1200 seems fine as an investment in them.
Governance facilitation
Fund it. Pay him double. Who else is going to do this?
2 Likes
I’m not sure if I can leave a comment here or if it’s only for community members to discuss the applications, but I’ll give it a shot.
@Valdorff
You made some great points. In the near future, AI won’t be able to fully replace humans in every situation, but we’re getting closer every day. I think it’s advantageous to start experimenting with AI sooner rather than later.
I appreciate your understanding of the limitations of sources and how that affects the utility of AI. As we mentioned earlier, this shouldn’t be the final product that satisfies all users at once. Instead, it should be seen as an experiment to determine if RP should explore this direction further through this low-cost solution.
Yet again, it boils down to taking a proactive stance toward the latest technological breakthroughs and how you can leverage that technology to your advantage.
If any of your competitors were to launch this chat and attract a substantial user base, how would you respond to that? #Rapid Research Incubator 
@peteris
I’m concerned that RP may not have a way to gather questions from newcomers to crypto. So, I’m curious about the basis for seeing these sample questions as weak. Do you think that any of these questions might never be posed by new or current members?
Unfortunately, we can’t make the rest open-source since it contains our unique know-how. We’re requesting $7k, not $10k. While the solution you propose is feasible, open-source models often come with lower performance, high fine-tuning costs, and high AI engineer costs, among other challenges, so all that could reduce the likelihood of your suggested approach succeeding.
Thank you! We appreciate all feedback!