January 2023 GMC Community Discussion of Submitted Applications

Grants / Bounties
40

GM! Just wanted to state the obvious in the case of the grant proposal to translate all Rocket Pool documentation to Spanish.

Adjusting the proposed compensation to USD-denomination or reducing the requested RPL amounts, based on the ongoing market movements, would be totally OK for me. I realized after the weekend that the initially proposed RPL-based compensation became a bit unreasonable in FIAT terms (even for a technical/specialized translation work). Happy to discuss this further if this proposal would be relevant for the committee.

41

At the risk of being self absorbed, I assume this comment is in reference to my questions on #trading, and if so I apologize that I gave the impression that Knoshua deserves 25$ an hour for his time. I’ll chalk it up to trying to type while chasing after my 20 month old. I’m agnostic on whether he deserves 50, 100, 200, or 500 per hour.
My intended point was that the pDAO treasury literally cannot pay people what they deserve for their work.

the thought experiment is as follows, skip for TLDR:

Waq requests 1860 RPL. Butta requests 2000 RPL.
Rocket scientists: Jasper spends a lot of time working on RP; probably spent 200 hours on his paper which created a lot of beneficial buzz, and has a strong twitter following, i’ll say give 600 hours total in the last two years. Val spends more time answering questions from noobs than I spend awake, I’ll say 1200 hours. Object 700/patches 1200 (i confirmed, patches has 74k posts on discord, most in #support, so that’d be like a post per minute). Yorick mostly shitposts but based on pure volume and i’ll give him 250 hours. ken- 500 hours, to include his work on LEB.
These 8 people (6 rocket scientists and 2 strongly aligned and influential community members) I mentioned are all technical, and fill a needed niche; they have all also been critically important for rocket pool’s success. At 100 dollars per hour, here is 4450 hours- or 14,833 RPL (30$/RPL); add on the beaconchain and rocket fuel and this is ~18633.
The total amount that can be given retroactively if we expend the treasury is ~19k.
But then all the other players working hours for RP on support: Mig is constantly in support- 300. Lilac; fornax- 200 and 800. deetoo- 300. Ramana- most recently people floated 22ETH (1200 RPL), but that’s just for rocket arb; he’s a frequent support contributor- maybe another 200 hours. Then all the theses: Xer0.eth, marceau- 80 hours each- and I’m sure marceau’s time is worth a lot. Then add some of the thought leaders: uisce, dondo, uisce, knoshua, a35, kevster, vvander, pieter, invis, calurduran, hodja, hertzen, enkriptix and many many many more- those who have shaped protocol policy in a major way- thousands of hours. Major Marceau achievements like the whale marriages; those who put forward liquidity early before funding was secured.
Then all the individuals contributing to the various governance debates, in governance discord and Reddit. Hundreds who had to process the various arguments for and against different snapshot vote allocations, or weigh the arguments for different amounts of RPL collateral; those that spoke and those that voted have been critical to how functional our governance is compared to other protocols, which is a major selling point for Rocket Pool.
Then the personalities: bankless, sassal, superphiz, i’ll add marceau again- these people have been critical to raising awareness of the protocol.
Then the projects that have integrated with rocket pool and given the unique flavor and visibility: POAP, ethstaker, rocket pool dashboard (in early days), rocketscan, etc

Essentially, the point is that hundreds of people have been involved in this process and in making rocket pool great, and the pool of funds is not big enough for everyone to be even close to compensated.

Problems with paying our contributors what they deserve in hourly wages retroactively:
  1. There are not enough funds to pay all worthy individuals even at a heavy discount.
  2. Paying some people and denying others for their time risks creating conflict (specifically conflict between GMC members and community members). Same with choosing to pay some an hourly wage less than others for similar work.
  3. What one’s time is worth differs greatly by where they live (can be by a factor of 10), and rocket pool probably shouldn’t be in the business of reinforcing those inequalities.
  4. Setting high hourly rates for retroactive pay will set unsustainable expectations of future work.
  5. Selecting a few people ignores the many many contributions of other community members that have helped elevate these few (answering questions, retweeting, giving insights, reviewing code, just consuming the content etc).
  6. Hourly rates ignore the benefits of efficiency (ie, should a person who works 10 hours and reaches 1000 people get paid more than someone who works 1 hour and has the same impact?)
  7. Giving large payouts now has the paradoxical effect of de-incentivizing future public goods because the treasury will be extinguished for future rounds.

My recs (which fortunately I don’t much of a reputation or relationships to protect :sweat_smile:)

  1. Don’t bankrupt the fund on this grant round. Maybe restrict payouts to 2 reward cycles.
    There are a lot of worthy individuals that did not apply, and because the GMC is just starting the protocols for distributing funds are in their infancy.
  2. Those projects that have an obvious path to sustainability/profitability should try to make money rather than relying on RP for grants. Projects that have an income source or have received donations for their work should declare these so RP is not double paying. Be very cautious about giving grants to for-profit companies (outside our community), as you are unlikely to get the same value as community members can provide.
  3. Grants should focus on seeding projects, not provide ongoing salaries.
  4. Awards should focus on attempting to quantify the measurable good that contributions have for the rocketpool community; hourly rates are somewhat immaterial to this.
  5. A member of the 7 person GMC requesting ~9% of all future funding is a conflict of interest whether they recuse themselves on that particular vote or not; it will affect every single other vote in a resource constrained environment.
  6. I hope that in the future more grants will be written where the writer is not the sole beneficiary.
7 Likes
42

Great comment. FWIW, I think it’s appropriate for applicants to be clear about how much they would expect to earn in their normal job market, for example, and then for the GMC to approve their application for a completely different amount determined by completely different factors.

I think I agree with this as a rule of thumb. I also think that we could really use a full-time researcher. It probably shouldn’t be a hard rule.

43

Quick comment about translations: let’s make sure that any translations we fund are delivered in a maintainable way, from a tech POV. The format of translation files is something the team should probably be involved with (@kane & @jcrtp)? This is especially important if multiple translators are approved this round.

Some things to consider:
  • The RP website is not open source.
  • Translations should allow for pluralization, number formatting, date formatting.
  • It should be possible, in the future, to collect translations from new sources (a vendor, via crowdsourcing, etc. This should not be an onerous process).
  • How should new copy be deployed? In particular, what if the team has English copy ready to publish but new translations are not yet available?
  • How can translations be shared between projects? We may want to avoid double-paying.
44

I must say some of the requested amounts shocked me in either direction.

Im not questioning the massive amount of time and passion many have poured into this project, especially since i havnt personally contributed anything much besides regular Twitter engagement and a POAP Design here and there.

But seeing massive amounts of funds requested for tasks that were surely (hopefully) done out of passion as a hobby by retrospectively applying senior software engineer hourly rates that may or may not fit to what was specifically contributed seems iffy to me.

In my opinion, those grants should never become something that actually attempts to pay senior hourly rates but something that nicely rewards contributions that were first and foremost done out of passion and as a hobby.

And from what i saw most requests did exactly that.

Really not trying to step on any toes, many are contributing so much and are absolute pillars of the community… i just think there really needs to be a philosophical discussion on what those Grants should really do, especially since this will set a huge precedent going forward i imagine.

love all of you guys

Edit:

I aknowledge that i might also have a complete misunderstanding on what those Grants are supposed to do. In that case i think it would be important to make sure there is a community wide understanding of what that exactly is.

4 Likes
45

I did want to jump in and say that this being the first round ever, I don’t think anyone knew how to value their work. It’s fine for people to have come to the conclusion that the number they give should be their usual hourly rate or X% of the value they generated for the protocol, so long as they are also cool with the fact that they likely won’t receive all of it.

1 Like
46

Note: as a retrospective applicant, I’m probably biased

I would like to offer a rebuttal on some points. Overall, I think the GMC can afford to pay people a similar order of magnitude to what they deserve.

Retrospective rewards are already capped at 50% any period’s total awards. Thus, vesting retrospective rewards across longer time periods, such as a year, reduces the burden. If our total backlog for ~3 years of unpaid labor is only 4500 hours, that’s not bad. Our annual GMC inflation is >$1m USD. We can afford the whole backlog at 100/hr and still be under the 50% retrospective cap. Cut the average pay to $50/hr and we can double the backlog to 9k worked hours and still be under the annual inflation. And this is just the first year! Retrospective rewards are front-loaded and will taper.

Additionally, many people volunteer their time without expectation of reward. Among the many wonderful contributors you mention, which I appreciate very dearly, I suspect most will not request payment for broad contributions, but only for concrete deliverances. Patches for example had to be forced and ghostwritten in order to get a request for Support backpay.

I think the contributions you mention could be served well by a quadratic funding round as suggested by CLR.fund. This can reward many contributors each a partial sum weighed by the community. I think this may be more palatable to contributors who shy away from recognition.

As for your recs:

  1. Agreed. In fact, I do not believe we should be tapping into reserves when possible. Vesting should be preferred to break up payments.

  2. Agreed.

  3. Disagree. Grants in specific cases should act like full-time employment even if they cannot quite match the salary for it. See Knoshua’s proposal. As per my analsysis above, I believe we can support looking at grants like an extended dev team. In fact, IMO, it would be a failure of the GMC if we did not.

  4. Partial agreement. I suggested earlier a base salary with an impact modifier. It will be very hard on the GMC to only rely on subjective metrics.

  5. Disagree. I don’t follow your argument. Every vote affects every vote. People should give a logically arrived-at value with the understanding the GMC is free to re-appraise as they will.

  6. I’m not sure I follow. If the beneficiary isn’t pre-specified it should be a bounty.

1 Like
47
  1. I think we should be fine here - we can’t spend more on retroactive than on forward-looking stuff by design.
  2. I agree that if projects can make money the should do so. Even if it can’t cover all costs, it can reduce RP’s costs and allow other uses for the funds.
  3. I tend to agree that “salaries” are not a great structure. Large recurring expenses rapidly drain our ability to further reward other things. There may be exceptions, but I think we should primarily be looking to get things off the ground, but not feed them indefinitely.
    In retrospective awards, @waqwaqattack asked for 310 RPL/period (I believe RPL was around $20 when that was posted), and @jasperthegovghost asked for $150k for ~14 periods ($10.7k/period). If this applied going forward, that would be 17% of the total budget at $30, or 34% of the total budget at $15. I think this is dramatically too much concentration. Waq posted looking for the same going forward. @jasperthegovghost - you haven’t posted for a forward looking grant - is this because the work is unpredictable, because its value is unpredictable or because you’re not looking to request funds going forward?
  4. Yes and no. If I create X value doing a task only I can do vs creating X value doing a task 60 people can do – supply and demand says we should likely pay less for the latter. In both cases we need good ROI vs X, and that’s much easier to achieve when there’s many folks that can do a thing. Relatedly, a person with a highly specialized skillset takes on higher opportunity cost by contributing to RP - that should not be ignored if they use their specialized skillset here. Note that a person with a highly specialized skillset doing something anyone can do should not be paid the high rate.
  5. Hmmm… I’m not as bugged as you by this one. I think the main issue is the #3 issue that it’s just a large piece to set aside in general.
  6. Me too :slight_smile:. And better yet, more bounties (I think we had 2?)

Relatedly to number 4, I’d like to reduce my ask on https://dao2.rocketpool.net/t/january-2023-gmc-call-for-grant-applications-deadline-is-january-15th/1335/4?u=valdorff to 4 RPL every 2 weeks (from 10) to account for the big price jump and for the fact that this job is not highly specialized.

1 Like
48

A variety of factors went into my decision not to request future funds. First off, I do not know what my degree of involvement this year will be with pending med school applications. Further, I think the vacuums I filled no longer exist. The IMC is the defacto bizdev team for protocols now. The RPIntern is well established, there is a grant request for Twitter educational materials, and our overall Twitter presence has become very solidified.

I believe I filled a unique transitionary role that is no longer needed, at least not in a manner that requires compensation beyond oDAO + stipend.

49

I understand the price has moved a lot since I submitted my applications. I will accept a lesser amount for my grants as long as the value of the work is still acknowledged. I will leave that to the GMC to decide.

50
  1. I agree rewards should not be forward looking:

  • It is hard to predict future value of the token and the contribution.

  • It disincentivizes new contributors if the budget is fully depleted.

  • There is a perceived psychological downside of the grantee not contributing at the same level of effort if funds are guaranteed.

  1. There should be a reasonable distribution of funding that attempts to promote marketing vs development. I.E. If a heavily unequal portion is distributed to developers, this disincentivizes marketers presently and moving forward. Marketing is one of the biggest weaknesses of our protocol. There was far more development work done, to balance this there should be a slightly different weight applied to marketing contributions.

  2. I also do not agree with using competitive industry wages as a baseline. Comparative value among other projects should provide the baseline for rate, then factor in total time spent and ongoing time spent.

2 Likes
51

This is why I think we need to start thinking about how to diversify into stables, etc. (imagine if the price was down right now instead of up). Maybe the GMC can flag the relative importance of that as time goes on.

1 Like
52

So… for grants, I don’t think it’s an issue. If you agree to do something for 5 RPL :man_shrugging:. Grant askers can consider whether they’d like to ask for (or accept) slightly more than they would in current-usd to account for volatility.

For bounties, I could see some challenges where a big price drop might mean bounties rot on the vine and a big price pump might mean we overpay. I think this only results in issues for dramatic price shifts.

I actually think retro is the most affected here.

I think this is fantastic, but more difficult to find. Keep in mind the GMC is all volunteer and there’s a lot of proposals to seek rough pricing for. In many cases I suspect googling for a market rate will be much easier than finding comparable crypto projects.

2 Likes
53

Thanks again @calurduran for the guidance.

All GMC members, and Rocket Pool community at-large are invited to take the following information into consideration for the assessment of our team’s proposal as it appears that a clarification is required with some added notes:

This proposal is made to localize the primary section in protocol’s documentation into 5 languages (OFD), and serve as a proof of quality for the translation of all remaining written content.

Given the undeniably sobering market conditions and relatively high cost of translation into 5 languages, poor quality would not be a tolerable risk if the scope involved entire documentation (approx. 100k). Therefore, our team is glad to offer the Rocket Pool protocol the opportunity to have its primary content translated by a highly experienced team, and depending on the output, assign the remaining work to an already-proven team by avoiding risk of any loss in terms of both money and time, and benefitting from the advantage of providing feedback throughout this process.

The scope of translation:

  1. ‘Overview’: Main informative section in the documentation aiming to guide new users around the basics of protocol, complemented by FAQ and Glossary.

While the scope was initially limited to ‘Overview’ section excluding any external links (for the sake of orderly completion of translated sections), our team decided to involve one article for the community to test the advantages, and thus necessity of translating the Medium content (blog articles in addition to Explainer series).

  1. ‘Rocket Pool — Staking Protocol Part 1’: A Medium article that is specifically recommended to be read by users under the ‘Introduction’ part of ‘Overview’ section.

That being said, the rate per word is $0.1 (or RPL equivalent) including the entire translation and review work. Rocket Pool is welcome to offer any changes to the proposal including selection and/or extension of translation content, selection of languages, or any other special requests.

The exact word count will be finalized at the initiation of project as files in the protocol’s repository will be the most accurate source.

Regarding considerations suggested by @Dondochaka:

We are planning to use Crowdin as a CAT tool for the entire project. Crowdin offers various integration capabilities such as synching translatable content from the protocol’s repository with our project so that you may receive translations as pull/merge requests.
The formats available are listed here.
New English content may be copied as is to all local pages until the relevant translations are complete and uploaded to RP repository.
All translations are saved in a sort of database file called Translation Memory [TM], which speeds up the entire process and improves consistency. TMs can be used for future projects as a tool to pre-translate files (i.e. applying the saved translations to the identical files). GMC is welcome to ask further questions with regards to use of CAT tools.

54

I think this is fantastic , but more difficult to find. Keep in mind the GMC is all volunteer and there’s a lot of proposals to seek rough pricing for. In many cases I suspect googling for a market rate will be much easier than finding comparable crypto projects.

I think my original quote was misinterpreted. I meant a comparative value among other projects / applications requesting funding from the GMC.

Fictional examples:
Project A - Generated 100s of new users and saved the protocol thousands of dollars.
Project B - Fixed copyrighting on a number of protocol’s documentation.

Project A should have a much higher baseline for rate (even if the professional industry rate is different).

I’m sure it would be hard to gauge value between all projects, but I’m sure some would stand out and be easier deciphered than others.

55

Ah, yes I see. I agree with that too. ROI is the ultimate metric, and it seems like that’s what you’re asking for.

I think the market rate bit is needed to understand what an acceptable “I” is in ROI. I don’t think we should ask people to spend time on something if we can only pay, eg, 1/30th of a fair rate - I think that’s literally less rewarding than volunteering.

Armed with a reasonable “I” and a best guess at a rough “R”, we can then rank and decide which projects to fund.

1 Like
56

I think yours is a very reasonable take, I’ll put forth my few small clarifications. The 4500* hours to expend the entirety of the pDAO reserve is my estimate is for 6 random contributors I chose; it ignores the hundreds of other contributors and probably 10’s of thousands of hours.

5.

This referred to the rocket fuel proposal. Once 100% of funds are committed (which will happen at some point), every future approval vote means a decrease in an already approved project. The conflict of interest occurs when a vote of ‘yes’ is effectively a vote to cut your own salary; given the size of the prospective grant, I think those ‘no’ votes could give the appearance of impropriety (although certainly not intentional given Waq’s ethos and commitment to RP).

6.

The rocketpool community is extremely collaborative, so next round I hope we see collaborative prospective grants, and more retrospective requests that primarily benefit people besides the author.

I think the choice comes down to whether we want the GMC to give large awards to a few very high achieving community members or spread out smaller awards to more participating members. The latter is obviously more equitable, but I would argue it is also much more beneficial for RP, because people who are acknowledged with even small awards are more likely to contribute in the future (see POAPs).
Ultimately, I think this is something that deserves to be discussed by the community, and not just left to the GMC to be the adults in the room and ‘decide what is best.’ So thanks to everyone for your critiques and discussion!

aside on knoshua’s grant proposal

I think this grant could easily pay for itself 10x by increasing efficiency and decreasing fragility. And his work on oDAO alone over the last year is worthy of this amount. And I think that the current oDAO implementation is almost non-sensical, to have such a critical piece of infrastructure run by individuals chosen for their non-rocket pool passions and not technical expertise or dedication to RP.

My concern is over risk of failure at the end of knoshua’s efforts, specifically:

  1. conclusions/implementation pathway are not agreed upon by pDAO

  2. conclusions/implementation pathway are agreed upon by pDAO, but not agreed upon by RP team (and consequently oDAO) [from its premise, any reduction of rewards to oDAO reduces the public goods funding that the team has stated is its raison d’etre]

Knoshua’s brilliance can be uncompromising at times, and good ideas/correct ideas are not always palatable to the community (see https://dao2.rocketpool.net/t/redesigning-no-rewards or the historical Jesus Christ)

I think the safest path is to reach some consensus FIRST as a community and team regarding some of the tricky oDAO questions, and then fund knoshua to chart the way forward; an easier alternative would be give stipends to a select group of community members (probably picked by knoshua) to act as a sounding board to refine good and temper controversial ideas.

That being said, I think funding the proposal as written is preferable to not funding it at all.

57

I’m proposing to review the state of the oDAO: Identify duties and their purpose. This is an aspect of the protocol that is currently undocumented and, in my opinion, not very well understood by the community.
Furthermore, I’m proposing to explore alternatives. This will be pretty technical. For some duties, there might be multiple alternatives to explore. For some, their might be no feasible alternatives without major changes to Ethereum itself.
There will be work-in-progress releases of what I got after 4 and 8 weeks. I’m also happy to give more informal updates as I go, like posting stuff in the #research channel in discord. There will be plenty of opportunity for people to give input, if they wish.
I think it’s crucial to do this first, to give people a way to approach these topics. We can’t reach consensus on something that quite frankly nobody has given much thought. I am trying to lay the groundwork for the pDAO to decide these things, not preempt any decisions.

10 Likes
58

This seems like a sensible approach to me as well. Back in the fall, I spent a few hours working on a draft charter outlining duties and expectations for the oDAO, but found it an unproductive angle.

  • There are indeed many technical duties, so either you’d gloss over these or quickly get into a lot of detail.
  • You can’t effectively define expectations for the current state when we don’t even agree to what extent the current state is desirable or how it should be changed.
  • Developing a high level community mission / vision for the future oDAO first would be more effective;
  • But even this benefits from a better understanding of the current state of affairs and the possible future paths first.
59

FYI for those monitoring, a new forum post with the award announcements should go up in approximately 30 hours +/- 2 hours.

4 Likes