[Informal Poll] - Increasing the SourceCred payments

Is it true that we are currently underrewarding the type of work that happens in the forums? There is certainly some very important work that happens in the forums (like risk analysis and new MIPS) but I am not sure those are proportionately rewarded by SourceCred. Unilaterally increasing SC rewards will mean we get more of what we already get, which is mostly opinionated discussion. If that is what we want then so be it, but more opinionated discussion on its own does not result in consensus.

Honestly your comments and my own thinking make me wonder if this poll should also include reducing or eliminate rewards as I think that is just as valid a test as doubling them. One thing we already doubled rewards and it is my own impression that there wasn’t too much change. We do have some new posters here like yourself @AstronautThis with new views and input which is appreciated, but I think a general poll as to why people post here might be in order. I mean honestly rather than just increase rewards hoping for more participation (what does that mean really in terms of community growth - support - actual change) perhaps a view towards what the real community goals are?

Do we want more participation by people who own and vote MKR in governance? Do we just want more forum posts? Perhaps we just want more people around who might be able to do work or to be hired? Or maybe we just want to focus on the tasks at hand that still need to be done?

These all seem to be valid topics of discussion. What I am wondering is if we got a report on whether the last doubling of SC rewards really made a measurable change in ‘participation’ and what that means for the goals of the community?

I think I am going to have to switch my vote to abstain because of the questions above.

2 Likes

I’m happy to be involved in how to frame that discussion. However, if that discussion takes place on the forum or mostly involves people from the forum, then many participants in that discussion (myself included) will have a conflict of interest.

At the moment, I think SourceCred is set up such that this is rewarded. If we use forum post count as a metric of participation, it will indeed look like a success but I don’t know if that is the right metric to chase.

Voted no despite being on the top10 sourcecred payouts in the last months.

Reason: I don’t feel I have been contributing that much to the forum, mostly doing chores like taking care of DCs, signaling and working in the Rates Group / MakerDao Open Market Committee.

I always feel a bit dirty when I receive my payouts.

Imho we should rather do more with grants or put people on the payroll of CUs (like Risk and Governance is already doing it) than just using an algorithm to pay people. This can (and will be!) easily gamed and having more SourceCred payments is just increasing this risk.

1 Like

in my eyes, sourcecred can and should be continually “tweaked” to incentivize different behavior and types of posts… e.g. we could consider to compensate Signal Requests from community members that have “cred”. etc…

we are just starting the exponential phase of maker… we need the community to grow …

2 Likes

I’m voting ‘Abstain’.

As a word of context, I believe the aim of SourceCred is to incentivize quality governance participation that is ‘unbacked’ by MKR in one’s wallet. It’s supposed to encourage new/potential members of the community to go through oftentimes-tiring process of familiarizing oneself with the forum. The DAO is only as strong as our off-chain & on-chain governance communities, and so with the barrier of entry so steep, a person’s effort to participate should be rewarded.

It is my long-term view that SourceCred should serve off-chain governance in a similar way that Staking Rewards serve on-chain governance. Practically-speaking, it should seek to incentivize and reward an expertise-driven culture, where quality input is appreciated.

I’ve been running ‘SourceCred Outreach’ for over a month now. Its aim is to identify promising community members and point them towards SourceCred. Now, whether they see the resulting DAI as a small gesture of appreciation (most likely) or it actually improves their financial situation, I believe it might have played a role in encouraging several of our outstanding new contributors, e.g., @Aes, @PaperImperium or @aburban90.

From my experience, I believe the community-building aspect of SourceCred is just as important in the long run as any financial payouts. For this reason, I’m not stressed about increasing the payouts as the outcomes are positive regardless of that.

One important factor speaking against an increase is the rising incentive to game the system. As much as we’re prepared to handle additional moderation duties, the relatively-lower payouts are a natural defense against manipulations of the system and/or damage to merit-driven discussion culture.

6 Likes

I voted ‘Abstain’ as well. I think sourcecred is a fine program, but maybe it should be limited to a year or so per participant?

EDIT: I have changed my mind (and vote) after reading other people’ s comments.

I don’t think you should discredit yourself. It’s because of people like you and the other Community Members that I wake up every morning and the first thing I check is this Forum and the Maker Rocket Chat. Every bit of your participating helps shape & build this DAO. So, I believe you deserve every DAI and Conti that you have earned via SourceCred. You should check out other Community Forums that don’t incentivize their communities. Pretty much ghost towns…

7 Likes

Huge :+1: here.

I might be as bold as to add that you may not be considering how valuable your input is here @ultraschuppi ! Even if you discount the (extremely) valid points Frank brings up, the “chores” you do for governance are part of what keep this efficient machine running. How much unpaid labor do you think you were doing before SourceCred? As much as we rely on MKR holders being incentivized to set good governance, without real people putting in tangible work the rocketship never takes off.

For some prospective we currently spend less than 350,000 DAI on SourceCred per year. Without pointing to any Core Units, it’s fair to say that would be a minimal personnel expense (would be right around 3 facilitators in the GovAlpha budget). For that small expense we not only bring in new voices and opinions to the DAO, but also get the chance to say thank you and give back to dedicated community members that would be out here giving their time anyway.

8 Likes

100% THIS!

3 Likes

It is what large organizations spend on their office supplies. Bang for the buck, I personally believe it help bring new ideas. With what is coming in 2022, we will want to attract the next round of community members!

6 Likes

Right now, it feels like there’s a pool to be distributed about the participants. (Am I right on this one?)
If the participants grow in a given month, the Dai received by each will diminish.

Why are we aligning rewards in this zero-sum game? We want people to tell their friends to come, not the opposite.

Should we not try to find a Cred baseline and assign it a certain amount of Dai, and if more people are participating, we give out more Dai (and vice-versa)?

Exempli Gratia: in a “quiet” month, we might give out a total of 3k Dai; if the next month we triple the activity, we might give away a total of 9k Dai (we should cap it at some number to avoid Governance Alpha going bankrupt).

I’m not sure if it’s possible to tweak the algorithm to do this.

Unless my premise was wrong, in which case I’ll see myself out.

6 Likes

An excellent point, especially considering we want to be growing to pay more contributors all the time!

Practically, it’s super easy to input different distribution numbers in the SourceCred instance. Budgeting might be a concern though, as currently it’s up to the GovAlpha budget to support SourceCred so increases to payouts would necessitate a bigger distribution to the Core Unit. I wonder if there’s a way to automate the size, otherwise maybe a Parameter Proposals Group might be reasonable.

Well, some people are arguing that they’re receiving too much.
I think it’s just about finding the right number (I’m making it sound easy so someone tries, but, shhh, it’s a trap).

2 Likes

Voted Abstain for obvious reasons (I administer Maker’s SC instance).

Such an interesting thread! Loving all the intelligent discoruse :heart: Composing a mega comment now that addresses all the things.

3 Likes

Give credit where credit is due. Quality posts are valuable and should be rewarded to the fullest extent the DAO is able. This community has so many bright minds. I love reading the forums to see all the dialogue every couple days.

1 Like

Voted Abstain for obvious reasons (I administer Maker’s SC instance).

Below is some additional context re: SoruceCred’s relationship with the DAO, as it relates to this change, as well as inline responses to objections and proposed ideas.

SourceCred Working Group → GovAlpha CU

SourceCred has been receiving a 5k DAI/mo dev grant from the Foundation for prioritized support and development of new features. This grant program was sunsetted at the end of March. So, as @LongForWisdom mentioned, the GovAlpha Core Unit (CU) has taken over paying SourceCred. For transparency, while we have not yet implemented it, SourceCred has proposed (and GovAlpha agreed to) switching from the 5k DAI/month flat amount to SourceCred’s preferred tithing model (used by SourceCred’s other partners), whereby SourceCred receives 5% of the Cred and DAI distributed. In addition, I will be paid 2k DAI/month to continue providing priority support for Maker’s instance, as well as opting myself into payouts, as is standard practice in SourceCred’s other partner communities. At current distribution levels (~20k DAI/mo), Maker would be paying roughly ~2,500 less per month than under the previous arrangement. If distribution was doubled to ~40k DAI, month, SourceCred’s tithe would equal ~2k DAI/mo, and total cost would likely be around 4.5-5k DAI/mo.

While there is some potential for conflict of interest here, SourceCred believes the tithe is the best way to align incentives. If Maker decides to increase the payouts, here or on other platforms, its costs will likely increase as well. SourceCred community members contributing in partner communities are paid via SourceCred (and vice versa), and we haven’t seen any issues with that yet. On a personal note, I also requested to be opted into payouts so I would be incentivized to contribute here more. The discourse on here is great, and I’d like to post more on issues related to SourceCred, as well as decentralized governance, etc.

Governance of SourceCred parameters

Thus far, parameter changes such as increasing the distribution amounts, changing the payout formula, and tweaks to the algorithm have been decided by the SourceCred working group, which is now under the GovAlpha Core Unit (CU). Eventually, the plan is for the governance community to make these decisions.

Since the budget for SourceCred is currently under the GovAlpha budget, if MKR holders were unhappy with SourceCred, they could defund it. As @prose11 noted, this arrangement limits how quickly we can increase distribution amounts. The Parameter Proposals Group they suggest seems reasonable to me as well. I will leave it to the community and facilitators to determine proper polls, Signal Requests, etc.

Gaming

While this is certainly a concern, I think we could double payouts without too much risk of extra gaming. In part because of mitigation strategies such as trust level-based minting, a feature SourceCred built for Maker, and the 10 DAI minimum to receive on-chain payouts. It is also just very time-consuming for new accounts to begin accruing enough Cred to collude effectively, due to strong moderation on the forum and the high barrier to entry to meaningfully participate in Maker governance :sweat_smile:. This could be mitigated further if only certain high-context contributions (e.g. governance actions) are paid more.

One thing I have seen is an uptick in brand new accounts opting in to SourceCred with no activity or likes. Some of which never post. My guess is they’re fishing for airdrops. However, they never receive on-chain payouts, and those ETH addresses aren’t currently stored anywhere in the instance. Keeping an eye on it.

Optimal pay

This one is tricky. The Maker community is very diverse. For some the payouts are pocket change, not enough to change behavior. For others, it’s a meaningful, potentially life-changing amount. People also assign different meaning to the payouts.

I am a little surprised at people thinking they’re overpaid.

I have specifically thought @ultraschuppi and @AstronautThis were if anything underpaid :confused: Ultrachuppi because they do governance tasks we want to incentivize generally (MIPS, Signal Requests, etc.). AstronautThis because they’re a newcomer that has been consistently racking up Cred (recruitment is a goal of SourceCred). I think they may be undervaluing their contributions. I do however think a danger of doubling payouts would be to increase the number of people that think they are overpaid. I fear the system loses legitimacy if it is perceived to be way overpaying (which it may well be in some cases).

My personal feel is that the amounts are about right for most contributions. However, I think there are contributions and contributors that are underpaid, in which case an increase in payouts could be a useful experiment. SourceCred can be configured to selectively value certain types of contributions higher. Which contributions will depend on what SoruceCred is good at incentivizing (and what it is not), and what the community thinks SourceCred should incentivize more (or less).

Headed to sleep here, but returning tomorrow to address ideas around incentives and some valid arguments against an increase.

5 Likes

I think SourceCred is great, as I commented several times.
I express some personal wishes for the future (I know you have been working on some of these, but if the bond between Maker and Sourcecred is increasing, I think it’s ok to ask):

  1. it would be awesome to have more frequent and automatic payouts.
  2. In a world moving to L2/sidechains, it would be nice to be able to receive the payouts in xDAI/arbitrium. In such a space, even daily rewards would make sense. Or even “streaming” of rewards (dreaming).
  3. Allow to make “Cred tips”. Often I see amazing posts of others, and I’d like to tip them somehow. I always push the like button, but sometimes I wish I could give more! :heart_eyes:
4 Likes

This is exactly what I have been doing with my coworkers to try and get them into crypto/DeFi. SourceCred should be the gateway portal for the future of finance via MakerDAO governance. It is a huge learning curve when you do not even know how the current financial system works. So, I think it is essential to incentivize new users to spend the time to learn the ropes. And after some time getting comfortable with the protocol I am sure they will want to own MKR which is a net positive. SourceCred is not a money-losing scheme, but an investment in the future of this protocol and DeFi more generally.

3 Likes

I couldn’t agree more. I’m arguing that the current model is promoting the exact opposite by being a zero-sum game.

3 Likes