MIP40c3-SP7: Modify Protocol Engineering Core Unit Budget

say “significant” = half.

Yes, as an example, experienced members of the SC should be paid more then twice as much as a member of the communications team.

However, there is a reason top companies standardize pay to a role and each employee is paid based on the role they are assigned.

Unless an individual is managing a large team of developers they are not worth a 2M+ package. Creating an environment where every SC developper we onboard gets 2M+ is not sustainable.

A few unanswered questions:
Is each member of the SC team equal?
What will be the starting pay for new SC members?
Should the team facilitator be paid equal to each team member?

1 Like

The purpose of a stock option is to align investor goals with employees. It is not designed to be guaranteed money. Maker will be more volatile over the next 4 years then facebook or Google stock however it also has the potential for much greater multiplicative returns.

I would also add the proposal is to mint new maker for the SC team at no cost. This is not the same as a stock option

A stock option would imply we instead give the smart contract team the right to buy maker at a set price and this option only has value if the price of maker goes up.

4 Likes

Hi all,

Three areas of comment as we put this MIP forward for Formal Submission:

  • Looking back over the previous month has seen a great deal of discussion on the pros & cons of this proposal. However, we have not landed on an agreement for the MKR reward, as a result the team has decided to put forward our original proposal for on-chain MKR voters to decide.

  • Due to ongoing discussion of the compensation framework, we reserve the right to switch from our MKR vesting plan to the MKR framework that the community has begun discussing, but has yet to agree on. We hope that a suitable overarching vesting framework will be chosen and backdated to the commencement date of the core unit(s). Therefore, we are submitting our proposal as it exists above, with the option, not the obligation, of switching to a community determined MKR compensation framework if/once it is agreed upon.

  • As per the update to MIP39c2-SP#7, specifically: “community discussion during the RFC process indicated that we should also give immediate attention to laying the groundwork to promote future protocol growth and innovation, in particular L2 development”. I have increased the number of team members by two, included in the budgeted costs.

6 Likes

Someone listen to this guy, he is the only person on this thread being reasonable…

2 Likes

Some things here:

  1. When will you be able to provide a multi-sig address?
  2. Will the initial distributions be manual (until we get the keg?)
  3. When do you expect to receive the first batch of funds?

Feels like the budget is actually more like 6.91M given that over the year you’ll be receiving the lump sum + 11 months of 510k. I’m aware you’re not spending all of that in the year, but it is something you need to operate, and governance doesn’t have access to those funds anymore (as I understand.)

If you can, please make this more clear, I’m fine with you editing the MIP at this stage if the goal is to increase clarity.

1 Like

@LongForWisdom , thank you, I have updated the MIP section “Smart Contract Implementation” to include updated multisig addresses as requested.

Also, a further clarification regarding timing: if the April 26th executive passes, the Core Unit would officially work for the DAO from May 3rd. For this to occur, we would seek forward payment of the first month, May ($510,000) to go into the Protocol Core Unit Multisig, and the continuous employment lumpsum ($1.3m) to go into the Continuous Employment Fund Multisig at the passing of the MIP bundle executive.

2 Likes

Done!! Can somebody please forward this Gentleman the DAI requested above. TY in advance.

May is looking like a Month for celebration! :dancer:t4: :man_dancing:t4:

3 Likes

Actually this begs the question of how much MKR IS left in the treasury. While I want to retain talent 1K MKR over 3yrs is pretty much retire money. Given the heat in the space probably reasonable but I would honestly expect some pretty high team turnover rates within these 4 year periods because of this - so plan accordingly by staggering quality hires.

Hence my next question. If turn-over is going to be high how is the team going to find and retain the next people? Also what sort of non-compete clauses are there (any?). I mean I don’t want a Maker smart contract dev just up and going somewhere else to fork the protocol or to compete with it.

It is going to be interesting to total up these Core unit costs and look at the current revenue rate because right now I am seeing some pretty big numbers being put up and I am only on the the second core team.

Begs a question who is integrating the financials on all these core units into a kind of business expense run report based on the asks here?

My other issue is how in the heck we are going to get any decentralized backup redundancy in these groups with these kinds of budgets? Is it even possible?

RWF is working to get this done. You can see the first report here. And @Aes is working on an updated version. Should have more detail in the next monthly financial presentation.

I keep a spreadsheet of net revenues based on vault SF, DCs. Last time I ran this came up at 115M/yr being the current revenue rate P/R 19.2, 57M/yr expense doubles this as a P/E, 83M quadruples it.

We talk about having sufficient funds for operations and whining about having a surplus in the 60-100M and not burning MKR to achieve this as fast as possible when expense run rates looks to be in the 50-80M/yr range??! I mean sure we can always print MKR, but we really need to be thinking about a secondary reserve as I can see pulling these funds from surplus during bad times could easily start minting MKR.

I am back to how much MKR does the foundation have left for these kinds of expenses that will be folded back into the DAO proper as a kind of ‘rainy day fund’. Remember Maker has SF’s pretty much at 0 for an extended period after BT.

I made an analysis (not documented) and keeping such an external buffer in the surplus is better in all cases. Having an external buffer makes sense only if you want to convert some DAI in USDC in case of a big issue with DAI. DAI holders will not like it.

But you are right that expenses (in DAI or MKR) are decreasing MKR profitability and limit our ability to increase the Surplus Buffer in line with growth (I discussed that here). That’s why I advocate that MKR holders limit expenses to a fixed amount (that can be revised as much as we want over time).

PS: Maybe it is better to continue that discussion in this thread. A Framework for Valuing MKR

3 Likes

@iammeeoh It’s an interesting aspect you’re bringing here to the discussion. Personally, I am not quite convinced this could be applied as a blanket measure across all types of roles within the MakerDAO Core Units.

It’s quite difficult seeing Software Engineering type of activities fitting into that box. It’s quite a collaborative and peer-review heavy approach comparing to other functions. Think of pair-programming for example.

For Risk teams, I personally agree that approach can be valid, particularly when it involves modelling parameters. There is relatively little review of existing models by independent teams to allow the research-like activities that you find in modelling teams even within banks and other economics institutions. That leads to potentially not ideal outcomes and a lot of trust given to Risk teams without checking frameworks. While this was OKish when the protocol had just a feel million staked, lack of model documentation and reproducibility by other teams is one of the biggest risks to MakerDAO as the protocol grows. Critical smart contract changes have audits. Critical Risk changes in a multi-billion dollar ecosystem might as well have some… Liq 2.0 was a first :slight_smile:

2 Likes

I have to agree here! Tenure is a really bad set of incentives that encourage rent-seeking. It’s been studied to death in the economics literature and academics with tenure on average publish less, in less influential journals, teach less, and their students report lower satisfaction from their teaching.

1 Like

I totally agree.

Perhaps it sounded like I was proposing tenure for the SC team. I was not.
Just to be clear, I was just expanding on the Reddit quote from rune:

I also think a good place to look for compensation standards would be research universities and how research positions at such institutions are compensated.

I agree that tenure is a very expensive instrument. It is most definitely not what is needed in a startup/experiment/young-DAO like MakerDao. At least not at this stage.

It has also been studied that researchers are more likely to commit fraud, publish fake results, target h-index pumps rather than honest research, etc, if they are under the pressure of “renew the contract” or “apply for the next job”.

From wikipedia:

Tenure is a means of defending the principle of academic freedom, which holds that it is beneficial for society in the long run if scholars are free to hold and examine a variety of views.

Anyway, MakerDAO is not academia.

So I 100% agree with you guys that the tenure system is not appropriate for what we are doing here now.

1 Like

I actually think there areas of MakerDAO that would benefit greatly from that intellectual freedom and inquisitive peer review. For me the low hanging fruit is risk modelling.

2 Likes

I kind of agree with this.

But in my opinion (as a tenured academic myself), the best approach is not to replicate academia “outside” academia, but to interact with academia.

For instance,

  1. fund PhD/PostDoc scholarships on the topics that we care about (at the ‘fundamental research’ level).
  2. Sponsor academic conferences, run workshops, etc
  3. Try to create incentives/rewards for our (research-oriented) Core Units to publish in good conferences/journals (to get peer-review external feedbacks), etc. → A good example is RWF in my opinion.

As far as I know the CMU has a substantial amount of MKR, donated to them by one of MakerDAO founders: MakerDAO co-creator donates $1.4 million in MKR to Carnegie Mellon | Modern Consensus.

That would be a good (top level) institution to start interacting with academia. :slight_smile:

3 Likes

As part of our broader effort to bring more transparency to the CU budget structure, we have documented the wallet setup of this CU and others;

Read more about it here: Introducing the CU Budget Transparency Map

1 Like

Thank you Petru_Catana!

@Derek I understand you are busy however at some point could you provide an update on PE unused funds?

Based on makerburn.com on 14 PE members are currently vesting so I guess we should expect unused funds to be returned at some point. Right?

Hi Zarevok, Yes! I have and intend to provide per-monthly categorised expenses from the PE team, but would like to provide descriptive context first around some of our audit, software and tooling costs so it makes sense to the community. This will include both surplus (salaries) and deficit amounts (audits and gas deployment costs).

Keep at me about this, early December should be a quieter time for me to complete this task and provide an insightful report.

4 Likes