KPI distribution - DFX Community members (1/2)

DFX Proposals
21

I completely disagree with this statement.

End of day you are also assigning weights based on the KPI criterion. Number of addresses that qualify for a criterion is not dependent on the weight you assign to a category. Just because fewer/higher addresses qualify for a category should have no impact on weight assigned to a category.

We (as a community) should be open to discuss weights that benefit end KPI goals.

As an example,
Tanman (me) suggested weights of 35% for engagement + 15% for LM/Holding
CaptainM suggested opposite - 35% for LM/Holding + 35% for engagement

These are valid conversations to have based on what works as an end KPI goal.

But trying to finalize weights (or points or %) based on how many addresses qualify against an activity sounds flawed.

22

A couple points I guess.

Firstly, I’m not sure what led you to believe that I’m opposed to KPI goals that benefit the community nor the suggestion that I somehow implied other comments were not valid
 Quite the opposite. I value any and all discussion and despite proposing some alternative viewpoints, do this only to spur ideas and ensure we as a community have considered whether our proposed allocations make sense. I was merely offering an alternate opinion and happy for it to be agreed upon, modified, or rejected by community consensus. Let us please keep the discourse focused on facts and opinions so as to avoid breaking down what I feel to date have been productive and interesting discussions.

My concern regarding broad categorical percentages is in fact specifically with concerns that it risks missing the point of the KPI goals as there’s the potential to both over- or under- incentivize.

Hypothetical example:
Pretend community consensus is that voting and LP provision are considered equally important. And let’s pretend 40 individuals voted while 80 people provided liquidity. If each category were given equal percentage weights despite different numbers of participants, the voters would receive double the KPI allocation which to me doesn’t make sense. If you further take it to extremes what if participants were 40 vs 160? 40 vs 320?

What I was imperfectly trying to propose is a rough framework that values specific community contributions against other contributions, a concept that will admittedly be easier to frame once we know the total community KPI options to be assigned as well as the number of participants in each category.

Others may disagree and despite my current viewpoint, I welcome those opinions. I look forward to thoughts from other community members to help move the discussions forwards.

23

I think discord level 2 should be enough. Because the number of people with level 3 is less than 25.

24

Can anyone clarify more specifically how the discord level metric is calculated? Specifically what degree of participation or comments is required to reach level 2 vs level 3 for instance?

25

I don’t know exactly either. When I looked at mee6 bot’s site, I read that the difficulty ratio can be changed.
I want to give an example from myself, I searched all my message histories. I sent about 25 messages, I’m level2 53xp. My Rank #42

My guess is that the number of people with level2 and above is less than 100

Note: the person who enters the discord starts at level 0 at first.

26

Hey @thecryptoforager - Nothing personal, purely a mathematical disagreement on what you are proposing.

I am stating weights should be based on end KPI objectives, without a consideration of how many addresses qualify for it. You are also in alignment, but seem to suggest weights should be fine-tuned based on number of qualifying addresses
which is where the disagreement is.

Key assumption: Weights for engagement (voted, posted in forum/discord) > just actively LP/holding tokens.

In an extreme example - say only 1 address is engaged (voted, posted on forum/ discord) and 1000 addresses are in LP pool. Given weight of engagement is higher and there is only 1 individual in pool - all KPI options will go to that 1 address. There is nothing wrong with this

This is the behavior we want to model in future. This is the pool of individuals who will likely help us achieve our goals (with additional incentives aligned to their work). We know this pool (1 vs 1000) is less now, and that’s reason we are having KPI incentives to increase it in first place. Just because our desired pool is less, we should not reduce weight for that category

In summary: I am proposing engagement should be weighted higher than active LP/holding addresses
knowing fully well that the higher weighted bucket has lower addresses, rewarding these engaged users much higher than non engaged LPs (who likely do not have time commitment to vote and be active on community forum/discord). To each his own


P.S:I will try to limit actively posting for a few days, so we can capture other viewpoints as well

2 Likes
27

Fair to say we can agree to disagree.

The concept I’m suggesting is based on the UMA KPI airdrop UMA Airdrop » Claim free UMA tokens

Look forward to opinions from others.

28

So looks from Discord looks like each message sent earns between 11-18 points (random basis).

Level 2 requires roughly 14-23 messages
Level 3 requires roughly 26-43 messages

Maybe consider Level 2 (not sure exactly how many this equates to
?) vs Top 50 Discord participants for inclusion?

29

I been reading all the great ideas and appreciate the passion and great discussions that are going on here! You have an awesome community. Just wanted to jump in and add something that might be helpful.

You can hold some KPI options to reward those who go above and beyond,new community members that begin contributing to raise your TVL and show promise, and
bounties. Maybe, new LP addresses: “All new LP’s will receive our KPI options”.
Im just thinking out loud on the last idea.
Something to build TVL, gain new LPs and something to secure new contributing community members.
If you don’t use them you will get your collateral back. You wont be losing or wasting anything.

UMA uses tip.cc bot to tip others who contribute or do something cool.
You will see in our Discord "xxxxxx tipped xxxxx 5 uDAO " as you scroll through.
You will also see members tipping other members their own uDAO to say “Thank You” or “I Appreciate You” in our community.

When your working on a project to raise TVL, you get an alert, than you see in the chat that you were tipped uDAO. It’s uplifting, shows your contributions are not going unnoticed and you are appreciated for what you do.

3 Likes
30

@Mikhail said:

Is it fair to say that the KPI option program should be designed to incentivize B (work) to attract A (capital)?

I think that’s the right way of looking at it. Since TVL is a health metric of DFX in general, you could see it as a proxy for effectiveness across many teams, including social media, content, admin, and biz dev. And one hopes to attract people who can offer skills in any of these areas.

2 Likes
31

I think this is the right approach overall-- With the idea being that you’d issue subsequent KPI options based on participation in the program itself. This is how UMA has done it. I’m not saying this is the only or even best way, since not enough trials have been run, but it was effective for us and I think a reasonable model to re-test.

So the way that would look here is you issue these out to the community and people who have helped DFX already – which seems somewhat necessary if you’re also going to throw out a wider net to others. The goal is that you’ll:

  1. At least not displease existing DFX supporters
  2. At most attract a portion of them to become active supporters when they were previously passive.
  3. Throw out a net with the other 50% of KPI options and hope to catch good talent. (This is why #1 is important, so existing supporters don’t experience it as dilution.)

Over the period of this first issuance, you create your core. Over this period you set up your core community and the group around which the future incoming members will cling and snowball.

Future issuances is very much a developing science. UMA has mostly done issuance to existing members + having a recruitment / onboarding process. We have grown only gradually, but powerfully, in this way. You could do the same reliably but might want to experiment with more creative approachs.

One possible way is that if DFX continues to do liquidity mining, you might consider paying out in KPI options. This would both attract liquidity and create a passive / always-on attractor to bring people into the community effort. At least that’s the theory - It’s possible to imagine people dumping them all the same but one still hopes some will get with the program!

4 Likes
32

I agree with you on the logic that if only 1 address engaged, and 1000 LP’d, that this 1 address should get all the rewards.
Generally I think the distribution should be as easy/clear as possible.

You could even assign 50% of options to LP and 50% to engagement and make it so that you either qualify for it or not.

Engagement KPI Option:

  • level 2+ discord
  • participated in at least 2 votes
  • DFX Token Holder
    (at least 2 of the 3 requirements need to be fulfilled, but everyone thats eligible, is eligible for the same amount)

LP KPI Opiton:
(not sure about requirements, but structure it in the same way, that 2 out of 3 requirements fulfilled makes you eligible)

Those are just examples and maybe use other requirements, or change them a bit. And maybe 2/4 or whatever makes you eligible.

1 Like
33

I think what’s presented in the first post is fair enough(both rates and requirements). I must say that I do not agree with those who think that there should be more filters. Because the number of active users is not that large. For example, if there were 50000 users, I would think that there should be more filters. But as it stands, I don’t think it’s necessary to apply any more of them. It should be noted that this is an incentive program. I think it would be more beneficial to appeal to more people.

34

Great to have more engagement in the forum. Few follow-up thoughts based on further discussions above.

While I do agree we don’t want to overcomplicate the criteria, I do think it makes sense to have select criteria included that help prevent gamification of incentives given how prominent this practice is in crypto.

On a similar note I agree that we want to appeal to more people, but with the strategic caveat of aligned incentives.

With so many factors to consider for KPI options and great discussions thus far, I wonder if it’s worth taking a step back to see if we can form general consensus on the basic categories to be included or excluded? Further discussions re: how best to allocate to each category and/or any specifics within each category (I.e. timeframe for LPs, # of voting sessions for voters, etc) could follow.

Proposed categories to date (please correct if I’m missing any):

  1. Voters
  2. Liquidity providers (stablecoins or DFX-ETH)
  3. Discord contributors
  4. DFX token holders
  5. UMA KPI call attendance
  6. Forum contributors

Additional suggestions surrounding KPIs:
7. Holding KPIs to reward certain participants (I.e. people going above and beyond, new LPs, etc.)
8. Liquidity mining rewards as KPI options
9. Concept of grouping engagement (voting, discord, holders) into a single category and liquidity provision as another category whereby you need 2/3 or some other agreed upon metric to fulfil the category requirement for inclusion.

I haven’t seen any objections to #1, 2, 3, & 4.

I think rationale for excluding #5 (UMA KPI call) has seen support due to gamification concerns.

Curious re: further thoughts on #6, 7, 8, & 9.

Re: #6 I think forum participants are worth capturing as it’s a valuable contribution. If people agree, should this be own category vs lumping within say voting category or discord category?

Re: #7 While I feel withholding a portion of KPIs could be valuable as a tool to encourage new/ more active participants, I wonder if not necessary as expectation of possible future KPIs based on categories for this round could theoretically achieve similar objective.

Re: #8 While I like this in concept, I wonder if it would discourage participation by some LPs (and I don’t just mean those farming to sell) and result in less overall liquidity in pools


Re: #9 I’m personally in support of more of the individual categories.

If anything missing or seen differently, please chime in! Look forward to everyone’s thoughts as always.

2 Likes
35

Thanks for the summary of categories. Personally, I think we should limit the scope to just #1, 2, 3, 4 in your list there. Good to keep things as simple as possible for our first trial of KPI Options.

There are only a small number of contributors to the forum so far and having a whole category that is pretty much a subset of Discord seems unnecessary. If our forum was more mature with a lot of varied discussions, I’d perhaps feel differently.

I don’t have any comments for the other items here as I think they just complicate things. I see this initial KPI Option push to be more of an MVP just to see how it’ll go. We can always do more rounds later.

1 Like
36

I read through all forum posts including some excellent discord feedback by superumans (EAsports, Sean and others). I refined the plan to keep it simple and also be broadly inclusive of our entire community for wave 1 of this KPI distribution scheme

Here is my final proposal

50% to internal community

  • Voters (20%)
    • 15% to voters who voted in either DIP-1 or DIP-2 or DIP-3
    • 5% to voters who voted in all 3 (DIP-1 + DIP-2 + DIP-3) as an additional bonus
  • Discord activity (10%)
    • 7.5% to discord rank 2 (and above)
    • 2.5% to discord rank 3 (and above) as an additional bonus
  • Liquidity providers (15%)
    • 10% to all LP providers (stablecoin or DFX/ETH) providing liquidity for 2+ months
    • 5% to LP holders providing liquidity for 2+ months + held 75% of farmed tokens as a bonus
  • DFX token holders (5%) (in wallet and also unclaimed), holding >= 20 DFX

This is super inclusive of nearly everyone within community and also recognizes those who went above and beyond with bonus allotments.

As a reminder, any actions performed on/after Sep 17, 9 PM UTC will not be included here to prevent gaming.

6 Likes
37

While I acknowledge the small number of participants on the forum I do see that as part of the issue we’re trying to address. In principle we are trying to use the forum as a place to hash out governance proposals/ discussions prior to voting, and as such varied and active participation is valuable as this shapes what is ultimately voted upon. I’d argue to date the forum has not yet achieved the goal of varied discourse likely in part as it’s an extra step compared to discord and people have little incentive to participate. Without people actually sharing their views on the forum we end up in situations like the proposal for adjusting liquidity mining rewards where clearly a significant part of the community is not aligned based on opposing votes, and yet those opinions were never expressed prior to going for a vote, which would have been helpful in case modifications could have been made prior. While I don’t feel strongly about rewarding past participation as that’s not the ultimate goal of the KPI options anyways, I do see value in incentivizing future participation.

1 Like
38

Aside from that I think the remainder of the proposal @tanman has re-outlined is reasonable in terms of allocations. I might just suggest clarifying for LPs that it’s 2+ months providing liquidity (as opposed to holding).

I do also think it would be helpful to get a sense form the core team how many KPIs/ associated value per KPI you’re thinking would be reasonable. That would again help look at the allocation percentages and ensure based on the number of participants that the KPIs are of sufficient value based on the recommended distribution to actually be motivating and thus achieve the actual goal.

Also number of people captured in each category / subcategory.

39

I agree @thecryptoforager. Forum posts are a valuable input to this community
but unfortunately they can be gamed too easily. Ex: A user can create 10 accounts and post in forum to get a significant pool of current/future KPI tokens.

So here’s the best way to ensure community members share their views on the forum and also not abuse it

Forum posts can be used as a tie-breaker to bump a user’s discord rank to next higher bucket. So as an example - If cutoff for a KPI reward is Discord Rank 3, but there is a Rank 2 member on the border who narrowly missed out (by a few ranks only
and not someone who is at bottom of rank 2). They can request a core team member to manually review his forum contributions and bump up his address similar to a Rank 3 user.

Given nature of snapshot, no user can be confident his rank will fall cleanly in a higher bucket. So there is incentive to post on forums to act as a tiebreaker

Note that the contributions in forum post should be substantial content, and not a random “I agree with this view” type post

I suggest we try this approach in KPI wave 2.

3 Likes
40

Thanks for the flag. I edited my post to fix the typo