Pre-Hatch Impact Hours Distribution Analysis

Hello TEC community,

I have been playing with the Hatch dashboard and analyzing the results from different parameterizations. I noticed in these experiments that the distribution of Impact Hours seems more skewed than I would have expected, and it made me realize that we had likely never questioned the outputs of the data from the praise process.

Although I recognize that people receiving pay from the organization only received 15% of the tokens relative to unpaid contributors, the data seems to suggest that some forms of work in the TEC were more recognized by the praise process than others, and I thought this fact deserved a conversation involving the community.

Looking at the praise data, it seems that a lot of small praises accumulate significantly more tokens than larger tasks with less frequent praise, even if they are quantified relatively much higher. This begs the question: do the results we see land within what we would call a fair or accurate distribution of tokens for work that was put into the efforts of the TEC?

Considering that significant governance in the TEC will be held according to these IH distributions (or funds available via RageQuitting), I believe this poses considerations for community discussion prior to the Hatch.

I don’t mean this as an offense to anyone who received more tokens than anyone else - in fact, I think Rawl’s Veil of Ignorance should be invoked here to some degree, and we separate the desire to maximize for any individual’s best outcome to focus on ensuring outcomes that are fairest to all.

We have established the Praise process to track and quantify contributions, but afaik we haven’t taken a step back to do any quality assurance on whether those processes are outputting signals that we are comfortable with as a community. When using new processes like the praise system to quantify contributions to an ecosystem, we are measuring with Wittgenstein’s Ruler. In other words, we first need to measure the ruler itself to know if it’s good at measuring anything else.

I have put some overall thoughts on improving the praise process into this document, and I believe we should do some preliminary data analysis of the distribution of Impact Hours to learn more, as well as discuss as a community whether there is wider consensus as to whether this is an issue, and how we can address and mitigate it.

A few items to consider in this proposal:

  • Engage the skills of a few data scientists to understand the distribution of IH tokens more thoroughly & whether there are alternative weightings of the data that feel more accurate to the community
  • Discuss as a community what a “fair” or “more accurate” distribution of tokens looks & feels like, understanding that this may be fuzzy & vary widely by perspective
  • If determined necessary, consider various ways to adjust the distribution to be more accurate to how we feel it should be distributed
  • Create a new mandate for praise-givers - dish widely in the TE ecosystem. We want to get IH & TEC tokens to Token Engineers too! Learn what’s going on and help the system recognize it’s value. Many TEs are too busy doing the work to be present in TEC meetings dishing praise for that work.

I hope these thoughts and recommendations can inspire community conversation around this important topic - comments, disagreement, and further discussion welcome!

Big love,


Background and Context

Impact Hours are sort of a "Team Distribution"

Usually in crypto projects this is 5-50% of the total supply that’s distribution is decided by a benevolent dictator (or small group) who, obscurely, gives tokens to an unknown group of people at an unknown ratio’s for unknown work.

In the TEC, we do it differently. We follow a decentralized, Signal Aggregation approach, that is accessible for all to Participate.

  1. Anyone who wants to can praise anyone for anything.

  2. Anyone who wants to can join the praise quantification calls every 2 weeks and participate in the process, after quantification, there is a sense making round to make sure that the results made sense, and it can take into the overall distribution as well. This is how the sausage is made, quantifying Praise into IH.

  3. Anyone who wants to can propose the design for determining the Impact Hour Rate (the rate that IH gets converted to Hatch tokens) right now. The impact hour rate can even be set to be $1 at the Target Goal, and IH could be meaning less.

  4. Anyone who has engaged enough to have CSTK or IH (well over 300 people, and another group of about 200 people that just need to give us their address so we can give them the on-chain power) will be able to vote on how Impact Hours are converted into Hatch Tokens.

#2 Is where there is the measurement of the ruler, this sense making work every week allows us to do a check on the distribution.

There are of course more details here:

Original Intents

While distributing governance rights fairly is part of the intent of impact hours, it is only one of the dimensions that we are using this tool for.

  1. Creating a lovey-dovey gratitude culture and strengthening the relational fabric of our community
  2. Rewarding volunteers for contributions to the creation of this economy
  3. Acknowledging general contributions to the TE Space
  4. Creating a feedback mechanism for work that is valued by community members
  5. Distributing Governance Power to people with skin in the game.

My Thoughts

I think “fairness” in a token distribution is a noble goal but in application it has been very difficult to accomplish both an objective distribution and a distribution that is good for the community.

I don’t think any community can say that their initial distribution was perfect at achieving their goals… Some were absolutely better than other tho. YFI and other defi tokens that had a “fair launch” for instance have had a lot of criticism around this approach.

Vitalik recently discussed this issue in respect to the fairness of Miner rewards. Mining is a very fair process but the requirement of objectiveness leads to an awkward expenditure of capital.

Every individual has a perspective on what fair is. To ETH miners, the mining reward distribution is very fair. To Vitalik… it’s not.

What is really more important than if it is “fair” by any individual’s perspectives (Mining for instance achieves this) is the question of legitimacy.

Is this a legitimate distribution. Here is what Vitalik said:

It is impossible to make a full listing of theories of legitimacy, but we can start with a few:

  • Legitimacy by brute force: someone convinces everyone that they are powerful enough to impose their will and resisting them will be very hard. This drives most people to submit because each person expects that everyone else will be too scared to resist as well.
  • Legitimacy by continuity: if something was legitimate at time T, it is by default legitimate at time T+1.
  • Legitimacy by fairness: something can become legitimate because it satisfies an intuitive notion of fairness. See also: my post on credible neutrality, though note that this is not the only kind of fairness.
  • Legitimacy by process: if a process is legitimate, the outputs of that process gain legitimacy (eg. laws passed by democracies are sometimes described in this way).
  • Legitimacy by performance: if the outputs of a process lead to results that satisfy people, then that process can gain legitimacy (eg. successful dictatorships are sometimes described in this way).
  • Legitimacy by participation: if people participate in choosing an outcome, they are more likely to consider it legitimate. This is similar to fairness, but not quite: it rests on a psychological desire to be consistent with your previous actions.

Obv a lot of these forms of Legitimacy are pretty bad… hahaha but our IH distribution has evolved into a very legitimate practice mostly via participation, process, and continuity.

I’m very excited to improve the process for the TEC post-Hatch and future Commons, but I would be skeptical of changing it too much at this stage as it could threaten the legitimacy of the process.

I definitely like the suggestion that we dish praise widely. It was easier when we were in the TE Discord, but I would love to see more praise given to TE people in general.

Problem: Most top tier token engineers are very busy right now in the bull market & they can only get praise if they join a channel that the praise bot is in, and someone recognizes them for their work…

This barrier of entry is there for good reason. If they are too busy to join the Discord, then they probably aren’t going to participate in the Commons either. Thats ok, the Hatch is just the first phase and they can join later when they can be more free to participate.

Or they buy their way in.

It is important, for our legitimacy, that every participant has hatch tokens because of the skin in they put in the game. The Impact Hour Rate is unknown right now. Most proposals are giving 70-85% of the tokens to the people who put in wxDai. The Impact Hour distribution is only relevant for the 15-30% of Hatch Tokens, and will be further diluted once the we upgrade to a Commons.

Once we launch, the broader TE Community will naturally be more interested in our work and we can always create solutions for improving a poor distribution IF it happens… e.g. a proposal can be made to mint Voting-only governance tokens and airdrop it to thought leaders as needed… or any other outside the box idea to address the issues we run into.



A praise case study was done, and I am hopeful that we can make praise quantification into an app of some sort, ideally tied in with Source Cred, and we can make a lot of important improvements for the next iterations of this tool.

I hope that we don’t do anything to challenge the legitimacy of the distribution by manipulating the results outside of the normal process. Dish praise to people and then we quantify, the final impact hour rounds are yet to be done.

1 Like

Hello guys ! :smiley:

I’ve been on the community since september, i probably join almost all the events at TEC held since then. And in my honest opinion the distribution of impact hours before removing the IH to the people who is getting paid it’s precius how accurate it looks.

In my opinion, we need to look TEC as it’s own thing its not CS neither 1hive or Giveth, I could see post hatch TEC doing an airdrop to them. I honestly don’t know how CS or Giveth work, but in the case of 1hive, 1hive are all the token holders. And the snapshot could be taken right after the second proposal asking funds, but thats a decision that TEC have to make.

Regarding the small task vs big task i saw this issue pretty long time ago and the problem i was seeing its people who do alot of small tasks like myself was getting in my opinion overpaid when doing praise quant that’s one of the reason i stopped going on the praise quant, it feels very bad getting +1ih for doing a retweet.

That being said, everyone involved enough with the TEC knew about that and right now isn’t the moment to change any rule, we are here how we are because the decision we take as a community just before launching the hatch is not the moment to make any big decision with the risk of breaking the social contract we all made.

I consider myself a builder and I really care about TEC and im super curius abut to see what people do, thats a social experiment after all. But the people who put time in. In my opinion it’s more likely to care about the decisions being made, puting money its just a couple of clicks (its not a bad thing) thats why if somone listen to me the lasts days on the params party’s im always very cautious with the CSTK ratio, but at the end of the day i trust the process and people who have CSTK are cool people that we want in or atleast that’s what i say to myself :stuck_out_tongue:

Looking for fairness, it’s an awesome value, but legitimacy, its way more important in my point of view and that’s something that also came up when we started to rotate the praise quantifiers. Someone said about giving IH back to Griff and Livi because of the hard work they did previously, but we didn’t create IH to them from no where that would go against the legitimacy of the project on a very early stage.

In my opinion, there is no ruler, there is a protocol and everyone in knowing about it. That being said the spread sheet could be more transparent and usable. There were many times I wanted to check it, but didn’t do it because there are many tabs hide and I was fearing messing it up (I really trust the data myself but its important for mutual monitoring).

The good thing it is not the last common we learn from here, I would not break neither the social contract or the protocol. In the worse case scenario we don’t reach our MIN goal and IH are gone and that would be ok its respecting the protocol (that’s why IH have a social value not only economic, people getting paid lose less, that’s very personal but I don’t like this mechanic of people being paid I know it have pros and cons not going to debate it here)

That’s why, if the project its good enough they will put money in and govern it (Let’s be honest the price to get in isn’t that high, we are talking about TE’s after all)


I think that’s an interesting discussion we can learn a lot of things. I’ve not been involved with the community outside the swarm as much as I would have liked to (coding requires focus :sweat_smile:), and I don’t have a big comprehension of what could go wrong with acknowledging contributions. I will give my opinion on what we could do and what we shouldn’t do to improve the system and the valuations.

We have knew since the beginning that the system was not accurate at all, that it is mainly subjective, and “nobody wants to know how sausages are made” as we used to say referring to the process of converting qualitative praise data into quantitative impact hour numbers, but I think something not accurate doesn’t mean that it is not fair.

I consider a system to be fair when its rules are clear, everybody plays with the same rules, and we can verify openly that processes are being followed as it was expected. Even if the final outcome is not the most desirable, it is fair if the process has been open to everyone and have given partial results that people could react to (by changing their own behavior, by introducing changes to the system, or by challenging it). This process has created legitimacy, or as zepti was saying, it has forged a social contract that is not desirable to break.

Acknowledging people we may failed to acknowledge these last days before the hatch could be a good way to go. Although, I would not “print” more impact hours above the natural tendency we already had. We don’t want to be like FED printing trillions in just one year :stuck_out_tongue_winking_eye::

That being said, I think @JeffEmmett suggestions on how to improve the praise system are lit! Understanding the praise data using data science could be really insightful to find its weak points, such as little vs big contributions. And the praise quantification should probably be done easy to do if we want it to be transposed to other daos.


Thank you, @JeffEmmett, for raising this issue and to @sem, @Griff and @ZeptimusQ for responding here. I am still processing these ideas and immediately appreciate three things:

  1. the raising of the issue itself,
  2. the importance of maintaining legitimacy/the social contract,
  3. Jeff’s great ideas for improving the process.

I know there is a lot more to say and share, but I just wanted to throw in my two tokens (did you see what I did there? ;)) and be part of the process! :smiley:


My questions in a dialectic process are always: How can we transform this into the most desired situation?, and what would be the best possible outcome?

Using Ostroms framework for institutional change, and seeing the TEC as a conflict transformational platform we could ask ourselves for the ex ante and the ex post costs of a proposed change and try to use a credibly neutral criteria to guide our action.

“If the expected costs of transforming the rules are higher than the net benefits to be gained, no further cost calculations will be made. Appropriators will retain their status quo rules that produce fewer benefits than would alternative rules, because the costs of changing the rules are higher than the benefits to he obtained. If the ex ante costs of transforming the rules are not too high, expected changes in ex post costs will also be evaluated, including the effects of proposed rules on monitor­ ing and enforcement costs.” Governing the commons, p 198

We don’t want to loose any of the legitimacy we have built from the process as its been done so far, and instead we want to gain more legitimacy from the sense making and feedback. So let’s have an open and safe conversation around this.

Personally i don’t have problems adding an extra filter to the actual IH distribution process, but i would only want the filter to have an overall better legitimacy and feeling than the one we have.

The other day i said this was like cooking, we are making an exotic dish, mixing a lot of ingredients and maybe we now that we try the flavor, it needs some balance before putting it in the plate.

I say this as someone who wouldn’t be here if it wasn’t for the work of some Token Engineering pioneers (that can be higher in the table), but also for the loving culture that derives from !praise allowing community building around contributions, thankfulness and recognition :smiley:

My thinking is that all contributions are valuable. And that we should try to promote the best feelings around the commons that will be born for it to start with right foot!


It is never too late for democracy.


:face_with_monocle: Let’s open up this Praise System algorithm and see what’s inside!:alembic:

I am happy to see this thoughtful and diverse mix of opinions here in critically thinking about the design of our systems and processes, and in analyzing the governance decisions that have led us here, and being open to that being an evolutionary process worthy of time, space and lively discussion and debate.

Thank you Jeff, for raising critical questions about the systems we are engaging in, and Griff, Zepti, Sem, Suga, Juan, for your thoughtful consideration and all of your responses.

The Praise System

No decision is static and these processes are all as we create them together - and they are experiments. These systems have not been tested and validated - but we now have an opportunity to do so. To use Token Engineering to validate our designs! (or at least analyze them)

Because it is in an Excel spreadsheet with formulas, does not make the praise system an objective system that accurately reflects the desired distribution of impact hours and governance that the community wishes to see. Nor is it legitimate, simply because it is a process. The Praise Case study was NOT a QA testing or objective analysis using data science.

We manipulate the system inside the system - so does it matter if we “manipulate” the system inside or out - whether we adjust the scores arbitrarily inside subjectively and without context to the outputs to “pump” scores or use algorithmic adjustments / weighting on the outside - only with data analysis, it’s actually scientific and has context. (Ironically and namely we manipulate the system inside, to raise the score of Token Engineers because we want them to participate and have more governance!)

With data, we can examine the system as a whole, see if it meets the needs and desires of the community, and use algorithmic adjustments to tweak the system. Legitimacy is just perception. What is more legitimate - manipulating data blindly or applying a data science approach to explore the system?

Raw praise input in the parameter configuration dashboard and voting on those parameters to adjust the IH rate will not suffice - we need to consider adjusting the distribution, not just the IH rate - Computer-Aided Praise, like we have Computer-Aided Governance. It does not mean we HAVE to act, but we should explore.

This is straight out of the book “Ethical Algorithms” by Michael Kearns and Aaron Roth - paraphrasing: we should never initialize complex systems with raw data from untested algorithms.

We can’t just say the praise system is legitimate because it’s a system. Nation states run by dictators think their systems are legitimate too. A social contract by definition is “legitimacy of the state over the individual.” Are these the things we want to cite? Do we want the tyranny of the process and our attachment to the process to overrule the ability to consciously respond to the present moment, when we have new information?

The Decision Space & Origins of the Praise System

We made the majority of the decisions and two votes - one last August and one last December - on the Praise system with no data, or knowledge of the process outcomes.

This is part of the problem with our current systems. In the US, we vote every four years for most things. We have so much better signal processing now! We can update our decision making with real-time data!!!

The reason this is surfacing now, is that we have data, and are able to see the effects of the system as an input to the dashboard. NOW, we have data we can use and potentially apply. We can ask ourselves and our community, was this what we collectively wished to see? Should we apply an action or modify the rules?

On an “open process” & democracy

Just because votes were held, does not mean that the system is democratic. Open democracy in civic settings also has communication/notification requirements for major votes, and allows for time to petition. The overhead of participation and attention cost in this community is very high, and is a barrier to open access for voting and petitioning with the burden on the commoner to keep up or stand aside.

Ostrom’s principle #3: Ensure that those affected by the rules can participate in modifying the rules. This is our chance to enact Ostrom’s principle #3. And have a discussion as a community, about modifying the rules.

Let us look at the present moment, the new information and knowledge we have and respond consciously, rather than reacting to a challenge to the status quo and entrenched system.

This is an opportunity to pause, and act consciously before we ask our community to invest thousands of dollars, and assign majority governance over our economy.

To me this is less about token distribution and “fairness” and more about democracy, governance and self evaluation, and the fact that the actions we take here will have a huge ripple affects, as we are at the seed level in creating these templates and blueprints that other communities are looking to. That is why the choice we make now is important! This community is amazing! And has done fantastic work in building governance frameworks.

It not personal, or about feelings or token allocation, it’s about data and advancing Token Engineering. We don’t need to stop, we don’t have to act, we just need to think :thinking: and exercise our democratic muscles as a community :muscle:

I propose that we do basic data analysis with data scientists, and hold a presentation of the findings in a well advertised (at least three days advanced notice) community hall discussion with the wider community.

It’s actually quite simple. All it takes is a few hours of discussion and analysis, and if we decide that the costs to change the rules are higher than the costs of the status quo - as Juan so wonderfully suggested - then we forge on. But at least we will have gotten feedback from Token Engineers (the community we say we are building for), made a choice consciously now to employ a system that we know is flawed having looked at the data and applying a scientific approach, and we will have allowed for true democracy to flourish.


PS - I have been thinking about Universal Basic Praise and would love to hear dialogue around that…


This is a really interesting discussion.

Discussing token distributions and payments is really hard and rarely is done publicly (outside of communities with sourcecred :-D)

Money is a really weird thing, and it has a tendency to be challenging in relationships and I’m really proud of how openly we can talk about all this distribution.


I’m here again :smiley: I don’t believe it myself this is fun and addictive (the politic side). Pls if something sounds rude or very straight forward its not my intention, my main goal its to share my thoughts and keep learning from you guys. For me, it’s very important, we stick together.

It would be awesome to apply data science and see the results, I honestly think they wouldn’t be much different tho (if we don’t modify the data). Humans are really good at using the eye to get the result. For example, a group of friends go to dinner together and at the time of payment we distribute it equally. Then a member complains that I have consumed less. In two minutes the accounts come out, this person has consumed more and add 10 euros and these two who consumed less pay 5 euros less each. It’s not 100% accurate but it’s very accurate and the cost of doing it 100% accurate takes more effort.

With this little exemple i’m not trying to say we don’t have to be accurate in the future. But the results it self should be pretty accurate. Do a new distribution at the last minute might not be the best thing. How will contributors act after a change? I think this would create more conflict than solution even if the change is little. I’m really against changing past decisions. We are on blockchain after all.

Another argument I have against it’s really not the moment, it’s go time. We had plenty of time before to check and update our rules. Specially on december i was really not happy how we approach praise (at that moment i wasn’t good enough to write like i’m doing now so thanks you so much TEC :yellow_heart:, still plenty of room to grow and that excites me)

I really agree with sem (i would change fair for legit) but would be nice more engagement from the community, specially stewards

Yeah that’s so true it’s not something I personally agree with. If someone wants to join our common and contribute to advance TE field , anyone is welcome specially TEs. That being said I would love to hear what would be the redistribution we are talking about and how that would benefit TEC.

I’m so sorry about the misunderstanding, I was not referring to what Jean-Jacques Rousseau wrote in 1762 there is no state here. What I really mean when using social contract it’s that there are rules, boundaries and agreements in TEC that we shouldn’t break for the well-being of the common. TEC cares about the individuals (TEC needs them) but the main goal it’s advancint TE. That’s how I see it. In the long run having a TE public goods will help the individuals.

The onboarding document (i can’t find the english version) was talking about IH around 20-200. The human eye :p. Every blog post that happened every 2 weeks I was understanding that all these people were adding this money to the hatch (20-200$) in the case we succeed.

That’s why talking about redistribution makes me untrust the process.

I honestly agree with you here, I’m not sure how well the process will end. It was very open and it’s scary to see the future but also exciting. I feel like trusting our governance culture but that’s my personal feeling.

Democratic doesn’t mean it’s good, Hitler took power via democracy. I really love the blockchain space culture “skin in the game”.

One thing it’s modify the rules for a better being of the community looking to the future with the consensus of this one. But if I don’t understand it wrong, what it’s being suggested here is to change the actual distribution which means change the past and remove value from some individuals to add it to another. How are we going to handle that?.

That being said, TEC should care more for the TE field than for individuals. If there is a strong argument that would benefit the TE field in the long run we should consider it, so please share. And if yes, does it need to be done asap? why?

I really think that what you are suggesting must be done. It adds value to the TE field 100%. But I’m personally really concerned with the topics mentioned above.

1 Like

I hope you all are having a great weekend, and that you can join tomorrow to improve the transparency of the Praise System:


NEW TIME: The data science session will begin at 4pm EST / 10pm CET - Apologize for any inconvenience, and if you cannot attend, we will have a follow up session scheduled farther in advance :pray:t3: Look forward to seeing you there!

TEC Labs

1 Like

Praise System Research & Analysis

:pray: A huge THANK YOU to the community for engaging in the kick off meeting to research and analyze the Praise System!

We will be holding a focused data science session this week to employ the help of our subject matter experts in analyzing the data and posting follow ups as things progress.

In the meantime, see below and look forward to sharing updates as the work progresses:

:movie_camera:Meeting recording:

:books:Notes and document library:


Here is the Hack MD where we sketched all of the ideas for what data slices to look at

We are working with Angela Kreitenweis, founder of TE Community & Academy. She called this an “admirable action” and feels deep respect for the community to be willing to question these systems and processes, and is excited to join the community on this token engineering initiative and a wonderful educational opportunity. She gives her full support and is working very hard to coordinate with data scientists for a live coding session to begin digging into this work as soon as possible.

We are putting together a hack session with a group (3-5 + data science Masters/PhD students) of the top data scientists in TE community next Tuesday and ahead of that, will have a focused working session in TEC Labs on Sunday (TBA).

Will post here and on Twitter events as they are scheduled.

1 Like

Thank you so much for coordinating this effort Jess, let us know if any session will happen today, mainly I want to learn and just watch during this sessions, and help however is possible.

1 Like

Hey Jess, you are telling TE they will have more IH tokens? They will feel disappointed when that doesn’t happen. Our reputation will decrease. Even if TE don’t have as much governance of the commons, they still will receive funding pool funds for their work and if they decide they can buy TEC tokens as well
Could you tell me what do you think about my arguments in the previous post, pls?

Data will tell you whatever you ask for it. Will data tell you all those IH whales (not whales of the common which look like its not being understood) had to sell their crypto on a bull market to be here full time?

Hi Zepti, I am not sure where you got that impression, I am not telling TE anything about tokens. I am asking for data scientist to review different parts of the data, based on suggestions submitted by the community in the meeting last Sunday, and in the Soft Gov meeting yesterday.

I don’t want to engage in debate or convince anyone as I am not advocating for any particular action, except to have a deeper look at the data, based on the suggestions by the community.

The data scientists that we are asking have no stake in this system. Only Shawn, Sem and Andrew Shinichi (who may or may not participate in the analysis) have any stake or Impact Hours. The others, who will be doing the majority of the work, are impartial and have no Impact Hours, and will not be affected by any decisions made by the community or the outcome.

The main look - at distribution, is not a subjective matter and has percentages that exist as fact. Anyone is welcome to submit suggestions for what aspects we should look at, and the sessions will be open, pre-advertised and recorded.


I’m so sorry for misunderstanding your intentions, its good its all clear.
Sorry again, peace :v: