Assignment – Evaluate the Coordinape Tool for Mental Mining

Strengths and Weaknesses of the current Mental Mining

The Mental Mining Program is evolving. Just recently we introduced a major change to our reward structure [①] and to the method of allocating those rewards [②]. These changes, in particular the use of the coordinape tool to allocate rewards, had been introduced as being experimental. We are now looking to systematically evaluate these experiences and are asking for your feedback.

Background information on the latest changes

① Reward Structure

previous reward structure

Initially rewards had been linked to Citizen’s XIO token holdings. Reaching a higher rank on the leaderboard one would also receive a higher percentage earning on their holdings (limited by a maximum cap). This system had always allowed larger holders to profit stronger than someone with small holdings.

current reward structure

To make participation in the Mental Mining more attractive to new Citizens (with small holdings) we have been experimenting with a different approach over the last weeks. Rewards are calculated solely based on the contributions to the discussion and are no longer linked to wallet holdings. This allows anyone to earn substantial rewards simply by adding value to the discussion. The main concern with this method is that people can cheat the system by simply creating multiple accounts and earning on each of those.

② Allocation of Rewards

previous allocation process

Until recently all grading had been performed centrally by a single person (expert grading). A predefined grading logic was used to determine the number of credits each participant received. There are some disadvantages with this approach. For one, the fact that the whole process had been performed by one person only made it somewhat subjective. There is also a natural limit to the amount of discussions (comments) just one person can grade by themselves, therefore limiting any future expansion of the program, unless additional resources for grading would be allocated.

current allocation process

Currently we have been experimenting with the coordinape tool. Each participant is asked to take part in the collective allocation of rewards using the coordinape tool. The goal with it is to crowdsource the grading process thus letting all participants collectively come up with a fair distribution of the available rewards. One disadvantage of this system might be seen in the fact that all participants are asked (required) not only to provide quality contributions to the mental mining assignment but to also take the time and energy to read through and grade all comments.

Background information on the coordinape tool we use:

This is a short (4:42) introduction video:

Results of our last two assignments

There are two Mental Mining assignments that have been graded using the coordinape tool.

  1. Assignment – Feedback on New Website
  2. Assignment – Consistency for Content & Terminology

You can view the results on the coordinape tool by logging in the tool with your registered wallet. Please do so and inspect the graph. The examples display the results from the second assignment. What information can we extract from it? How does it help us? (If you do not yet have a registered wallet please follow the steps described in this post and wait for your address to be added manually on the backend.)

detailed screenshots














Some questions you might want to provide feedback upon

  • What do you like about the new system? What benefits do you see?

  • Which problems do you see? How would you reduce some of the friction points?

  • What potential threats do you see? How could people try to game our system? How can we identify potential cheaters? How can malicious behavior be prevented?

  • Can the coordinape tool actually help create a more valid grading result?

  • Should we require all participants to log into the coordinape tool and actually take part in the allocation process in order to become eligible for rewards? Or should this be optional?

  • Looking at the results (s. screenshots above) do you feel the outcome is valid? Has the tool lead to fair results? What questions come to your mind?

Bonus question:

  • How should core team members be treated in this process? Should core members be encouraged to participate in the Mental Mining Assignments? Should they be allowed to earn credits that convert into XIO?

This assignment will be active for one week and close on August 31st. You will be able to perform the grading on this discussion with the coordinape tool until Sunday September 5th.

6 Likes

First of all, I’d like to acknowledge that a single person grading is a tremendous amount of work, and that it is not a scalable solution. A move to a more decentralized approach seems unavoidable, and coordinape may be a good step in that direction. I personally had issues with it in the beginning, because it is a new platform and it takes some time getting used to, but I understand why it is needed.

What do you like about the new system? What benefits do you see?
The main benefit is that it allows for mental mining to scale. On the road to more adoption, there will be more and more people participating, and it is simply not possible to have a single person deal with the marking. Furthermore, one single person cannot be objective 100% of the time.

Which problems do you see? How would you reduce some of the friction points?
The main problem I see is that grading properly takes a significant amount of time, and that we can’t expect everyone to do it very seriously. I must confess that even I did allocate some points randomly. There is an individual pressure to allocate all points, but doing so fairly is a lot of work, so people may just end up allocating parts of their points randomly, or giving points to users they know. I noticed that their was a bias in round 2: if a user gave me points in round 1, I was tempted to give them points in round 2, regardless of their contribution.
At the moment, there is no incentive to grade fairly, and a bias towards rewarding the same people every time (since they are likely to reward us in return). This is not an easy problem to solve.
The mitigation that comes to mind is a form of reputation system, where a user who marks fairly (i.e., not randomly, and consistently with the rest of the crew maybe?) obtains some form of advantage, for instance, more points to distribute, or a compensation. On the other hand, bad actions would result in penalties (less points to distribute, or a decrease in points received).

What potential threats do you see? How could people try to game our system? How can we identify potential cheaters? How can malicious behavior be prevented?
The main threats are false identities (people creating several accounts to comment and farm mental mining rewards), and collusion (a group of people only rewarding each other, regardless of their contribution). The previous system was an excellent counter to false identity issues, and it is more difficult to address under the new system.
Detecting all malicious behaviour is almost impossible, but the reputation system mentioned earlier could help. One measure of how honnestly a user is behaving would be how close or far his input is from the average (e.g., if they consistently allocate more points to a user whose contributions were rated as low by everyone else, or if they do not allocate much to the top commenters).

However, I believe that ultimately, if there are no rewards for participating in coordinape, only a small subset of citizen will do so, and it will be easier for them to collude.

One potential solution to this issue is to recruit “super voters”, people whose behaviour has been consistently good, and who would have more voting power than the rest of the citizen regarding mental mining rewards. These super voters would receive compensation for their work, in exchange to forfeiting their eligibility to rewards (to prevent corruption). That would be a form of hybrid model, not fully decentralized.

Finally, coordinape being separated from the forum makes it extra tedious to use, since we need to go back and forth between the forum and coordinape to see the contributions. Is there any way to integrate coordinape to the forum, so that we can distribute points from there directly?

Can the coordinape tool actually help create a more valid grading result?
Yes and no. I think it still needs supervision, until there is an established base of well-behaved voters.

Should we require all participants to log into the coordinape tool and actually take part in the allocation process in order to become eligible for rewards? Or should this be optional?
Both have their benefits: we need people to participate in coordinape, so “forcing” them to do so is a way, but low quality coordinape ratings are actually detrimental to the system (random allocation etc…). If users are forced to participate, but there is no accountability, they may just vote randomly.

Looking at the results (s. screenshots above) do you feel the outcome is valid? Has the tool lead to fair results? What questions come to your mind?
Honestly, the screenshots are not easy to interpret and I am not 100% sure what I am looking at.

Bonus question:

How should core team members be treated in this process? Should core members be encouraged to participate in the Mental Mining Assignments? Should they be allowed to earn credits that convert into XIO?

Definitely, the core team members are citizen as well, and they should be encouraged to propose ideas and solutions regarding issues that may be out of the scope of their role. The core should be eligible for mental mining rewards.

8 Likes

It’s is a lot of work to grade everything and that’s whats good with the new system. The grading is done faster and “more efficient”. It is definitely a nice to see that new comers (people with less holdings) get paid the same amount as everyone regardless of you status/wealth.

I quoted “more efficient” and that is because simply the points can be arranged and speculated upon. People can randomly assign points and pay for no work done, but hey one problem solved another is born, we can’t have both right. I was also thinking in the direction of lets say like 5 task average or just the average of X amount of task done in month. Lets take 5 for example. The total points allocated within 5 epoch’s would be averaged and at the end of the month it’s paid out. It would bring more consistent assignment of points and more consistent participation.

Already mention people can just randomly assign points to someone they know or they can make a pact to assign just between them two etc… The problem with people having more wallets would be solved to some extend with the averaging because if they only allocated between their wallets the majority would win. And that problem isn’t that big off a deal if someone opt’s in only the people that participated in discussion on forum

It is definitely a good way to decentralize the mental mining. It has ups and downs.

Nah i think if we force them to participate they will randomize their allocation even more. But it would be fair for them to vote. Also the average could alleviate it to some extend.

I think it is quite honest work in my honest opinion. I think the natural curve (gauss curve) is nice and as it should be (68% of all participants should receive from 40-60% of total allocation and i think that should be implemented if it wont happen).

image

PS: I also noticed i was opted out and couldn’t allocate in epoch 2. Is everyone opted in by default or someone opt’s in the participants?

Actually they could be a part of mental mining. If we keep this system they could participate and they should be allowed to earn credits if people vote so why not. But as their precious time and knowledge is already beyond what we discuss here :stuck_out_tongue: and as this i incentivization program i think they honestly don’t need that. They are already neck deep into it :stuck_out_tongue:

4 Likes

What do you like about the new system? What benefits do you see?

What I like:

  • It encourages you to read everyone’s comments if you want to attempt to allocate fairly, which I wasn’t always doing before.
  • The mapping system in coordinape has a clean way to view where you get your votes from, which was interesting to me.

Benefits:

  • Reducing the burden on one core member to evaluate voting.

Which problems do you see? How would you reduce some of the friction points?

Problems:

  • Learning curve to use. - discourages new citizens.
  • Much more time is required to evaluate posts, which many people don’t have and therefore won’t allocate fairly.
  • disconnection between discord, the website, and this voting system.
  • reward dilution. As soon as we get more than 20+ people in this system, people will become less likely to participate. If you are at most likely to receive 500XIO, which is 50-60USD per assignment, and claiming that costs 30-40USD, then people are not going to spend hours writing considered posts and evaluating others.

Solutions to each problem:

  • Make a coordinape discord channel or pinned post with an explainer video or do a mental mining campaign for a simple graphic or medium article for our use.
  • Having an explorer role to evaluate posts separately and ensure people are allocating fairly through independent perspective. If this explorer sees a valuable contributor not receiving points, they can step in and award more to that person themselves.
  • Another solution being nominated super voters like @Meerlion suggested. Maybe these people have to read all comments and get paid 20% more XIO as a result?
  • Make rewards scale dependent on number of XSI participants. 0-10: 10,000 XIO. 11-20: 15,000 XIO and so on.

What potential threats do you see? How could people try to game our system? How can we identify potential cheaters? How can malicious behavior be prevented?

  • Aforementioned cheating with fake accounts or friends colluding. This is easy to see if you have an explorer or core citizen checking the coordinape vote streams, you could make note of any suspicious activity or reward citizens for raising it to core attention.

Can the coordinape tool actually help create a more valid grading result?

  • Yes. However work needs to be done to ensure incentives are in place to ensure at least some people are reading all the comments, and that fake/malicious activity is reported.

Should we require all participants to log into the coordinape tool and actually take part in the allocation process in order to become eligible for rewards? Or should this be optional?

  • How would this look if they didn’t use the tool? If that would involve some people just getting XIO based on core member reviews and some based on community reviews then that doesn’t seem fair or scalable. Make all use the app.

Looking at the results (s. screenshots above) do you feel the outcome is valid? Has the tool lead to fair results? What questions come to your mind?

  • Seemed relatively fair but there is work to be done around education, reward scaling, and incentivizing good behavior.

Bonus question:

Should core team members be treated in this process? Should core members be encouraged to participate in the Mental Mining Assignments? Should they be allowed to earn credits that convert into XIO?

  • Devils advocate here. No they shouldn’t, I think core members by virtue of their work and positions, and the fact they have part or full time jobs at BZ get to have their thoughts and views heard already.
  • XSI is an opportunity for citizens that only have a few hours spare a week to have their voices heard and earn a few $ on the side.
  • I appreciate that there are positives to allowing them, but the negatives in terms of diluted rewards, and opinions being shaped/skewed by seeing core members opinions before citizens can voice their own would create biases and more negatives than positives in my opinion.
7 Likes
  • Devils advocate here. No they shouldn’t, I think core members by virtue of their work and positions, and the fact they have part or full time jobs at BZ get to have their thoughts and views heard already.
  • XSI is an opportunity for citizens that only have a few hours spare a week to have their voices heard and earn a few $ on the side.
  • I appreciate that there are positives to allowing them, but the negatives in terms of diluted rewards, and opinions being shaped/skewed by seeing core members opinions before citizens can voice their own would create biases and more negatives than positives in my opinion.

You’re making very good points here. My (biased) opinion, as a soon-to-be core member, is that each core member has a very specific role. On my side, I’ll be working on security; as such, it will not be part of my job to take time to think about and formulate answers to e.g., marketing issues within the mental mining program. I’ll still do it, because I care about the future of BZ a lot, but it would be nice to receive at least partial compensation for that. Like in any job, if you put in extra work on something that you do not strictly have to do, it is nice to get some form of acknowledgement.
Maybe there could be a capped reward for core members? Or their credits could be divided by 2? Would it be acceptable to you if the rewards for core members were capped at, for instance, 200 XIO per task? That would not be a big enough amount to dilute the rewards of the rest of the citizen, but still be something.

3 Likes

I think core members committing time outside of their paid roles to add value is commendable and deserves recognition, but shouldn’t come at the cost of community engagement.

I like your 200 XIO cap idea, other ideas which wouldn’t detract from community incentives include:

  • BZ management incentivizing XSI amongst the core members by creating performance bonus multipliers.
    For example, X core member chooses to fully participate in both XSI assignments for the month then they get a 5% pay rise instead of a share in the XSI reward.
    Downside to this could be creating (potentially) far more core activity on the forum which could drown out the community, also some core members wouldn’t necessarily like to spend their time this way; so maybe there could be multiple ways to earn this 5%. For example the aforementioned XSI way, plus a bug finding way, a project management way, a tool evaluation way. Essentially some way that the core can be recognized for contributing more than they need to in the pursuit of BZ improvement.

This has turned into a bit of a rabbit hole now, and makes me think of a book where a “Chief Incentivizing Officer” was a role that was adopted.

2 Likes

^^^ This is a very interesting approach. We could apply some statistics here to determine how consistent a users grading is. I’d be interested in experimenting with this approach.

Along these lines – we could try to make this more difficult by introducing some minimum requirement to become part of the mental crew in the first place.
An interesting approach has been introduced by the coordinape team just last week. A vouching system. (See: Vouching - Coordinape)

It comes down to the question: Should we introduce some kind of minimum requirement before allowing new people to join the Mental Crew? (Many different minimum requirements can be thought of, e.g. certain time (history) on discord or the forum, certain amount of quality comments made, approval by core team or council member, or some kind of vouching system as the above mentioned.)

1 Like

What do you like about the new system? What benefits do you see
I see mental mining as a way to express your opinion, but have it informed/shaped by what others think. This system encourages you to read what others are saying so that you can give a fair and informed allocation of credits. By taking on board what others have said you challenge your own original position. I did this on the last assignment as some of my points changed over the duration.

Which problems do you see? How would you reduce some of the friction points?
It did take a reasonable investment of time to learn the new system, it was also difficult to align the feedback given in the forum with the people listed in Coordinate. It would make more sense to provide a rating against each piece of feedback as you read it in the forum (this would not be visible to the contributor). At the end of the assignment you would they get a tally of the contributions made by each participant and the score you gave each comment. You could then use this as a guide to allocate your 100 credits. General comment - if the reward for a tasks is minimal the value will be close to zero once you pay gas fees. At the moment it costs the equivalent value of 250 to 350 XIO to make a claim.

  • What potential threats do you see? How could people try to game our system? How can we identify potential cheaters? How can malicious behavior be prevented?
    This has been well document by others. Not sure if Coordinape can keep the allocations anonymous, but this would prevent a reward feedback loop - ‘you reward those who rewarded you so that they will continue to reward you in the future’ … and so on. I think it is important that there is oversight of the process by BZ to ensure that feedback that is useful and adds value to BZ, is rewarded regardless of the coordinape score. I think the coordinape score should be say 40 to 50% of the assessment score with the balance coming from BZ based on the usefulness. Also consider paying a premium to feedback that is the catalyst for real change within the area being discussed.

  • Can the coordinape tool actually help create a more valid grading result? I think our unconscious bias will always taint our marking. I think there is a tendency to give more points to someone that agrees with or provides similiar feedback to your own. If you think someones feedback is irrelevant, poorly thought out or simply wrong, would you give them any points. Therefore feedback that is radical or sits outside the normal range is likely to be poorly rewarded. This overtime may discourage diverse and challenging feedback - this would be counterproductive and the opposite of the objective of mental mining. As stated above we need to ensure that coordinape is only one part of the reward calculation.
    *** Should we require all participants to log into the coordinape tool and actually take part in the allocation process in order to become eligible for rewards? Or should this be optional?**
    I think coordinape should be optional but if you don’t participate then you don’t get a score and therefore your rewards would only be based on the points award by BZ

  • Looking at the results (s. screenshots above) do you feel the outcome is valid? Has the tool lead to fair results? What questions come to your mind? Not sure how to interpret the screenshots but looks like most people allocate across a reasonable number of people rather than favouring just one or 2 contributors. I think if we have 20 or 30 people participating it will be difficulty to fairly allocate your 100 points and may result in people making uninformed allocations. I also think being the first couple of people to provide feedback may lead to a higher allocation of points due to name recognition. When you go into coordiape at the end of the assignment, you flick back and forth between the forum and coordinape and you tend to focus more on the first few posts. Bonus question:

  • How should core team members be treated in this process? Should core members be encouraged to participate in the Mental Mining Assignments? Should they be allowed to earn credits that convert into XIO?
    Given the goal is to gather quality feedback on a topic, I don’t see why you would want to discourage participation. I assume all core team members already have an effective mechanism to provide constructive feedback on the operations, issues and challenges faced by BZ. If they want to participate in mental mining and help to encourage and facilitate a productive exchange of ideas then they should be encouraged. However the rewards they receive should not diminish the incentive for ordinary citizens to get involved. Therefore BZ core should be rewarded for their contributions outside of whatever rewards are offered for mental mining.

3 Likes

I like your idea of reward scaling based on the number of participants. Assuming there is a correlation between the number of participants and the volume of quality feedback then the extra cost will be offset by the new ideas generated.

2 Likes

I believe the switch is for a good cause but a little more work is needed on it because it can be manipulaated.

2 Likes

i will be straight: i don’t like this new system. I feel that with the current system, only a handful of citizens could actually gain some rewards for their effort, and i feel like the overall rewards went down quite a bit. Also, having to use a third party website to “rate” my fellow citizens is really an extra step that is keeping away the fun from the XSI

i would reduce some of the friction points by not having to manually assign points to others. We need an automated system.

i am not sure if this is possible, but i could create another account, and using that account to give me a lot of points. Also i could ask a friend citizens to give 100 points to me, and then we split the rewards.

it can, because it is “organic” grading system. But still, the effort is just too much.

well, at this point it should be optional. I feel like it would be easier for people to just write and forget about it, and get some rewards.

yes, it is valid. Yes it is fair…but still, i think that this would appeal only to a very small number of citizens / Crypto users.

well, i think team member are getting a lot of XIO tokens already, i don’t see the point in centralizing more xio tokens in their hands.

2 Likes

It comes down to the question: Should we introduce some kind of minimum requirement before allowing new people to join the Mental Crew? (Many different minimum requirements can be thought of, e.g. certain time (history) on discord or the forum, certain amount of quality comments made, approval by core team or council member, or some kind of vouching system as the above mentioned.)

Hey @jorn , I like your suggestions, I suppose the question before we can answer this is what purpose does XSI serve for BZ primarily?

  • If it’s a tool for legitimate and well considered views from the community that weigh into BZ decision making then making users wait a month as discord/forum members or answer X amount of questions with quality feedback first is worthwhile.
  • However, if it is an on-boarding tool that is designed to attract newcomers then having them wait any amount of time or be evaluated first will be a roadblock that will turn many away. (Unless they dive into other areas of BZ first and discover the value this project and team can offer)

Maybe there needs to be tranches of ‘ease’ for BZ missions. Liquid fire being the easiest, followed by more time demanding missions like XSI, and then obviously core being the most demanding as a full time job.

2 Likes
  • What do you like about the new system? What benefits do you see?

The new system will be a lot faster by having multiple graders instead of just one, which one grader can only do so much in a timespan. The benefits are that graders will get done reviewing comments/discussions and appropriating how much each participant gets for their input.

  • Which problems do you see? How would you reduce some of the friction points?

Sense there will be multiple graders each grader will have their own method of determining how much reward goes to a person. So in this sense grading would be subjective based on the graders understanding and human nature. To reduce friction points there would have to be a set of master guidelines that each grader has to check of with an amount associated with that guideline. It could be like a troubleshooting tree for giving out rewards.

  • What potential threats do you see? How could people try to game our system? How can we identify potential cheaters? How can malicious behavior be prevented?

As in everything now a days we are going to have scammers and people that will try to take advantage with getting as much rewards as possible. A good way to exclude bots that people will make for commenting and discussing would be to have a verified ID when the person logs in. Scanning similar emails and data points would help to ID potential cheaters. We could give out rewards to other graders for being like a neighborhood watch type deal.

  • Can the coordinape tool actually help create a more valid grading result?

I believe it can because multiple graders can view a comment/discussions and give their input on what each person receives as reward, so it would be more of a collective decision.

  • Should we require all participants to log into the coordinape tool and actually take part in the allocation process in order to become eligible for rewards? Or should this be optional?

Yes, everyone needs to participate in order to get rewards, as doing so we will have more collective input on what everybody is thinking and how they see one peer.

  • Looking at the results (s. screenshots above) do you feel the outcome is valid? Has the tool lead to fair results? What questions come to your mind?

First nothing is ever fair even when we try to make, but as a collective the room for error would diminish because we are all grading and rewarding at the same time. Error will be always present we just need to minimize the error and using coordinape, I think will help minimize the error. Some questions that could be made are; What would be considered fair and what will be considered foul, another could be if the error, i.e. a very large discrepancy between the graders. How would the issue be resolved.

  • How should core team members be treated in this process? Should core members be encouraged to participate in the Mental Mining Assignments? Should they be allowed to earn credits that convert into XIO?

Core members should always be involved because it is their vision that they want input on from the world. It’s pretty amazing how different people can give different ideas from one single question and/or idea brought to a table.
Yes, they should get XIO

1 Like

What do you like about the new system? What benefits do you see?

  • I really like that the new rewards structure is equal among all the users. With the old system it might have felt a little bit unfair, especially to new users.

  • As far as grading goes the previous solution to let one person grade everything isn’t in any way a scalable solution. Coordinape is a great solution to counter that problem since it let’s everyone distribute their rewards as they see fit and have their opinion have real consequences, so it makes the user feel as a part of the system.

Which problems do you see? How would you reduce some of the friction points?

  • First problem that comes to mind is unfair distribution (giving points to friends or because you know someone). Have to admit that I distributed some points a little bit randomly aswell as I didn’t delve into all the answers 100%.

  • Coordinape itself doesn’t seem to have any significant friction points since everything is pretty straightforward. Nonetheless I still believe that the forums should be integrated with Coordinape since it’s kinda hard to follow as to who you are crediting the points to and it feels a bit tedious. As for the solution maybe putting in place voting system in the forums would solve it. If we take Reddit for example then there are upvotes which almost always determine the quality of the post. Same could be done with mental mining points and there could be an option to sort by the most points or something like that. Even though there’s a risk of not noticing new posts but there will always be downsides to a system.

  • In the first mental mining that I participated in I saw two people exchange almost 100 points between each other and that’s why I think a reputation system would be nice to see. In Estonia we have a tech selling website which uses reputation system really well. If you make a deal with someone they can rate the experience as: positive, neutral or negative and there’s always a comment as to why you think so. Maybe not implement it exactly like that but I think adding a rating with a must have comment makes the rating system really good since there’s an explanation as to why you think so. This system could also be tied to mental mining points. For example if you get a lot of negative reviews you wont receive as many mental mining points at the start of the round.

What potential threats do you see? How could people try to game our system? How can we identify potential cheaters? How can malicious behaviour be prevented?

  • The system seems to be easy to game. Multiple accounts, friends giving points to each other and no control over mental points assigning.

  • Identifying potential cheaters could be done in multiple ways by doing an IP address check, ID check but these can be countered. I think the best idea would be to use brightID since there are real people who confirm your identity.

  • Countering malicious behaviour is always a hard thing to do. Only thing I see that can counter it is the reputation system as explained earlier.

Can the coordinape tool actually help create a more valid grading result?

  • If it’s improved and made more convenient to use then yes, but in the state it is in now then no.

Should we require all participants to log into the coordinape tool and actually take part in the allocation process in order to become eligible for rewards? Or should this be optional?

  • I think it should be required since then you will actually give feedback to the users so they can reflect on it. It’s kinda the same with universities where at the end of the semester you have to give feedback on the subjects and I really like that.

Looking at the results (s. screenshots above) do you feel the outcome is valid? Has the tool lead to fair results? What questions come to your mind?

  • I think there’s definitely a bit of first come, first served bias there. The results seem kinda fair with a few exceptions.

How should core team members be treated in this process? Should core members be encouraged to participate in the Mental Mining Assignments? Should they be allowed to earn credits that convert into XIO?

  • I think core members should definitely be encouraged to participate since they are at the core and can definitely offer valuable ideas and expand our ideas.

  • Core members should be treated as equals to earning XIO since their time should be rewarded as well as anyone elses.

1 Like

I 'm not a fan of introducing minimum requirements to join the mental crew. We are looking for a broad spectrum of input which draws from a diverse range of experiences. Therefore each mental mining tasks may attract the interest of different citizens - by placing minimum requirements it could discourage casual or one-off contributions.

1 Like

It would be useful to understand what BZ does with the feedback we provide through these assignments. Jorn - you manage the process but how do you distill the feedback into a useful format to share with others at BZ. To go through all the responses and pull out the beneficial feedback , someone most read it all - I assume the person that does this also does the grading. I think that person is in the best position to allocate the bulk of the points (say 60%). The wider BZ team (that manages the area that the assignment is addressing) could then allocate 20% based on the benefit they get from the respective pieces of feedback. The remaining 20% would come from coordinape scores.

2 Likes

So we could use the vouching system. I would propose that a person gets opted in by having 3-4 (could be adjusted) mental mining member votes. If it turns out he did good (by looking at his point score) in the coming months/assignments (say like X months or Y number of assignments), he could be accepted permanently, if that is not the case he needs to get 3-4 vouches again. I guess that would really incentivize to stay and participate and really contribute to the system. Because until now it could be decided if the contribution was good enough and ranged with credits, with the new system the idea above would be the a good solution to actually distribute the rewards to the ones that really contributed to assignment. It would bring a little bit more work to stay in but i guess that’s the price we have to pay.

There could also be a minimum points to be aquired to be able to participate in the next epoch (something low like 50 points).

3 Likes
  • What do you like about the new system? What benefits do you see?
    The thing that I like most about the new system is that rewards are no longer based on your wallet size and you’re now rewarded solely based on your contributions to the community. The benefit is that there’s going to be more quality contribution from community members who want to be included in the reward pools which in turn benefits the community as a whole. Better contribution, better ideas, better projects, more of a reason to hold XIO.
  • Which problems do you see? How would you reduce some of the friction points?
    The biggest problem I see is biased grading or not serious grading. Im also not sure the average community member is going to want to use cordinape. I think the learning curve might discourage them along with new citizens. To combat this, there just needs to an educational tool that makes the process simple for the average joe.
  • What potential threats do you see? How could people try to game our system? How can we identify potential cheaters? How can malicious behavior be prevented?
    One of the biggest threats I see is people making multiple false identities with multiple wallets to upvote themselves or try to get most of the mental mining rewards. Im not sure the average community member is going to want to use cordinape.
  • Can the coordinape tool actually help create a more valid grading result?
    it has its good and bad but it decentralizes the mental minding.
  • Should we require all participants to log into the coordinape tool and actually take part in the allocation process in order to become eligible for rewards? Or should this be optional?
    I don’t think it should be mandatory because when does anything get done well when someone fees “forced” to do it. You’ll probably get the most quality contributions from people who actually want to be there.
  • Looking at the results (s. screenshots above) do you feel the outcome is valid? Has the tool lead to fair results? What questions come to your mind?
    The screenshots are a bit confusing as I am not very familiar with the tool besides what I’ve been reading on here so I won’t pretend to know.

Bonus question:

  • How should core team members be treated in this process? Should core members be encouraged to participate in the Mental Mining Assignments? Should they be allowed to earn credits that convert into XIO?

I think anyone who has valid and actual contributions should be rewarded. Team members add a a more expertise contribution to the conversation which is a very valuable thing to have. This shouldn’t be discouraged by not rewarding them for said contribution.

3 Likes

It’s better from a decentralisation point of view, and it removes the bias provided by one single person making the evaluation. But it does have it’s trade-offs…

It’s a little bit too complex and involves performing multiple actions besides just posting a comment on the forum and waiting for the rewards. This will turn off some new users I think. What attracted me first to the XSI was the simplicity… Posting 3-4 comments on the forum and then just waiting for the rewards. I would reduce this friction point by not keeping it 100% decentralised, as in having someone from the team that will always supervise and help with any problems regarding coordinape or other aspects of this new grading system.

Here we have a fine line between measures that are too soft or too harsh I think. For example KYC is definitely out of the question, since it defeats the whole purpose of DeFi, so alternative measures could be explored I guess. There’s also the problem of how to differentiate between users who collude and users who simply share their points in a more or less random way… Also having to log in to another app and sending points to other people is one thing, but implementing complex requirements for this grading and voting will deter people from getting involved in mental mining altogether, so overcomplicating the grading process shouldn’t be something we want.

One other problem that I see (unrelated to coordinape tho) is the complexity of the assignments and the number of questions we have to answer in relation to the rewards. For example in this assignment we have to answer to 7 questions and we will only receive a small share of those 10k XIO. If we compare this with the simple assignments we had during the XSI period we will see a significant increase in the number of questions and the complexity of answers that we have to provide… And in the same time we see a significant reduction of the rewards… That’s not something I agree with, especially since now we have to do even more work to grade the comments on coordinape, we also have to post longer/more detailed answers, and all of this to receive some peanuts. /rant over

Probably, but it shouldn’t be used in such a way that puts all the burden on the users.

Yeah there’s nothing wrong with being required to log in and take part IMO, but we should keep the process as simple as possible, in order to avoid everyone having to do lots of complex things for rewards that are not worth it…

I will have to take another look at the screenshots in the following days, cuz for now they seem pretty complex and I can’t follow what’s happening there exactly.

Not a big fan of this, because of two main reasons: 1) Since the core citizens have access to more inside info about the project then it creates a conflict of interest, since they already know stuff we don’t know, stuff that could certainly be useful in making better & more well informed comments… and 2) Core citizens are well known and it’s easy to gather points just because of their popularity or the fact that someone sees a familiar username so they decide to send some points that way.

4 Likes

I had a look at the vouching system and although it would enhance the level of integrity in the system, it will also add an additional layer of complexity and compliance. I think if we adopt this aspect we are at risk of over cooking the process.

2 Likes