My thoughts on pollen distributions

Today, I attended the meeting that was mainly to talk about pollen distributions and people gaming it. I did not make any interventions as I wanted to write a forum post so I can explain better my feelings.

Let’s be realistic, the pollen distributions have a relatively high payout per week. I think the payouts should be made smaller so there’s no reason for gamers to do it and legit contributors will still have room in the long term. I love sourcecred and I will continue working to improve it, but that takes time.

I may not be right, but I feel that the proposals are the best mechanism to use once we find a way to make that the proposals ask for a USD amount instead of HNY due to its volatility.

The time spent finding gamers and punishing them can be used to come up with really cool ideas or projects that can be potentially funded by a proposal.

All the kind of contributions fit into a swarm, and I think that if you want to get some kind of higher incentives for contributing you need to coordinate a swarm and request a payout for the equivalent amount of work made by the swarm and distribute it among the members with a Multisig or a DAO. I believe that Pollen shouldn’t be the way we distribute $40-50K USD worth a week mostly based on discord reactions and likes, rather a smaller amount.

If you have done something relevant, I’m pretty sure the community will be aware and support your swarm proposal!


How much smaller were you thinking of? This just sounds like punishing the contributors (devs, support, desingn) for the possibility of having gamers.

I might sound spoiled here, but I think the system should just be made more bulletproof - ofcourse I understand there are multiple projects going on for the devs, but just decreasing distribution because for the general contributors each time a gamer finds a gap in the system will probably destroy the value of 1hive/Honeyswap.

1 Like

I think payouts should be smaller to reduce the incentive to game the system but also to leave room to be increased in the future when more members join.

However, i think we should also have an active system to penalize gamers and cheaters, just like we have a reward system for those that are playing by the rules and being positive contributors.

In the short-term i feel like creating a “suspect” role for these cases would be the most simple way to do that and also serve as tool to tone down a bit of the current frustration of multiple members towards how gaming is being in practice allowed.


I don’t think the payout should be made smaller, when more people comes in and join in. The payout will be smaller since more people will get pollen. I think that doesn’t tackle to problem, even if the payouts are smaller, people will still try to game the system.

1 Like

I think the payouts should be significantly smaller. 50k a week ~ 200k a month ~ 2.4m a year on a system that we’re aware has large fundamental flaws and doesn’t reward contributors fairly is wasting money. Perhaps once we’re sure it can deliver reliably we can consider increasing it. Until then I agree we should reallocate those funds using the proposal system and create payments to swarms or individuals for specific work.


hello~ ^.^/

I think there is 3 group in pollen system

1- community members that’s do actual work , coding , designing , management and marketing (etc)
these type of members who add the most value to project should get most out of pollen.

2- community members that’s do semi work . helping in discord . engaging to meeting calls , making post in 1 hive and shearing there ideas & feedback (etc)
these type of members help the community by providing better system and more diversity idea .
3- community members that’s do tiny work , memes and talk in discord
these type of members still important and should not try to limits them because they can help the speed of growth


First off I encourage everyone to get involved in the discussion. For those who are unaware, there have been some instances in the community where people appear to be taking advantage of the SourceCred system. This is a summary of the concerns and potential solutions to these problems that have been discussed in great detail in the pollen dev channel on discord.

Initially, members of the Fauna swarm noticed that there were particular instances in discord where there appeared to be ‘emoji liking’ groups where users were essentially giving themselves cred. This became more evident when we ran the percentage change from version 1.1 of SourceCred to version 1.2 (changes detailed here: Updates to SourceCred). There were two users that immediately stood out, and upon further investigation it appears that they were engaging in unfair behavior.

Users ‘gaming’ cred:

The users that we documented that were ‘gaming’ cred are:

1. F-THV (anonmfb1 on discourse)

  • Evidence: 3 accounts all created 1 day ago at the same time with 20 likes on discourse

2. Wafey123

  • Evidence: 2 users consistently liking and emojing all posts/messages
  • Users honey distribution has been frozen pending further investigation

There was then further discussion once these users were identified how we should proceed. There were not necessarily any rules in place, and these were not hard coded into the system to prevent this from happening. There was much back and forth discussion for both sides of the arguement. Ultimately one users (Wafey123) payment was halted for the next source cred distribution. F-THV, and the other accounts acting as bad actors have not been halted as of yet.

This poses further questions as to who should be allowed to make these decisions? Should a users funds be able to be frozen? What should the punishment be for ‘gamers’?

Check-in Channel:

Check-in channel was initially disabled by the fauna swarm, & then the pollen swarm came to agreeance once both groups came to the realization that check-ins can be done in a better way. Right now there is an overabundance of cred being minted in the channel for content that is usually a repost of content that has already been posted elsewhere. For example, one post in check-in received 22 super-emojis, and 17 regular emojis. If we assume 1 emoji = 1 cred from a verified user, then (22*2)+(17) = 61 cred. Discourse posts are rated at 4x when compared to discord, and therefore 61 emojis on discord would be equal to 15.25 likes. Now if a user was able to get that every day on average, then they would be getting a very large number of cred just from checking in. It is substantially more difficult to get 15 likes in discourse, and discourse is usually original content or ideas not a repost of what you accomplished already that day.

We have considered so far opening the channel back up with 0 minted cred, or may also consider 0.25x like the meme channel (although that may goto 0x as well).

Discourse Concerns:

Right now there are some loop holes on the discourse side of things. Any user can like a post, and after the user becomes level 2 verified then all the likes will mint cred to the users that they liked. Additionally, right now the daily cap of likes is 50. There is even a badge for using them all in one day, which may unintentionally incentivize users to use them to get the badge. Some users may think the badges generate cred when they receive them. Additionally, my thoughts are that the amount of daily likes should be reduced to something reasonable, like 5 likes per day.


Pollen Distribution:

The curent amount of pollen distribution is currently 50hny per week - with 25 going to the weekly cred and 25 to overall. This amount was set early on when the price of honey was much lower. If we assume a price of $1000/hny, thats $50,000 per week being distributed into pollen. If we annualize that, thats $2.6 million dollars a year. There is no other token or network in the ecosystem that is paying nearly that much for an incentive program for the community. It may we wise to reconsider this number, and lower it in the mean time - especially with all the concerns of gamers.

How Do We Proceed?:

The way forward is not an easy one, but through further discussion we can implement solutions and fix these problems. Again this is open to the community to voice their opinion and come to a reasonable solution. I believe that the rules should be hard coded into the system so that there is no ability for anyone to game, although thats easier said than done. So what do you think is the best way to tackle these issues? Should a cafe call be scheduled to further discuss?


I used to give people a lot of emoji so noobie to get something to increase hyn holder and have diversity ~ but I notice I got a lot of hate doing that ~ so just stop doing anything discord ~ just cafe and General what I spent my time there

Wasn’t there a poll made already on how much Honey should be distributed via Pollen per week?

If this community vote is getting overturned, and I think 250 is a lot, I voted for 50, then why bother asking the community what they think if in the end ones actually holding power over everyone, the developers (who are working on these products) overturn the decision that the community made?

1 Like

it is a bit strange. But I do think most of this “people who previously were paid a lot suddenly are paid basically nothing” can be attributed to the 50%+ growth we have per week, among some other things like gh weights and some seeds being inactive like burrrata

@sandpiper It’s mainly a result from retro-activity and lack of time stamping. Let’s take the memes as an example, but this is also true for check-in, (and soon design I assume) and the forum: When I started meme-ing, there were about 40 memes in total. The next 40 memes were maybe 40% mine and many seemed to enjoy these. Looking back the memes were maybe getting a too large share of total cred, but that was the situation back then, before the gamers arrived 2 weeks later. By retroactively zeroing the memes (and checkin etc), we take out earned cred from everyone that ever posted a meme (/ did a check in etc).

Besides slashing the absolute cred, it also therefore also decreases the historical value/% they had. As the total amount of cred minted past weeks skyrocketed due to both gaming and x00% increase in users, the historical cred had already plummeted for those who were active before the game season started.

I know we are supposed to change the system and not the players, but currently, in order to punish a handful of bad players, we are destroying the system for everyone, retroactively. I don’t think there is an easy solution, but the combination of slashing everything that isn’t GH retroactively and letting the players continue isn’t one.

  • Those cred buckets we discussed before would surely help, at least for the GH cred that is currently buried under discord and discourse spam.
  • Having time-stamps for the channel weights would help early checkins for devs, early designers, and even early memers and others to retain their absolute value, even though the relative/historical value will still go down a lot due to the absolute increase in users/cred, which is ok.

Still, the gamers gonna game. If Simon still gets 1 HNY when the meme channel is zero, then he is gaming elsewhere and big time (and we already have shared many examples), as he is not contributing in any sensible way.
Ha, even Pouria is still above me in the new test distro with memes 0. We need better analytics imho. Or scrap the entire system.


No, au contraire. As a reminder, my post in that voting thread:

But is clear to see the gaming going on, let’s just check the number of votes vs the number of profiles with a profile picture:

250HNY: 8 profile pictures out of 36 voters (22%)
50HNY: 24 profile pictures out of 31 voters (77%)
25HNY: 17 profile pictures out of 19 voters (89%)
35HNY: 16 profile pictures out of 19 voters (84%)

Of course, there are legitimate voters that don’t want or haven’t uploaded a profile pic and there are fake accounts that do, but the overall difference is very clear.

So, in fact, we stopped fake accounts from overturning the community vote and went with the real majority.

1 Like

Yeah no, that’s not how voting works.

Voting should be 1 vote per person. Judging the votes by “they have profile pictures” is just an excuse to have your own way.

I don’t want to only criticize, so here’s my solution, which is what you should’ve done - Cross-check what discord accounts are linked to Pollen/Discourse (if they’re not linked to Discourse, find a solution), then, determine if those Discord accounts are verified on BrightID, if they’re, their vote counts.

That is objectively better way to judge the voting poll than just “profile pics bro”.

1 Like

Thanks for the summary @Blazar.

I think the community should make a decision based on a Poll, but in my opinion they should be removed from pollen distribution.

This channel should only be used for updates, as most of the content has already received creds on other channels.

In my opinion, cred should be 0x, I don’t really see the value. It’s great fun, but we should find another way to give rewards.

I don’t really agree to put a cap on the number of likes per day, because there are people who post really good stuff that makes an impact on the community. Discourse has a pretty good tracking system and gamers are easy to spot. Again, the community should vote if a person should be removed or not.

In my opinion, the system works well as it is and shouldn’t be changed just for few bad apples. Even if you are new, you will get some rewards.
Of course as always there will be problems, but no system is perfect. The important thing is that people work hard to make it fair for everyone.

Yep, that’s better. Please do so and share the results.
Hint: all these fake accounts were registered the same day, the day the vote went on.

You can easily distinguish them from real accounts.

Still, the system needs to implemented by which Discourse accounts can be linked to and verified by BrightID.

I think cred buckets may be the fairest way to go from here, something like 40% GH, 35% discourse, 25% discord. Biggest piece of pie shared between those contributing code, followed by discourse which is generally better thought out views and opinions, followed by discord which is largely chatter and some support (support is also offered by discourse)

I havent recieved my first distribution yet, so in a sense i dont have much saying in this.

But if there is such obvious evidence that some members are gaming the system, why not allow everyone to vote to ban them?
It looks like everyone is more or less in the same page

1 Like

How I see this is that we should definitely lower the payout. These rates are insane and it incentivizes people to cheat. Having actual proposals to fund swarms, and having pollen be more like a nice bonus is the way to go in my opinion.

The checkin channel was definitely a huge cred farm, and we definitely have to think about how to go about it. Putting checkins more in the swarm related channels could be a solution, however I wonder… wouldnt people just start spamming in their own swarms then? One solution would be to have fauna/seed members who represent a swarm make sure it doesnt go out of control.


hello will, i like that you added individuals for the proposals, since it seems the simplest way to solve this.

in regards to pollen, i would just keep adjusting the weights, 0.1 to unverified, 0.2 to verified, 0.6 to bee, 1.5 to seed. not sure how the other roles are achieved so i don’t know what to say for example of honey category on discord.
i obviously want to make money but i don’t understand properly the system to know if pollen rewards should be increased or decreased.
by personal experience, i’d rather be cautious than spewy