Activity or… positive impact? One is easier to game than the other. But the other is certainly easier to quantify. Thing is, if activity is a metric, aren’t we creating an incentive to participate not to say something relevant but to push some convenient metric?
I think the risks of this are small given that participants can be removed. One of the proposals I’m most interested in exploring is how to automate the process of issuing reputation/merit for specific activities we would like to encourage. There was some discussion on this here: Membership requirements for the Aragon Cooperative?
Let’s say for example we decide we want to encourage and reward participation on the forum, because we believe that active participation in the forum will lead to a more engaged community and richer discussion/dissemination of ideas in the community. We could pre-issue some amount of merit tokens, and give an address which is controlled by a bot/script the right to assign them based on statistics of the forum. (eg: http://forum.aragon.org/u) – This by itself is pretty game-able, if the merit tokens provide meaningful influence or control over economic resources perhaps this leads to bots and other forms of collusion on the forum.
But as that happens we can adapt and tweak the policy. We could have a rule where you cannot have more than a certain percentage of total merit supply, and we can only issue merits to validated members.
I think we can track multiple success metrics depending on specific goals of an initiative, but at a high level the most important metric are member satisfaction, retention, and growth (imo).
How would you go about removing participants that are not saying anything bad, simply not very useful?
I don’t get this token thing, it’s just way too complicated and subjective IMHO.
Can’t we just reward with actual ANT proposals that gets validated by ANT token holders? It’s the way it works in District0x I believe? As in if you propose a district that gets validated you get rewarded with their token?
Should I rename myself “Party-pooper of the Aragon COOP” yet?
Maybe we can have some KarmAragon? Works exactly like Reddit, one doesn’t get any financial reward from participation, only social reward. Maybe it’s what you’ve been suggesting and I just didn’t get it?
To clarify, the amount of KarmAragon generated every trimester could be an interesting metric, but do we need a Coop for that, what would it add compared to just doing this on the forum or ideally a decentralised platform (Akasha!)? So back to the Coop, one success metric could be the ratio of AGP suggested validated VS those suggested and refused? There could be more metrics certainly.
Diversity! for the users willing to use a decentralised platform that certifies their origin (Civic?), success of the coop could be how diverse a crowd it represents.
Its a subjective decision, so a vote of the existing members where they basically review the situation and determine if someone is actively trying to game the system or not, if the consensus is that they are then they can be kicked out. So someone that creates a bot to generate karma would likely be pretty easy to flag and remove. The harder it is to tell, the less damaging it probably is to discourse.
Yeah I think it is a bit similar to the whole donuts experiment that is going on.
I think this is generally a good idea as well, Might be an interesting AGP proposal if you want to flesh it out more? Perhaps a ANT reward for accepted AGP proposals?
I don’t really think about it as a “token thing” but rather as some way to allocate authority within an organization. If the goal of the cooperative is to represent the views of active community members in signaling on AGPs, having a mechanism which links weight/authority in the organization to activity in the Aragon community makes a lot of sense.
WRT why do this with the Coop org and not directly with Aragon? I think the main reason is that it is very experimental, likely to have lots of issues that we need to work through. If we did it with ANT it would likely be relatively need to be relatively small amounts (almost like micropayments) which honestly are not a particularly good motivator in most cases. By allocating reputation/points/merit in a separate org that is used for signaling, they are not micropayments, as even basic participation would likely result in meaningful stake in the organization itself.
I think there is significant precedent for membership based organizations to self-govern membership requirements. I wouldn’t really consider this part wildly experimental. I suspect it works just fine up to a certain scale, and then parts of the process would need to be delegated to either some form of automation or dispute resolution process (like the court).