You are viewing a single comment's thread from:
RE: Does Steem have a Tragedy of the Commons dynamic?
Some interesting ideas which if implemented, I can see could successfully improve the Steemit ecosystem. A few thoughts that spring to mind are:
- I don't think Steemit will ever downvote again no matter how bad the situation on the blockchain gets. Which means we're reliant upon another user with enough clout to be able to clear out $100 worth of rewards which at the moment doesn't exist. @ac-cheetah has a good "downvoting trail" that could take up this mantle but it would require far greater backing than the 120,000 SP that it currently has.
- The idea of a "decay" in voting weight as users become more powerful is a great idea. But - the 1st generation bidbots won't change so any 2nd generation bots would work until users reach the "optimal" point and then users would move back to the original voting bot. Or just use both - users of TipU, etc. buy votes from every service going.
- The ranking of posts by something other than payout is something that I'm particularly interested in and something that I'm hoping to implement in my new front end. Primarily ranking posts by comments and quality of comments as an indicator to their quality.
There are times though when I feel that Steemit is already past the tipping point of where it can be saved. I'm sure I've said this before - 75% of all power owned by the top 1,000 users (i.e. most users with 10,000+ SP) is delegated. Mostly to voting bots. Unless these 1,000 users change their behaviour (which is unlikely), the battle's already lost.
Last night, @cmp2020 was showing me an AI course that he purchased from udemy for his AI curation project. One of the courses I saw on his screen reminded me of this part of your comment:
It was: Recommender Systems and Deep Learning in Python
Maybe that'll eventually be useful for you, or maybe you can find some free references on the topic by searching for "Recommender Systems". That's the first time I had seen them named that way.
On this:
I don't think a tipping point has been crossed, but with the size of some of the bidbots, I'm getting concerned that one might be approaching. For Steemit's stake, I'm aware of somewhere around 38M SP. One of the bidbots is controlling somewhere around 32M. Unless Steemit is controlling some stake that I'm not aware of, it seems like we're getting close to a point where even Steemit would need to leverage a decentralized solution.
Thanks for the links - I'll take a look when I find some time 🙂
o1eh occassionally shares some statistics on the use of bidbots which I find interesting. He's got combined bidbots having more than double the power of Steemit although I don't know if he includes mister-delegation or the steemit account within this.
I don't expect Steemit to ever ban the likes of UpVu through fear that the 30+ million Steem delegated to them will leave the platform. If Steem is to remain a purely "blogging" currency, then I think that our best hope is to have a front-end that makes these things disappear so that new users don't know that they exist and don't continue to make them more powerful. And accept that they're there, draining the reward pool and propping up the price.
I agree on this. I don't really want to see them banned, anyway. I'd just like to see them encouraged to get better at curating. You're probably right that if no one's going to challenge them, and there are no relevant competitors, the best approach is to just make them less prominent in the front ends.