You are viewing a single comment's thread from:

RE: Open letter to fellow Investors

in #lif6 years ago (edited)

Vote trading (effectively self-voting) is the most braindead move, yet it is currently the move with the highest rewards on Steem. With the same amount of SP, even the best curation works will go nowhere close to the rewards acquired from trading votes. No sane investor will be okay in buying and holding SP with such an imbalance, where content-agnostic activities that work against the platform dominate actual content/curation activities by a large margin.

There's nothing wrong with trading votes, but a good economy will be impossible without aligned incentives on desirable behaviour. We just need to tweak the code so that curating (directly / indirectly through curation services) earns just about as much as vote trading.

Here's a thought experiment:-

  1. Every vote is money on Steem. Imagine that every vote we give out, we get back 0% of it. It'd be safe to assume that no curation will happen. Because if I curate, I earn exactly nothing while the other guy who's not even curating gets everything.

  2. Now if we get 100% of our votes back for every vote we give out, we can say that curation will at least happen much more often because there's no opportunity cost between any kind of voting behaviour. But there's absolutely no incentive for content creators to do anything at all since it all goes back to the voter. So it's actually the same as scenario (1) above. The effect is symmetrical.

  3. Today in Steem we get about 25% max (average about 13%) of our votes back for every vote we give out, while self-voting or vote trading all the time can bring us back 100% of our votes (or more). So naturally, vote trading /self-voting is the dominating force on the network, quite close to scenario (1).

  4. Now if we get about 75% of our votes back for every vote we give out? Certainly an improved situation compared to (2). Again, we can assume symmetrical effect so it's the same situation as (3).

  5. We can expect scenario (3) and (4) to collapse into scenarios similar to (1) and (2) over time as the profitability gap is too huge between voting behaviours.

  6. So now we're onto the balancing point, what if we get about 50% of our votes back for every vote we give out? Sure one could still self-vote / vote-trading away to get back 100% of their votes, but this middle-point would actually encourage more curation and one may even be able to get more than 50% back by curating well. The point is for folks to vote out more whenever they come across good content instead of just saving up to take it all back. There's also the cost of building a social network that perhaps many are willing to take between getting back 50% (or more) vs 100%. We've suggested tweaks that could make this possible, but so far a lot of the investors are playing mental gymnastics and don't think it's an urgent problem that can be solved. Maybe because by playing the game as it is now, they can easily use Steem's inflation to pay themselves as well. So why change? In any case, we've already proposed the changes in our posts past couple of months.

Sort:  

The fixed rules on author vs. curation rewards always seemed like a mistake to me. As your thought experiment shows there will be a subjective balancing point where the hassle and effort of developing a voting scheme will outweigh its benefits.

But it's subjective and each SP holder will have his or her own tipping point.

So it seems to me that there should be a market in curation. Let each SP hodler determine how much curation value his or her vote is worth. Ideally there would be some sort of feedback loop where the more generous you are with your curation the more it is worth - possibly in visibility terms.

Part of the difficulty with steem is how multi-dimensional it is. Optimizing on one parameter will necessarily de-optimize another.

Isn't changes like these ultimately in the hands on STINC? It seems to be clear curation rewards need a boosting as authors are leaving without being founded and low quality content drives away those who are deciding on if their time is well spent on Steem ecosystem. Serious curation just isn't worth it right now and that is something that would set our social media ecosystem apart from the rest as there's many more coming up.

What if a post would continue to generate revenue even after seven days? Then curation rewards could would continue to flow over time and people would be rewarded for writing useful and timeless posts, rather than knowing that their post might be somewhat useless after seven days.

Generate revenue even after seven days... My thoughts always run in this direction.

Yet presumably it would so dilute the reward pool as to make up-to-the-minute posts much less appealing to the cell phone set. This 7-day limit goes after the NOW Market occupied by Twitter, Snapchat, Reddit and so many others.

Quality, long-term content becomes ultra-niche, niche market, quite a different kettle of fish. But in an EOS-centric universe it might just work out fine. If there is demand, a market will emerge.

True, but we could still let older content be able to gather rewards anyway. The limit has probably turned away a lot of content creators.

"Let older content...gather rewards" I hope so! Since the Mainstream Media confines itself to PURE BS there is a world of citizen journalism just waiting for the newcomers to speak up.

It's something to consider and I guess it might be even required change to capture a bit more professional journalism that take a lot of work to create.

I'd imagine the required work to do the change isn't crazy either, you could just essentially keep resetting the 7 day window internally.

There are many thoughts in it i have.. but what i appreciate the most there are some specific parameter values which can be tried out to get a more balanced system.

I stopped vote selling on Minnowbooster recently but i know there are many who just continue with it and get a lot more payouts. But for me it was a two-folded decision

  1. moral not to support bad actors with my vote which i cannot influence if i sell a vote
  2. economically for the long term success of Steemit because the social aspect of the platform gets slowly destroyed if this is the defacto standard of behaviour

The social and economic factors cannot be separated. If we exploit Steemit or any other future platform it works short term for a few like a pyramid scheme (Steemit is not) and we never ever get to mass adoption because word of mouth spreads like wild fire in the savannah about the anti-social behaviour with massive bot usage.

Regular people are not interested to be part of such systems.. because then it's like a horde of globlins with clubs is just waiting for new victims.. exaggerated view but basically the selfish attitude to attract people just for this purpose to exploit a even bigger reward pool with bots, vote selling and so forth cannot lead to a long term success.

It is the same with every business, if i start as CEO and immediately want a Mercedes and golden faucets but don't even have any product sales or customers.. it won't work.. i'll destroy the first green shoots immediately. I see a similar situation on Steemit currently. The growth of the platform is impeded by this imbalance.

I'm not a iPhone socialist but a long term investor who has always the choice like any other Steemian to use no, a bit or massively bots and services for higher gains but the more we move the balance to the reward robbing or distorting reputation bots it gets a very bad taste and economic desaster in the long term is my opinion.

I think Dan Larimer knows exactly what's going on and will get to the next level with a EOS platform in the near future but Steemit has still potential with the strong community of witnesses and Steemit Inc.

Thanks For Sharing!! Imagine if Facebook paid us!! You've got a new follower!! Hey You got a minute? CLICK on this link, enter your e-mail, and then watch the video!!
http://www.winwithroberthollis.com/pjcswart