Humans are better than bots at valueing posts

in #steem8 years ago

The new rules for regeneration of voting power have caused some controversy. Here is my take on the topic.

I think the change is probably a good thing for Steem community. It forces users to really think how valuable posts and comments are. Some users have voted everything with full power and it's not a good thing. Users should do that only for great posts that really deserve it and vote less valuable posts and comments with less power.

Why? Because it's fair. If you vote everything with 100 % power, you are basically saying that "nice comments" and "great posts" are equally worthy. But they are not. Just because somebody commented something nice to you doesn't mean that you should allocate big monetary rewards for him. But of course it's a good thing to like what he did, and for that a vote with a little power is a great tool.

Some users have suspected that cognitive load is too much. Can people decide easily how much voting power each post and comment is worth? At least for me it was pretty easy. Apparently I got the voting power slider earlier than some other users, so I didn't have any prejudice about it, I just started to use it. After a day or two it became a habit and doesn't really require much brain power to decide how much power I should use to vote, it's just a gut instinct, really quickly done.

You can use something like this to decide the power percentage for a vote:

80-100: Great post! I really enjoyed reading it!
50-80: Good post.
25-50: OK post or great comment.
10-25: OK comment.
1-10: "Thumbs up", "it's nice that you wrote this", official posts from Steemit.

Greedy people whose first priority is to opimize their curation reward can also vote more freely. They have bigger ammos for their gun, which they can use to vote for posts that have high chances of getting big reward. The rest can be done with small votes and more relaxed attitude, just like using Reddit or Facebook, upvote and like whatever you feel.

So, using a vote power slider is a great thing in my opinion. It will cause fairer distribution of author rewards. Authors who write great posts will get more money. But how we can get people to use it? This is where the new rules come in. They will incentivize users to think more what is actually great content that deserves a full power vote and what is just something nice that deserves a thumbs up.

The main reason why the new rules were made was to fight against bots who harvest curation rewards. With current rules they can afford to vote for lots of posts. Automated bots don't do their voting decisions based on how great the content actually is, instead they just try to guess what post might become popular. The more we have these bots, the worse it is for the platform. They just don't do as good job as humans when curating posts.

If the automated voting bots want to continue what they do with the new rules, they need to decrease their voting power. This means that humans will have a lot more power over the system. A full power vote is more powerful and the bots have to decrease their voting power so the power relationship will change much more favorably for human users.

I haven't seen any better proposals from critics of the new rules to decrease the power of the bots, so I support the hardfork. Although I suspect that something like 10 full power votes per day would be more optimal choice instead of 5.

I haven't had much time to hangout in here lately, so I've probably missed a lot good posts about this topic. Sorry if these arguments are already discussed.

Sort:  

It's quite ironical that @wang agrees with you )

A couple of corrections/observations:

I absolutely agree that the slider is a good thing. However, it has nothing to do with the actual change in daily vote target. Percentage strength voting has always been available through the CLI, and bringing it to the UI would not require a hardfork or a change in the vote target.

Although the stated reason this rule was made was to fight bots, it will effect not only bots, but all active curators. And the main beneficiary of reducing the influence of active curators will be the least engaged members of the community, those that only cast 5 votes or less a day at full power.

The slider has already been available in the web UI, it just isn't enable for people below a certain SP level or something.

My point is that the combination of the voting power slider and the new rule will be better to the whole system. New rule will create proper incentives for users to actually use the slider. It will force users to think better how powerfully they cast their votes. Results will be: 1) Much fairer distribution of author rewards and 2) decreased power of bots.

Yeah, it's true that most active curators could lose also, but I'm not so sure it will be so bad as some have claimed. If the active curators use the slider, they can adjust how fast they deplete their power. If they make sure that they have always enough power to vote for a supergreat post, they can still make a lot of money. After all, they are active so they have the best chances to spot a popular hit before anybody else. And if they are more interested in rewarding good authors, with the new rule they will have more power to do that, only negative side is that they can't do it dozens of times per day.

The slider doesn't force less-active curators to think carefully, they just vote maximum power all the time. In fact that is not a terrible thing for just about anyone to do because of how vote power adjusts to your voting frequency (once your vote power is half consumed your votes naturally become half as powerful even without moving the slider).

The slider does give a fine-tuning tool that people can use, but whether that is really a good thing or not I'm not sure. I think a lot of people just like to vote if they like something without really thinking about it, and that is probably a good thing if you want wider participation and better curation results.

@@ -1,20 +1,21 @@
+%3E
The slider doesn't f
@@ -106,251 +106,396 @@
ime.

  • In fact that is not a terrible thing for just about anyone to do because of how vote power adjusts to your voting frequency (once your vote power is half consumed your votes naturally beco
    +%0A%0AThis is true, but hopefully the new rules make them think a little bit how wise that strategy is. You never know how much time you are going to spend on Steem and how much you end up voting. A quick visit might end up a lot longer, as it many ti
    me
    +s
    ha
    -lf as powerful even without moving the slider)
    +ppens in social media.%0A%0AAdjusting the power is not even a difficult concept, most people have used something similar in video games
    .%0A%0A
    +%3E
    The
    @@ -805,8 +805,139 @@
    results.
    +%0A%0AWider participation maybe... but better curation results? I don't see much good in people voting %22nice comments%22 with full power.

Hmm... something really messed up that comment.

The size of the reward is also good feedback for the author. He then has a better chance of judging what his readers/readers in general want to see from him.

The comments section is good for this, too, in theory, but not everybody likes to leave comments, and just read and upvote, instead.

You got my 100% upvote and it's a good point.

Greedy people whose first priority is to opimize their curation reward can also vote more freely.

Regardless of yours or indeed my views on the hardfork; what depresses me, is this communist view that if you try and make money from Steemit you are greedy.

If someone wants to optimize their curation rewards, that means they are trying to vote on the highest paying content as early as possible. This will reward the content producers, so it is a symbiotic relationship; to call one side greedy is just plain churlish.

Cg

Maybe I should have said it in a different way. English isn't my native language so I sometimes use words that are not suitable to say what I want to say.

By greedy users I meant people who want to prioritize their short term income. Although now that I think it more, it's kind of funny because curation rewards are paid in SP – not really short term. Opposite would be users who vote for content they believe is the most valuable for the whole community (and don't care how much they get rewarded themselves). Of course this will benefit them in the long run when the community gets better. When it gets better, the price of Steem will also go higher and they benefit that way. That has been mostly my way so far, although I openly admit belonging to the greedy camp in some other areas of life.

You bring a valid point to the table of Steemit, one that should always be discussed until there is a solution. Got my Follow. I will continue to follow your pieces in the future.

Thank you for your thoughts on the new voting system. The impact we will see after the implementation. I guess with every new system there are benefits but also challenges. The bots will also change and we will see what the result is.

Great post with a great point. Thanks

Great article and thanks for sharing. Happy to upvote and share this on Twitter✔ for my followers to read and see. Following and looking forward to reading more of your posts. Cheers. Stephen

I think there is room for both bots and humans. The current bots are very unsophisticated, and a good human curator could easily outperform. If Steemit one day scales to millions of users, then a well designed not may end up with the upper hand.

The is potential for both good and bad curation from both bots and humans. Also, it is really unknown what the true outcome of the voting changes would be, because one of the largest variables is how users and bots would adjust their behavior to adapt to the changes. It's one of those things that we wouldn't really know until everything played out.

Coin Marketplace

STEEM 0.19
TRX 0.13
JST 0.029
BTC 57889.17
ETH 3155.19
USDT 1.00
SBD 2.42