Remember, 25% more vote power after HF21

avatar

Would you want Google or Facebook algorithms to be used to uncover content they consider valuable for you to vote on and hide what they consider you shouldn't vote on? Think about it for a little while before answering - especially if you are a content creator.

So?

Oh... but you don't want to downvote...

Upvoting and Downvoting is the algorithm that is used on Steem to uncover valuable content and hide what the community believes should be hidden. So far it has not been very attractive for most people to downvote and as a result, it has created a stigma in downvoting, making it a negative. No one really minds giving a restaurant or a hotel a bad review for shit service though.

What I still believe is the game changer for Steem in the HF21 is the 2.5 downvotes that can be used to order content alongside upvotes. While everyone focuses on what they are losing (10% out of the author pool for the SPS and then going to 50/50 so curators earn more - taking away from authors) they should really be looking on what is being gained.

Come HF21, everyone effectively has 12.5x100 votes a day instead of 10. That is an increase of 25% - 10 up, 2.5 down. I believe this changes things more than the percentage though because downvoted content doesn't get good curation returns.

What this means is that while some people have been aiming to frontrun on content that will get large votes, if those accounts will also receive downvotes later, they will stop voting on them and with the 50/50 split, there s more incentive to curate content that will get higher votes and not get downvoted.

People like to maximize after all and "wasting" a vote on content that'll get downvoted by the community will not be high on an maximizers agenda. This means that those voters are still incentivized to vote, but have to be more sensitive to what they vote on.

If everyone is using their 2.5 full downvotes a day ( I suspect it will be more like 1-1.5 on average), that puts more in the pool and more in the hands of those 10x votes that are incentivized to find content that won't be downvoted so that they can maximize their 50% curation. If it does work this way, it means that the maximizing behavior of curators benefits who?

Oh, the content creators. But, with the convergent curve there will also be incentive to have votes land on content that isn't going to get downvoted and is going to get community support also. This means that there is more chance that good content gets rewarded and when in the spotlight more as bad content is reduced, it could get rewarded even more. And what happens when good content gets rewarded on Steem? oh, it also could get rewarded in PAL or a few other tokens also.

Not only that, what happens is that over time content creators that do consistently produce interesting content will get driven consistently upward and then they become influencers on Steem and attractors of more wanting to be like them. That is actually what an influencer on a social platform does, it isn't to hawk products - it is to attract more users who want to be like them and do the hawking through their own social circles. As I have said, your data isn't valuable, how leveragable your network is.

what happens when there are people on Steem who are truly getting paid well for being here? More people come wanting a piece of it, and that will coincide with the drive of mass interest in crypto. At 10 dollar Steem, my account will be worth 500,000 dollars for 2.5 years digital work. Tell me, how many people on Youtube are in that category? How many people WANT to be in that category? Right now my account is worth 20k dollars. I am holding.

At some point, accounts like mine are going to attract people in just like accounts with 100k followers attract more followers on Instagram, it becomes a point of mass that has gravity and as prices rise at the same time global interest in cryptorises and at the same time global economies fail, Steem is going to be a shining star of an example in socialized distribution and creation of wealth. 10 dollar Steem means that over the last 3 years, there will have been three billion dollars distributed to the community and, still no central owners.

At what point has Youtube given 3 billion dollars to its userbase? How much have they made though?

People really don't seem to understand where all of this is leading it seems and while they argue about what they are losing now, what they are really losing is their ownership of a future position by not participating in the growth of the platform.

What is going to attract people to Steem in the long run is seeing success happen on Steem. Successful bloggers, gamers, application developers and the like and once that is seen and attracts a wide range of new users, they pour in to find that this is much more than a social media, it is an economy that becomes more stable the more participants.

I think that while in the short term HF21 is going to be quite painful for many, in the long-term, the community algorithms start to find their equilibrium and the upvoting / downvoting balance will be both relatively healthy and accepted as part of managing a community where there is no centralized policing.

It is going to be immensely fascinating to see what actually happens in the days, weeks and months after HF21 because for the first time in Steem's history, the management of content will be up to the community and incentivized to maximize upon. I think content producers like myself may be happy once things settle, if they have decent hustle. And if a content creator doesn't have hustle, they aren't likely going to make it anywhere in this world with their content.

Participation is key in a community.
Upvotes and Downvotes.

Taraz
[ a Steem original ]



0
0
0.000
30 comments
avatar

@tarazkp - I understand the intellectual side of your argument. The numbers about downvoting makes sense. The purpose of downvoting makes sense. The long term benefit of downvoting makes sense.

However, I have a negative emotional response to downvoting because it is confrontational. People are afraid of "flag wars" but more importantly they are afraid of a big stakeholder destroying a little guy.

Will the community rally behind this type of misuse? Maybe. That is yet to be seen. Seems like this will create a situation where you police the accounts that are smaller than your own, but not have enough power to take on account with a larger stake.

Just my 2 cents!

0
0
0.000
avatar

However, I have a negative emotional response to downvoting because it is confrontational.

That is up to you.

People are afraid of "flag wars" but more importantly they are afraid of a big stakeholder destroying a little guy.

This happens now and because there is no incentive to help, no one helps.

Seems like this will create a situation where you police the accounts that are smaller than your own, but not have enough power to take on account with a larger stake.

I think there might be a few surprises.

0
0
0.000
avatar

I've been saying for a long time that the average user will not downvote for the reasons you're mentioning. If any solution is dependent on that it will fail. It is absolutely confrontational and people don't come to a content site/service to fight with people. They want to be entertained and have fun, not attack people or be attacked.

The only hope for things like the EIP is at the whale level. If whales flag other whales stopping the large scale abuse, that could make a positive difference. Upvoting and downvoting to decide what is trending will never work when you mix in a financial incentive. We need to know when to accept that something just didn't work and focus on solutions that might work.

Some ideas:

An algorithm that take a lot more variables into consideration.

Something that maybe looks at page impressions, comments, upvotes, resteems, how quickly the upvotes came in, at what point in the posts life the votes came in, if known bots voted on the posts, etc. All to determine if something is ACTUALLY trending.

or something even simpler like what Steempeak has done, which is just make the trending page a list of posts curated by curation groups. This isn't ideal, but it's MUCH better than what's happening now.

If anyone sat around and thought about it, there's lots of potential solutions. People are dogmatically attached to this idea of downvotes as a means of sorting the content, and I think that just needs to be abandoned.

Observe user behavior and design features that empower them. Don't try to bend the user to your will, humans just don't work like that, they'll just leave.

0
0
0.000
avatar

I do believe a trending page with better algorithms would work much better to get the good content up there, than downvotes coming from individuals. Individuals are subjective. What each one likes varies and it only needs one person with a big account to override the views of a lot of smaller accounts. Whereas, if the algorithms go more on activity, it doesn't matter what the post is earning, it matters what is being promoted by most of the community.

The upvote already allows for a certain amount of say in whether or not you agree with what someone has posted and its value. If you don't like it, you don't award it anything, if you really like it, you can give a 100% upvote, or if you like it, but feel it's already rewarded enough, just give a token 1%.

Posted using Partiko Android

0
0
0.000
avatar

I've been saying for a long time that the average user will not downvote for the reasons you're mentioning.

The average investor?

The only hope for things like the EIP is at the whale level. If whales flag other whales stopping the large scale abuse, that could make a positive difference.

The most active stake on the platform by far is the orcas and dolphins, they are also the ones who will benefit the most from a price rise. The 30 odd whales here can sell now and do well enough, nearly everyone else needs the price of Steem to go up.

An algorithm that take a lot more variables into consideration.

And then we have google.

Something that maybe looks at page impressions, comments, upvotes, resteems, how quickly the upvotes came in, at what point in the posts life the votes came in, if known bots voted on the posts, etc. All to determine if something is ACTUALLY trending.

This isn't just about trending, it is about the entire ecosystem. This means that authors are only a part of the ecosystem, not the entirety. If this isn't a place where developers, investors and consumers want to be, the authors can still create but, they get nothing.

0
0
0.000
avatar

The average investor?

Yes, the average user, which won't be an investor. Is this an investment platform? It's a blogging platform that a user can choose to invest in, and will eventually hold a stake in if they continue to use it. The thing that gives it value, which would make it attractive to investors is USERS.

Just think about it like this, imagine all we had was Steemit Inc. in it's current state and NO other developments, but we had 5 million daily active users

or

We have SMT's, Communities, a super scalable solution, we fell on the perfect economics balance, but we have 50 Daily active users.

Which one is more valuable?

The most active stake on the platform by far is the orcas and dolphins, they are also the ones who will benefit the most from a price rise. The 30 odd whales here can sell now and do well enough, nearly everyone else needs the price of Steem to go up.

I'm counting Orca in "whale" in what I said previously, but anyway everyone wants the price to go up, the question is what's the most effective way to increase the price? I'm for the EIP, but I don't think it will improve the price. I'm hoping it just stops some abuse and improves the internal morale, and it would be great if distribution improved, but make no mistake, no one outside of Steem cares about this.

And then we have google.

The bad thing about Google isn't the fact that they use algorithms. We aren't Google if we use algorithms, that's crazy. The bad thing about Google is they're collecting data from you, typically without your consent and selling it to advertisers and making huge sums of money from it without giving you anything. The bad thing about Google is that they are mining YOU for value. If we use an algorithm to get a more accurate picture of what is actually popular that just makes us functional, and capable of doing the thing we set out to do.

0
0
0.000
avatar

Because there are potential financial rewards at stake, I think it's still risky to downvote. The main reason is that even if you are voting something, the person who wrote it might see it as you downvoting them. With all the flag wars that seem to get started, the risk seems high. Just my opinion. I mostly try to keep my head down and just talk with my friends.

0
0
0.000
avatar

They will likely take it personally, because of the current system, not because it is. 10% of the votes on reddit are negative, it is part of the system there though so no one pays it much heed.

0
0
0.000
avatar

I haven't spent much time on Reddit, so I didn't know that. I guess it makes sense. If you're used to getting a few downvotes with each post, then each one isn't as traumatic. I remember my first ones and even though they were spam and didn't affect rewards, it was still jarring. Hopefully things will go down like you said if the change goes through.

0
0
0.000
avatar

Up voted and resteemed eith great respect. But really i could not understand to where you are going with this post...
All what i like to do is to upvote and resteem.
Cheetah is attacking my posts here ... so as i am only just operator of this account till its real owner my x-boss is coming back to control it... i will try to minimise the quantity of posting and to help others by resteeming and upvoting.
Well all in all good...
By the way... please send my best wishes and give her bbbbbbiiiiggggg kiss @smallsteps.... i am very proud of her....she is enjoying reading the pigs stories as seen in the 2 films....OMG... she is really fascinating. ..
God bless her
Bye for now as cheetah will come and down vote this comment as usual

0
0
0.000
avatar

Cheetah is attacking my posts here ... so as i am only just operator of this account till its real owner my x-boss is coming back to control it... i will try to minimise the quantity of posting and to help others by resteeming and upvoting.

To me it seems like you have a very strange owner of the account to be targeted by cheetah. Should have perhaps started your own?

0
0
0.000
avatar

I have 2 more accounts in my name plus 1 account on another platform.
But this account is for my x-boss whom he was and still very kind with me.
I can not explain my honoured feelings for helping him. He is really a genuine clever man but still in hospital .
Thank you

0
0
0.000
avatar

If everyone is using their 2.5 full downvotes a day ( I suspect it will be more like 1-1.5 on average), that puts more in the pool and more in the hands of those 10x votes that are incentivized to find content that won't be downvoted so that they can maximize their 50% curation. If it does work this way, it means that the maximizing behavior of curators benefits who?

It maximizes the benefits for those who actively go out and seek things to downvote.

In that single statement is why the rest of your post doesn't matter. It's a fantasy. It's a fantasy build on observations about human behavior which no one believes because they are simply not true. I won't say that this is a lie because I believe that you believe. But it's wrong. Elementally.

The core problem is not a technical issue but instead an utter misunderstanding of how and why people use social media platforms. They do not go to a social media platform in order to find or be exposed to content that they don't want. They go to a social media platform in order to find, consume, and enjoy the content that they believe should be rewarded. At worst, they get exposed to content that they're not interested in.

Following, upcoming Communities, the underlying idea of "curation," it's all about communicating to the system your preferences and expecting to get more back. When people receive content they're not interested in and in fact think is overrated they don't immediately think "oh, I need to downvote this," they think "this social network is a piece of crap; it's giving me stuff I don't want." They don't see it as their responsibility to do things for other people at their own expense.

We been told over and over again that people don't downvote because it's too expensive in terms of voting power – and in part that's true, but they don't downvote for much more important reason:

They see content which requires down voting as an elemental failure of the underlying platform.

And that's because it's true.

So, let's ask the question from the other side of the coin. Who is it that will have the most use for downvotes?

People who go out of their way to seek content for down voting. People who have been involved in the Whale Wars before and have consistently used downvote vote power to abuse and harass others. People who take it on themselves to go and police others. Bot owners who realize that downloading content which is associated with their bots effectively provides a bit more of the reward pool for themselves.

The mechanic rewards most the people who are most likely to abuse it.

That's why it's a bad design and why ultimately it will lead to less creator reward and less creator activity on the platform. Best case, it leads to another Whale Wars spike but the participants are too busy going hammer and tongs at each other to notice any of the rest of us. Worst case, the creative space on the STEEM blockchain is made to that much more vulnerable to someone at random coming along and bumping some of your already scant reward for reasons absolutely unrelated to your content.

It's very difficult to get excited as a creator or as an analyst about any of this mechanism. I wish more people were talking about how questionable it is from a position of process.

0
0
0.000
avatar

The core problem is not a technical issue but instead an utter misunderstanding of how and why people use social media platforms. They do not go to a social media platform in order to find or be exposed to content that they don't want. They go to a social media platform in order to find, consume, and enjoy the content that they believe should be rewarded.

This is not just a social media designed to leverage the behaviors of people who " find, consume, and enjoy the content that they believe should be rewarded" because on social media like you mention, they have little direct control over what content gets rewarded.

So, let's ask the question from the other side of the coin. Who is it that will have the most use for downvotes?
People who go out of their way to seek content for down voting. People who have been involved in the Whale Wars before and have consistently used downvote vote power to abuse and harass others. People who take it on themselves to go and police others. Bot owners who realize that downloading content which is associated with their bots effectively provides a bit more of the reward pool for themselves.

"In that single statement is why the rest of your post doesn't matter. It's a fantasy. It's a fantasy build on observations about human behavior which no one believes because they are simply not true. I won't say that this is a lie because I believe that you believe. But it's wrong. Elementally."

0
0
0.000
avatar

This is not just a social media designed to leverage the behaviors of people who " find, consume, and enjoy the content that they believe should be rewarded" because on social media like you mention, they have little direct control over what content gets rewarded.

That this is also completely and incontrovertibly untrue. Users of functional social media platforms have complete direct control over what content gets rewarded – via thumbs up, +1, Like, or one of the other dozens of system-signaling mechanisms which are built into pretty much every platform currently popular. That is exactly, definitionally content getting rewarded – and with a more active positive feedback loop because interacting with that content and asserting to the system that they like it generally means they'll see more content like that, specifically targeting their preference profile.

You appear to be falling into the traditional Steem trap of imagining that the only way that people get rewarded is with pay. It's not the first time and it won't be the last, but not recognizing that there are other values at play in the space which are part of why people interact with social media platforms will not help you make good decisions when it comes to improving people interacting with things that you care about.

Hell, the last year of statistics falling out of the Steem blockchain about what and how people use the platform and the dwindling number of people choosing to do so should have been indicative.

"In that single statement is why the rest of your post doesn't matter. It's a fantasy. It's a fantasy build on observations about human behavior which no one believes because they are simply not true. I won't say that this is a lie because I believe that you believe. But it's wrong. Elementally."

You could have, at any point, decided to actually provide evidence or at least argumentation that support your claims – because I'm ready to. You could have even cherry picked any of the points that I made and argued against it.

But you couldn't even be bothered to do that much, and that's kind of embarrassing.

But I'll bite. We'll play.

Explain to me how after the oncoming hard fork people will have more downvotes available and simultaneously be exposed to more content which they believe deserves downvoting? Especially since we've been promised Communities and other application-based siloing of content, which has always resulted in people being exposed to more content which they like and less content which they don't like. Is a normal user expected to increase the number of posts and comments that the platform intends that they downvote?

Of course not. That would be silly.

So who is going to be using this larger number of downvotes?

People looking to use those downvotes.

Who is going to be motivated to use those downvotes?

People who have an inherent motivation to not just reallocate the reward pool but deliberately directed away from other people and content that they believe does not deserve reward. People who want to engage in petty harassment of other people. Bought owners who want to make it less profitable for other people to run competing bots. A couple of the rare active whales who are deliberately targeting the krill-eating of competing whales.

This isn't hard. This is just looking at people interacting in a social environment, looking at the tools they will be handed, and working out the obvious.

Now, if you would like to make any kind of coherent argument in defense of your position that this is going to be a net gain for content creators, one that actually makes sense and is in accordance with what we actually see happening on not just this platform but every platform, I'm certainly ready to hear it.

As is, you haven't made an argument based on what people are actually doing, or even the financial environment that crypto-commodities in general find themselves in, but you've made unwarranted assertions as a crypto cultist. That is not a reason to believe you. On the contrary, you have literally made the assertion that there are "still no central owners" of the Steem blockchain as a service. We both know that's just not true, whether you take it as meaning a controlling interest in the amount of commodity available in the blockchain as a voting force or a more concrete level of control of the day-to-day operations and direction of the platform.

So there you go.

I'm willing to point at the behavior that we have seen, actively, on the Steem blockchain for the last several years and use that as evidence to support my contention that your belief is an utter fantasy which does not make sense in the context of what we can actually observe.

Your turn.

0
0
0.000
avatar

Explain to me how after the oncoming hard fork people will have more downvotes available and simultaneously be exposed to more content which they believe deserves downvoting?

Personally, I'll visit the page linked as 'Trending', I'm sure i'll find immediate exposure to content I believe needs downvoting.

0
0
0.000
avatar

Personally, I'll visit the page linked as 'Trending', I'm sure i'll find immediate exposure to content I believe needs downvoting.

You know, I'm not going to lie. I can't tell if this is elegant sarcasm or heartfelt belief.

Poe's Law strikes again!

Oh, it's you, Abh. It's elegant sarcasm. For a moment there, I was really concerned.

(Which neatly outlines part of the problem of the whole Trending system as a whole, especially when it comes to positive feedback loops and engagement, but that's a discussion for another day.)

0
0
0.000
avatar

You appear to be falling into the traditional Steem trap of imagining that the only way that people get rewarded is with pay.

if people are here for what doesn't include financial incentive, that upvote button works much the same. If they want algorithms that filter content, they can be built in multiple ways, get a dev. Some front ends already do this. use them.

Now, if you would like to make any kind of coherent argument in defense of your position that this is going to be a net gain for content creators, one that actually makes sense and is in accordance with what we actually see happening on not just this platform but every platform, I'm certainly ready to hear it.

you see PAL? that isn't affected by Steem rewards. Content creators can be there and even buy in there without relying on anyone with Steem for their earnings. However, PALnet is build on Steem, using Steem earnings and by people who are interested in developing and investing into Steem. As a content creator, that gives them a space to do as they please (they could also build their own and tokenize it) and that is a net benefit of Steem investors and developers for content creators. When there are hundreds of such experiences, built on Steem by Steem holders and investors, content creators will benefit heavily. Not many redditors, facebookers or Instagrammers did that much on those platforms until the platforms were built.

People have been on those other platforms for years doing it for likes... this isn't Facebook. It isn't Instagram. It isn't Twitter or any other random centralized platform. this is Steem. Perhaps you need to review what Steem is for you, and realize - that is only your opinion. If people believe your opinion is worthy, they will follow in your footsteps perhaps. Lucky you.

0
0
0.000
avatar

if people are here for what doesn't include financial incentive, that upvote button works much the same. If they want algorithms that filter content, they can be built in multiple ways, get a dev. Some front ends already do this. use them.

No. It doesn't matter how many times you say so, it does not "work much the same" to upvote content here as upvoting content elsewhere. The only thing they have in common is that they send a feedback signal to the original poster that says "I like this." But after that, the world is very different.

On platforms which are succeeding, that signal is used to help identify some sort of description of things that they like and the system can be used to find more of that or creators who do more of that. You have a positive feedback loop of the machine helping find things "more like this."

On the Steem blockchain, you have some magical numbers twiddling in the back which means that maybe, possibly, the creator will get a fraction of a cent which they will in turn have difficulty spending and the general social order will yell at them for doing so, some people who happen to be on at the right time and judged that the post will be uploaded by other people for criteria which may have nothing to do with the actual content itself will also get a few fractions of a cent for much less personal, intellectual, or effortful "curation" – because that system is still almost tailor-made for machines rather than people to do, and the average use case will see absolutely no effect from their interaction at all.

And God forbid that the topic, field, or person that you are or often interact with is on the outs with someone with a bit more stake than you have because now they'll have more free value-eroding downvotes that cost them literally nothing and cost you more from a smaller pool.

Yes, this is a brilliant idea.

you see PAL? that isn't affected by Steem rewards. Content creators can be there and even buy in there without relying on anyone with Steem for their earnings. However, PALnet is build on Steem, using Steem earnings and by people who are interested in developing and investing into Steem. As a content creator, that gives them a space to do as they please (they could also build their own and tokenize it) and that is a net benefit of Steem investors and developers for content creators. When there are hundreds of such experiences, built on Steem by Steem holders and investors, content creators will benefit heavily. Not many redditors, facebookers or Instagrammers did that much on those platforms until the platforms were built.

I don't see PAL. No one sees it. It doesn't have any kind of presence.

But let's take it just like you said – all you've suggested is that an entirely different system based on a new token with the caveat "and if they don't like it, they can go create their own," is going to reward content creators better by – magic? Sorcerous SMT's don't just make things happen. If it's based in Steem, it's based in a POS system, and it very much a problem that stake, and its distribution in the population, is part of the problem.

"When there are hundreds of different niche experiences based on the Steem blockchain, content creators will benefit heavily." Somehow. By magic. Because the value of the commodity will go up because – you will be able to buy so much more with it? A cynic would point out that more creators in the space will devalue the commodity, forcing the reward pool to be spread out more thinly. The basic assumption is that the value of Steem itself as a token will increase but that doesn't follow at all. We've already observed, on multiple occasions, that the value of Steem doesn't actually track with the value of the content on the platform.

That's because they are utterly unrelated.

People have been on those other platforms for years doing it for likes... this isn't Facebook. It isn't Instagram. It isn't Twitter or any other random centralized platform. this is Steem. Perhaps you need to review what Steem is for you, and realize - that is only your opinion. If people believe your opinion is worthy, they will follow in your footsteps perhaps. Lucky you.

Yes, heaven forfend that we actually do something or consider something which is successful. Let us never look at the user experience and consider whether a sane person would invite themselves in for it. Let us not observe our own history and double down on the things which were the most broken. We certainly must not consider what the most active parts of the platform are and assess whether or not they will be doubly rewarded and whether or not the majority of users think they should be.

"What Steem is to me" doesn't matter. It's not just my opinion, it's my use case. I'm trying to think past that, to imagine better ones, to figure out what users want and actually provide them with something that fulfills their needs – because that is the sort of thing which lives under the constant threat of potential success. I realize it's a mode of thinking which runs counter to "well, it's cryptocurrency, so everything is fine!" But it is the kind of thinking that has the chance of actually understanding why things fail, analyzing their failure modes, and building mechanics which work better.

I would hope people did not follow in my footsteps. Took inspiration from me, sure, but don't follow me. I'm going in directions I find interesting. They might not be as interested in what I want to see and do as I am. Now, I encourage people to apply their own critical thinking facilities, to look around them at what's available, and make choices that work for them – and for them to truly understand what it is that they want and what it is that they want to experience.

If other people read what I say and come to those reasonable conclusions, yes – I'll be very lucky.

0
0
0.000
avatar

I struggle to make it through a paragraph of your writing style so, I don't think that I will be one of the people taking much inspiration here, but there might be some who make it through. Good luck!

0
0
0.000
avatar

Ok, I remember... and this is long decent hustle indeed. Any snacks?

0
0
0.000
avatar

Oh, I have to do a snax update. thanks for the reminder :D

0
0
0.000
avatar

Your theory is sound, just like the theory of communism, socialism and democracy is sound. It remains to be seen as to whether it will actually work that way when implemented.

Personally, it won't make me want to use downvotes any more than I already do, which is pretty much never. My upvotes do the job for me. If I don't think something should have rewards, I don't give it any and if I think it's under valued, then I'm going to give it 100% and might even resteem it.

I think a trending page with different algorithms would do much more than making downvotes more attractive. A post like this one gets a lot of conversation and activity going, so if that activity were included in the algorithm, then it would be up there. Some posts are great, but its hard to say anything more than, "great post!" so if the algorithm can also take into account the amount of upvotes it's getting and from which sources (ie. the quality curators) then that could still be up there.

Whether or not you like the traditional social media sites, when it comes to the trending algorithms, they're doing it right. I landed on Reddit for the first time a month ago, and got drawn into the recommendations, with all the conversation that was happening there. It's not a conversation I wanted to get into (roasting isn't my thing), but it was addictive. So I simply put a block on notifications for it. I can't even remember how I did it, it was that simple.

Posted using Partiko Android

0
0
0.000
avatar

Your theory is sound, just like the theory of communism, socialism and democracy is sound. It remains to be seen as to whether it will actually work that way when implemented.

Which is why it needs to be trialled. Since all econoies have failed so far "communism, socialism and democracy" and we are trying to do something different and succeed - it is going to take trial and errors. luckily, this is possible on Steem if people stop thinking like they live in a government of "communism, socialism and democracy" and act the same.

Personally, it won't make me want to use downvotes any more than I already do, which is pretty much never. My upvotes do the job for me. If I don't think something should have rewards, I don't give it any and if I think it's under valued, then I'm going to give it 100% and might even resteem it.

keep an eye on abuse at least.

I think a trending page with different algorithms would do much more than making downvotes more attractive.

It isn't enough because authors want to get paid from the pool but a great deal of the pool is going to crap that will should never appear anywhere near trending. Hard to advertise success if the content that should get rewarded, isn't.

Whether or not you like the traditional social media sites, when it comes to the trending algorithms, they're doing it right.

The way Steem interfaces do this is up to them, the way Steem does it is up to us. With the incoming tokenized experiences, authors have more ways to earn than ever yet, everyone still wants it out of the Steem pool. How many bought into PALnet in the hopes that it would turn into a place for content producers? about the same number who bought into Steem originally probably. Not many. Same processes over and over, many people don't want to invest in their future but still want to have support for it.

0
0
0.000
avatar

It isn't enough because authors want to get paid from the pool but a great deal of the pool is going to crap that will should never appear anywhere near trending. Hard to advertise success if the content that should get rewarded, isn't.

Yet with different algorithms, that crap wouldn't reach trending. Then, as less of it reaches trending, then the drive to get the price high enough for trending disappears, because its not the price that's dictating it.

The way Steem interfaces do this is up to them, the way Steem does it is up to us.

Yet the Steemit interface is still often the first interface that people encounter. Can people now sign up for an account via another interface without actually having to go through the Steemit one at all? If so, that would be good. At the moment, most people find out about the interfaces after starting on the Steemit one. How many have turned around and not even bothered getting an account after seeing the Steemit trending page?

I'm most curious how the new curve will affect rewards for comments. Will they mostly all be dusted? What do you think?

Posted using Partiko Android

0
0
0.000
avatar

then the drive to get the price high enough for trending disappears, because its not the price that's dictating it.

it is not their drive to reach trending (other than some of the app announcements), their drive is the profits.

Can people now sign up for an account via another interface without actually having to go through the Steemit one at all?

Yes, there are a number who are creating accounts nowe including @steemmonsters and partiko i think.

0
0
0.000
avatar

I remember when i got here the thing was post /spam twice a day etc to make $. What a stupid concept! Some can do it but i see a lot that probably shouldn't.

Now they will be able to take a breather and enjoy writing that one really cool article whilst reading / curating others really cool article and still win!

0
0
0.000
avatar

Now they will be able to take a breather and enjoy writing that one really cool article whilst reading / curating others really cool article and still win!

While the community is small and niche, this is pretty much what has happened on smoke.io over the last year. People post less, upvote more.

0
0
0.000
avatar

I still have my doubts though...particularly concerned about how this would affect new users...it is a waiting game though

0
0
0.000
avatar

I’m guessing most people won’t be using their downvotes initially(I probably won’t) but only time will tell. Hopefully the changes will be good but I’m worried the curation changes will end up allowing big accounts to snowball while small accounts stagnate(a common social media issue).

Posted using Partiko iOS

0
0
0.000