What will it take for Steem to reach "the masses"?
What will it take for Steem to reach "the masses"?
Introduction
Pixabay license, source
In this post, I will write out some thoughts about the longstanding question of how to bring Mom and Pop Mainstreet to Steem?
Before I get into that, though, I want to point out another question that we first need to ask ourselves, "Do we really want 'the masses' publishing on Steem?" Certainly, we would like everyone to have a Steem wallet and to do transactions with STEEM and SBDs, but as far as social media - a quick trip to Twitter, Instagram, Reddit, or Facebook reveals that there is a great deal of low-value content whenever "the masses" are present. In contrast, there has always been widespread messaging that we only want "high quality content" on Steem.
I would argue that a drive to produce only "high quality" content is at odds with a drive to "bring the masses".
So, before we talk about "how" to bring the masses to Steem, we need to decide "whether" to bring the masses. Personally, my answer to this latter question is, "yes", so I'll move on with the topic. Anyone who says, "bring the masses", however, needs to recognize that it will make Steem into something very different from the ivory tower platform that some people seem to envision.
Against that backdrop, I have come up with seven topics that I think might be useful in order to bring the masses. I'll list them here, and then move on to write a few sentences about each. At a high level, here is the product of my recent brainstorming on the topic:
- Improved ranking of posts
- Ability to use with little cognitive effort
- Custodial services
- Single-language presentation of content
- Engaging front ends
- Rewards for engaging in dialog, not just top-level posts
- Services that provide passive income based on staking
Improved ranking of posts
As a social media system, at its heart, Steem is just a ranking system. People post content, and the blockchain ranks it according to consensus. I don't think it will surprise anyone if I suggest that we could be better at this. I spend a lot of time thinking about curation, so this will almost certainly be the longest section of the essay.
In short, it seems to me that there are two potential ways to change our ranking capability. (i.) Algorithmic changes; and (ii.) Behavioral changes - which, in turn are driven by incentives and persuasion.
So, we need to measure the existing ranking system and then drive improvement. How do we measure it? I'll suggest that one way to look at it is that "if post A pays more rewards than post B, there should be some sense in which post A is 'better' than post B.". And how do we know if that's the case? My suggestion is that we can compare it to one or more alternative ranking systems. Here are three candidates:
i. Measure post rank by the rshares value of the median vote.
ii. Measure post rank by the total post value minus the highest valued vote.
iii. Measure post rank by the number of votes.
Then, we can compare those metrics against the one Steem is using: If all three alternate measures suggest that post A is ranked higher than post B, then we would hope that the reward payout would be ordered that way, too.
Having measured it, what's next? How do we effect change? I have come up with several possibilities:
Algorithmic changes
If we consider curation as a ranking system, then we quickly realize that an overvalued post is just as harmful as an undervalued post. In fact, this may be one of our biggest obstacles. One answer to overvalued posts is the downvote, but that comes with its own set of problems. I feel like I have beaten this to death, but I think an algorithmic solution is needed that will encourage voters to self-regulate. I have suggested a Steem curation rewards distribution that is modeled after a 2nd price auction. Other possible algorithmic changes might also be found.
Macro-curation vs. Micro-curation
It occurs to me that there are two levels of curation. I'll use the terms micro-curation and macro-curation to describe them. Micro-curation is looking at a post and voting to reward it based on its quality - in isolation. Macro-curation is looking at a collection of posts and voting to place them in an order that attracts readers. This collection might be a single community, a tag, or the website Trending page.
To improve this, I imagine a tiered system where minnows focus on day-1 voting, dolphins focus on day-2 voting, and so on. Yes, the larger stakes voters would be sacrificing curation rewards, but the hope is that the value would be replaced with gains on the value of their staked SP. Micro-curation happens fast, macro-curation takes longer. So, to encourage this, it might make sense to bring back the 2-tier 30 day payout window?
The reward for micro-curation is curation rewards. The reward (in theory) for macro-curation is a growing audience and a rising Steem price. Which brings us to:
Curation tools
It would be useful for high-stakes curators to have tools available that can tell them which posts are currently ranked, "out of order". This could be done by comparing the current payout against any/all of the above alternative ranking measures.
Curation rewards
Finally, lets talk about rewards to curators. There are two things about Steem-Power incentives that are fundamentally upside down. As investment and staking goes up, the interest on SP goes down and so does the ROI percentage of curation rewards. This works in exact opposition to the "network effect", where the more you have, the harder it becomes to attract more. I don't really see a way to avoid that, but what we can do to increase curation incentives is to increase the rewards percentage for curators.
The current model is that 50% is supposed to go to authors and 50% is supposed to be split among curators. However, with SBDs consistently at artificially high prices, authors are currently raking in somewhere around 80+% of a post's payout. A naive solution to this imbalance is to increase the curation percentage to 65% or 75%. A more sophisticated approach is to let the authors set the curation percentage at posting time and use it as a lever to compete for votes. (We should not repeat the Golos mistake by letting authors adjust the curation percentage in between posting time and payout time though.)
I know that authors will be resistant to losing a percentage of rewards, but if we believe that it will raise the price of STEEM, it really should be a win-win for authors and curators. In principle, by increasing investment, authors would receive a smaller percentage of a bigger number, and the increase in value could more than offset the loss in percentage.
(Also, it might even benefit authentic authors by making abuse rings less profitable.)
Ability to use with little cognitive effort
Beyond ranking of posts, another way to attract more people is to make sure that the platform is easy and fun to use with low amounts of cognitive effort. Most people are not bloggers and probably have no desire to be bloggers. If we're only recruiting people who are willing to spend 1/2 hour to multiple hours on a post, we're missing a lot of people. There need to be use cases for people to post with little effort and spend a higher percentage of time on actual engagement.
This is why I have been a strong supporter of @famigliacurione's SteemLinks idea. If we want Ma and Pa Mainstreet here, we need to give them more options like that.
Custodial services
Yes, I know, "Not your keys, not your crypto", but the reality is that Ma and Pa Mainstreet do not want to deal with the complexity of 4 separate keys, recovery accounts, and so on. They have a hard enough time just remembering a single password. There are trade-offs. If someone wants to manage their own keys, they should be able to, but I would also like to see the emergence of a custodial service that can do the key-magic transparently, and let a person interact with the blockchain with just a password or standard 2-factor authentication.
Single-language presentation of content
If I'm a little bit curious about Steem and I visit steemit.com and I see a page full of foreign language posts, with no way to find the English ones, I'm gone. Period. End of discussion.
If we want to recruit, "the masses", there needs to be a fast and simple way to filter posts to just the visitor's preferred language.
Engaging front ends
People don't go to Facebook and Twitter for rewards. They go there because they enjoy it. If we want to attract Ma and Pa Mainstreet, Steem needs to be the same. The reality is that there will never be enough rewards to support everyone as a business, and that shouldn't be the goal. Like Youtube, there will be some high earners and some who collect little or nothing. To keep the people who collect very little, the platform needs to be as fun as possible. For most users, if Steem is to evolve into a large scale ecosystem, rewards should be just an afterthought.
Rewards for engaging in dialog, not just top-level posts
In a sense, I guess this is back to the topic of ranking and curation from the first section. But I'm separating it because the first section had top-level posts in mind. In this section, I'm thinking about replies. Basically, we need to find ways to ensure that engagement is rewarded at the reply level. This also ties in with the creating the ability to participate without large amounts of cognitive effort.
I am not a reddit user, but I have come across many of their threads, and what distinguishes them - to me - is not the top level post, but the conversation that develops around it. If Steem's curators are almost exclusively rewarding top-level posts, we're discouraging this sort of organic activity.
One quick way that curators can already find conversations to reward is to take a look at the replies on posts in the #penny4thoughts category and the SteemLinks community.
In the past, the suggestion came up to partition the rewards pool into separate sections for post-rewards and reply-rewards. I don't have a firm opinion on this but it might be worth bringing the idea back into the discussion.
Services that provide passive income based on staking
As I noted earlier, the more investors we have, the worse the ROI gets. That is the mechanism that is designed into the blockchain. It was done for good reason, but it's an incentive structure that works against increasing investment. Also, not every investor wants to participate in posting initiatives like club5050. We need ways to attract investors who just want to park their money and forget about it for 20 years.
I don't have a lot of good ideas on this front, but I think this is an important topic that the community should be thinking about.
Conclusion
So there you have it, my 7 ideas for bringing the masses to Steem. In real estate, they say that the important things are "location, location, location". In Steem, I think the important things are "curation, curation, curation".
As the single largest stakeholder, I think Steemit has been using the steemcurator and booming accounts to good effect for both micro-curation and macro-curation in the various communities, but there is always room for improvement, and we can all look for ways to improve things, instead of just passively depending on a single stakeholder.
In the spirit of continuous improvement, here is a restatement of topics from my recent brainstorming activity:
- Improved ranking of posts
- Ability to use with little cognitive effort
- Custodial services
- Single-language presentation of content
- Engaging front ends
- Rewards for engaging in dialog, not just top-level posts
- Services that provide passive income based on staking
What do you think about these ideas? What are your ideas for bringing "the masses" to Steem?
Thank you for your time and attention.
As a general rule, I up-vote comments that demonstrate "proof of reading".
I love the thread b/t you and @danmaruschak.
To me, the most illuminating thing pointed out there was:
That's indicative of a more general issue, which is just that there is too little development happening (not just code, but business, marketing, community, operations.) I know there is a beginner community, a few active nodes, etc. etc. But in the end, there really is too little substance to the whole shebang.
That being solved would at least allow for a chance that some mechanism for managing over- and under-valuation, scamming, prospecting etc. be managed.
I don't know enough about an ecosystem like this to know what attracts the right kind of devs, or what barriers need to be removed for onboarding them into the STEEM devops, and so on. But if that were a well oiled machine, all the details you two talked about would be ironed out, even if in fits and starts, or iteratively over time with experimentation.
I think the biggest barrier to mainstream adoption is the "crypto is a scam" narrative. A lot of people believe it is, or worry that it might be, so they steer clear. I'm not really sure how you can measure whether something is or isn't a scam, but if there was some technical way to assess scammishness and Steem doesn't score highly then that might be beneficial.
Putting aside the >$1 SBD issue, I think high curation rates contribute to a rich-get-richer dynamic. For long term health I think a currency needs a wide distribution, but the narrow self-interest of large stakeholders is to focus on passive incomes and guaranteed returns which I think works against a healthier distribution.
Maybe, but it would also reduce the incentive for overzealous self-voting, which contributes to the same sort of dynamic.
I wonder if it would be possible to make it a witness-defined parameter so that the witnesses could set different levels and try to find the best balance... The more complicated the change, the less likely it is to be implemented, though.
Update: I forgot to reply to your first point. I totally agree that fear of crypto being a scam is a big factor. How to measure scammishness is an interesting question in order to reassure possible users and investors is an interesting question.
I think self-voting is a bad idea. It should probably be made impossible at the chain level, and then de facto self-voting with other accounts under a person's control should be considered a form of abuse. In mainstream social media it's usually considered gauche to like/upvote your own content, and the monetary element of an upvote here tends to make it look even more slimy. Leaving whether or not self-voting is acceptable to ambiguous social dynamics was bad decision early on, in my opinion.
To me, the problem to avoid is overvaluing a post. If a post is overvalued, then it reflects a problem with the ranking system. I don't care if it got that way through self-votes, bidbots, or any other mechanism. Likewise, if the post is valued reasonably, I also don't care if it got its value from a bidbot or self-vote. I'm just interested in the end result.
The problem I see with self-voting (and bidbots) is when people write a 5 cent post, and they know it's a 5 cent post, but they vote it up to 50 dollars. That's the main scenario that led me to list improved post ranking as the first of my suggestions.
I think the original vision of steem had a significant "wisdom of crowds" element in post evaluation, but that model doesn't seem to work that well when post valuations are dominated by large-stakeholder votes. There's a tension between saying the value-proposition of holding Steem Power is that you get a say in what is or isn't valued, and the long-term value of the post-rankings mapping to something that seems to make sense to "normies". If there was a large population independently voting with somewhat comparable stakes we'd probably be in a different environment. Our current reality is a small population where most people's votes barely register and a few have huge influence.
I think it makes sense to think of over- or undervaluing a post as a negative externality like pollution -- it undermines the credibility of the platform. Good stuff not getting rewarded leads to people thinking there's no point in trying to post good stuff, bad stuff getting rewarded makes people think they're likely to get a raw deal.
Agreed. I think the original idea was for posts to "bubble up", but it's never really worked out that way because larger stakeholders lose out on curation rewards that way, so they prefer to vote at the front of the line. I think the piece that some large stakeholders might be missing is that their stake gets more valuable if they can encourage investment by helping the smaller curators to increase their rewards.
It seems like Steemit and one or two witnesses probably get that, but some of the "smaller" big stakeholders - not so much.
I think there might be a tragedy-of-the-commons thing happening where the optimal long-term collective action is to grow the value of the ecosystem, but it's easy for each individual to prioritize short-term value extraction while expecting somebody else to do the work to cultivate growth.
You may be right. I definitely think there's a tragedy of the commons dynamic in play with posting abuse for some authors who don't have any substantial skin in the game. With curators, there's a counterbalance, because people with more stake have a bigger motive to protect it, but some of the same incentives are still in play.
Interesting topic, hopefully we will be more aware of the decisions to be made and that among other aspects the quality of the publications are excellent.
Ah, but due to the fact that they are very good, they should not plagiarize them or disappear them from them as has happened to appropriate these. Greetings.
Yes, you're right. Thanks for raising the point. Plagiarism is a perpetual challenge that goes beyond the quality of the post, and I sort of ignored it in this essay. If we had more people, plagiarism would almost certainly be an even harder challenge. Although, that could be mitigated by having more curators and better curation tools.
I think a simple repetitive advert on a platform like Youtube can help bring the existence of the platform to the awareness of people. That is how most new tech firms in my country do it, and they get results.
Thanks. Yeah, good point. Advertising and promotion are also important. I was thinking more about how to keep people here once they visit the site, but you're right that we also need to get them here in the first place.
I don't remember where I heard that self-voting in the past had been reduced to 50% because the whales self-voted carrying most of the daily reward, from there the whales took another strategy of passing all their power to another account and vote your posts with what haejin is doing.
I am sure that the steemit team can make the changes that you suggest but they have another mentality that I honestly do not understand.
I am not saying that the steemit team is doing a bad job, but I do not see in them the interest of attracting new investors, if only Justin Sun could use social media to market the platform, I am sure that large investors would join the platform as @rme did, although I do not know how he met steemit.
Big investors like him is what this platform needs, I do disagree with the system that @upvu uses to use the borrowed steempower for them to make millionaires at the expense of others.
By delegating our steempower to @upve, we are enriching the Koreans who monopolize the entire trend zone, there I agree with you, when a foreigner enters for the first time and sees the trend zone, they will think that this page is Asian and the most it is likely that they are not interested in interacting on the platform.
You make many important points that I hope the steemit team will read.
@steemcurator01 please read this post.
Great point about a page looking like it's intended for an Asian audience. I think that experience is likely even more visceral for lots of visitors. For somebody browsing who expects English websites, pages in eastern languages in particular probably alert them to a potential scam (if they didn't already have their defenses up just for the crypto angle in this site's case).
I am one of the users who does not bother to see the trend zone, since it gives me the feeling that all those Asian accounts are from the same steemit team, because I do not think that all those accounts were created in the same year and at the same time, Asians took over the trend zone when the Tron Foundation took the right to steemit.
Hi @remlaps ,You have more cardboard than a recycler ha ha ha, it is an expression to say that you are a man who has studied a lot, it is a joke, I hope it does not bother you, in relation to attracting the masses to STEEMIT, I think there is no need to strive, the current pandemic and its confinements have led us to work from home, and to learn other methods to earn money, this panel is well known and the need for a salary is the basis for many to be here, thanks to GOD there is this panel in which several people generate their livelihood or help each other
What an excellent researchful work you have done here @remlaps this is great reading in between the lines
Thanks! I'm glad it was informative for you.
O yes it was Indeed sir
Inside talk, steem on!