5 Stupid Situations Generated By Steemit Growth - And How To Avoid Them

in #steemit7 years ago

June 2017 will probably go down in history as one of the best months for Steemit growth. I think the entire population of active Steemit users doubled. The rough numbers will probably tell you an even more appealing story, but I don't follow the rough numbers, because, in this specific use case, they can be misleading. In the early days of Steemit, one person could use the cli_wallet to create literally thousands of accounts. If I'm not mistaken, I think that one of those persons is still alive and playing around the blockchain these days and those thousands of accounts are still lingering around.

So what counts is the number of active accounts. And we went from roughly 9k monthly users to over 21k (on a bad day).

This is wonderful news. It really is.

But as in every growth situations, if you grow too fast, you will experience what is known as "growth pains". There are at least 5 types of growth pains Steemit is experiencing at this time and I felt compelled to write a post about them.

I do believe that if we don't address them, they could drag the whole ecosystem into a stale.

1 Follow for Follow

This was seen before in other social media platforms, more prominently in Twitter. I still remember the early days of Twitter, when you would get a "Please follow me back, I just followed you" direct message, almost every hour. Ok, so one may ask: "What's wrong with that? Why wouldn't I return the favor if someone was so nice to follow me?"

Well, because, if you don't assess your potential of interaction, you will not bode well in the future. As time consuming as it is, and as boring as it can get, a minimal due diligence is required. Following blindly each and every person clicks the "Follow" link in your profile will not create you an audience. It will inflate the number of followers in your profile, and that's it.

If you don't know the difference between an audience and just a number, well, maybe you shouldn't be on Steemit at all.

2 Meaningless, Polluting Comments

This is by far the most annoying growth pain of Steemit thus far. Sometimes it doesn't even take one minute and my posts are displaying a nice(?) string of comments, all pretty much with the same message: "Please, pretty please, upvote this comment and give me some money", sometimes not even disguised in something completely meaningless, like "Oh, nice post, you are a nice person" etc.

For all of you who are hoping to get rich by that, stop. Go get a job at McDonalds, if you qualify, or do whatever you can do for a minimal wage in your country, because you won't make even that with this approach. Believe me, it doesn't work like that. So, if you don't want to experience frustration, poverty and despair, stop posting automated comments with no contribution.

You will be soon filtered out and you won't even realize that.

3 Extensive Self-Upvoting

With this, we get to a more delicate level of growth pains. If the first two are generated by a specific demographic - namely people attracted by a quick buck - this one is a bit more nuanced.

After HardFork 19, many people realized their vote power increased significantly. And they started to just post stuff (short posts or meaningless comments) and then upsteem the shit out of them. What happened was that the reward pool was emptied individually, without an interaction.

At first sight, this may seem like a legit way to get rich: gather Steem Power, upsteem yourself, power up and cash in profits. Rinse and repeat. Wrong. Deadly wrong. If you don't engage and if you don't allocate parts of the rewards pool to other people, the system will collapse. And by that I mean that your rewards will literally be less than a thousands of a cent. The blockchain will continue to work, but the value associated with Steem will go down the toilet.

Nobody will buy Steem if the entire ecosystem is made of thousands of isolated chambers in which every individual is voting for himself. Nobody.

But plenty of people will buy if the ecosystem will boast interaction, foster ideas, create and implement collaborative projects and so on.

4 The Deadly "Minnows Squeeze"

As more and more minnows will join the platform, there will be more and more Steem allocated to those accounts. At this moment, the account creation fee is probably 1 Steem. It doesn't look that much, now.

In an ideal world, this fee should be paid by the ones who want to join. But in an initial effort to attract users, Steemit Inc decided to pay this fee, for the user. That's how you explain the "join and get X free Steem" marketing stuff you see floating around about Steemit.

But this supply of Steem is not endless. At some point it will end. Try to think what will happen if there will be 100,000 millions users joining Steem tomorrow. There will be 100,000 Steem allocated to those people.

And here comes the funny part. Out of those 100,000 people, I think less than 1% will become power users of the platform. The rest will try it for a few weeks and then leave. But, the vast majority of them will leave their Steem in their account. So we will end up with dozens of millions of lost Steem in unused accounts, lingering around the ecosystem. This will act as a pressure on the price, not because of the amount of it, but because of the uncertainty. Nobody can predict when these dozens of people will remember they had 2-3 Steem in their wallet and start dumping all, thus dragging the price down.

I know it sounds Sci-Fi, but we've already seen a lot of Sci-Fi already unfolding in the crypto-universe.

I really think Steemit should do something about the account fee creation. At least secure its value somehow. As strange as it may sound now, making people pay to join might be a good idea. I'm not sure how many will start the "follow for follow" thing or how many will post meaningless comments, if they are already invested.

5 Chronic Lack of Education About the Platform

By far the most dangerous growth pain of all. I think less than 1% of the newcomers know how the platform works. They come with a Facebook/Twitter/My Space/Hi5 bias (or, even worse, a Tsu bias) and they expect thing to "just work". They have no idea about the reward pool, about witnesses and their role, about the role of SBD in the platform and so on.

In more than 45 years of wandering on this Earth I learned that the most frightening, dangerous and deadly peril for the human race is stupidity. Or ignorance. Or confusion. Name it as you wish, but the deliberate act of ignoring the mechanisms of the world never ended well. On the contrary.

I cannot stress enough the importance of education, with regard to Steem. If one in 2 people will actually understand how things are working around here, then we will never see "follow for follow", polluting comments, extensive upvoting.

And even the "minnows squeeze" could be transformed into a "minnows explosion" that could bring the price of Steem to the moon.


I'm a serial entrepreneur, blogger and ultrarunner. You can find me mainly on my blog at Dragos Roua where I write about productivity, business, relationships and running. Here on Steemit you may stay updated by following me @dragosroua.


Dragos Roua


You can also vote for me as witness here:
https://steemit.com/~witnesses


If you're new to Steemit, you may find these articles relevant (that's also part of my witness activity to support new members of the platform):

Sort:  

And your answer is the best example of education around here. I will start do this more with the ones commenting in this manner.
I was thinking lately that there are too many similar posts, explaining basic functionalities in steemit. You made me wonder with the last point and I think they bring value for the new joiners even if it's too much for me.
How else do you see education happening?

Tutoring new "lieutenants" - people following you and commenting constantly (no-nonsense, obviously) and growing them.

Hello. I read your article and am wondering if the number of new users is causing our wallets to shrink, or is it because the value of steem dropped?
If you know the answer, would you kindly enlightnen me?

Both answers are "yes", but in different ways.

The most important "yes" is the drop in price. STEEM lost more than 50% in less than 3 weeks, it was trading at $2.4, now at $0.98.

The second most important "yes" is the one related to new accounts and the fact that these minnows are also participating in the reward pool. I don't have any calculations on that, so I can't give you a specific percentage, but I see it as insignificant, compared with the price drop. In time, the "minnow squeeze" will become a problem, though, if there aren't any adjustment made at the blockchain level.

P.S. If you're looking for enlightenment, I would dare to recommend meditation and compassion. Chances to get enlightened by a random answer on a blockchain-powered social media are ridiculously low.

Thank you for taking the time to enlighten me. I appreciate it.

The top five problems faced by Steemit are so well explained by you @dragosroua . It's my second month on Steemit and I already have seen what people do well or not so well.

Follow for follow, valueless comments, self voting, and lack of education are problems that originate from people. People need to be guided and they need to put in some personal effort as well. I can proudly say that I am involved in none of them and in some weeks I have educated people enough to give them a head start. My blog is full of education and guidance for minnows and newbies. The effort goes unrewarded at times but it's okay. I'll carry on.

As far the minnow squeeze through Steem allocation as joining fee is concerned, Steemit does need to listen to you. I agree. I think the delegation of SP is better than giving Steem and that delegation should be for a period of 4-6 weeks only. Steem given is a Steem lost.

Talking practically about educating people about Steem and Steemit, please have a look at what I have created for the community and how much money it could bring in to Steem if people get to know how to buy Steem from crypto exchanges. Please help me spread the message by resteeming it.

Thanks a lot for this amazing post. I agree and I am resteeming it in hope of spreading awareness about Steemit.

Once again, you put things in perspective. I was feeling down about why so few upvotes of my posts and comments.

Now that I see the inherent problem with upvoting my own post when I post a new article, I will make sure that box is cleared and see what happens.

I see nothing wrong with voting your own posts, as long as you keep voting other posts as well. That's the real problem, keeping all the voting power to oneself. It's not like it's yours, anyway...

 7 years ago  Reveal Comment

Confronting these follow for follow people seems to work most times. Some people are a little slow to realize it's not the way to act here. This guy yesterday posted a comment on my post which I had already seen him post the same thing twice already in other people's posts. I checked out his blog and he commented the same "follow for follow" message like 40 times in 30 minutes. It went back way further than that too. His newest blog post was his celebration that he got to 100 followers and about growing the community, even though people had been flagging him down to a Rep 12. I told him what he was doing and that people don't like that and he should actually engage with his comments. He messaged me back and said he didn't realize and that he would stop. I went back and looked a little later and there were no new "follow for follow" comments. So direct confrontation might work for a bunch of it.

Yes, it usually works (as long as they are not using bots) but there's a "time to reward" balance to every activity you do around here. I tutor them as much as I can too, but at the end of the day there's a threshold and mine was just reached :)

All super good, educational points. I had the follow for follow/upvote of upvote dementia during about my first week on Steemit; and I see so much of it now, just seems totally ridiculous.

Glad you're cured now :)

It is the best part about steemit, that it has a learning curve

The best part is that it actually works, IMHO, but I like how you think.

My experience is that the Steemit learning curve is quite steep and requires a great amount of time and diligence. Anything that could make onboarding more smooth would be a big improvement.

@dragosroua
I really appreciate your post. The trip to my present reputation score has been hard and slow. And I have made some mistakes along the way.
I have read your post. I always want to read the posts I upvote. I will save this post. You have made some important points. Your thoughts, along with other thinkers, can help save Steemit and have our Steemit family to grow into something good and lasting.
Thank you

Francis

What if there was a set time period after which an account is flagged as dormant and the steem pushed back into 'active' use?

That's actually a very good proposal.

I really like the article and I think this shows most important issues on steemit platform. I think the second feature of oSteem will be to clean/hide all comments that doesn't create value to the platform.

Very difficult to assess value, hence, almost impossible to implement this. It's a painful, growing process that Steemit has to go through and it will end up with the filtering of meaningless, polluting comments. I guess reddit went through that too.

For articles it's hard to evaluate value. But for comments isn't so hard. With the plugin I'll filter all the comments that have less than 10 words and contain one of the expressions: "great post", "valuable infos", "Awesome", "good post", "really interesting" etc I know this will not eradicate all the useless comments, but for sure it will remove at least 60-70% of them.

Awesome good post. Itz really interesting . thnx alot for sharing with us. this is nice. Good work. Upvoted. Followed.

Did I miss anything? :)

2 Meaningless, Polluting Comments

I don't know why, but this point kinda reminded me of Freelancer.com and the shitty bids on projects and messages from South and South-East Asians that help to drive the price and value down on services and bids offered.

Yes, Freelancer.com, Elance.com. Those were the times...

And yes, I think you missed something: I also wrote something about this 3 weeks ago, can you have a look? Very much appreciated, here's the link. (And the link obviously goes to something completely unrelated, most of the times a copy-pasta meme.)

@dragosroua, I agree with your points. Education is the key, and fostering a successful Steemit culture is dependent on both educating the minnows like me, and constantly improving the underlying system. My idea for the self voting "problem" is the apply a maximum weight to self votes, which also reduce voting power by a higher percentage. For example your first self vote in a 24 hour period uses 5% weight, but uses 10% of your power. The next uses 4% weight and takes 15% of your power, and so on until you hve no power left and your upvote is worthless. I don't know how hard this would be to implement or how to gather support for the idea, but it would stop the provlem pretty quickly.

Very interesting proposal too!

How do you account for delegated SP and bots with this?

I don't think the huge drain on the rewards pool is coming 0.01 at a time from the people who engage in F4F and the 1% of people who control 93% of the SP probably wouldn't be very affected either.

The delegated SP doesn't really matter. With this system your 100% upvote, even being worth $100 would only give you a $5 self vote, but your voting power would go down substantially, taking longer for you to get back up to that 100%, $5 vote. After 3 self votes in a day, you would have only 55% voting power, which would take more than 2 days to regenerate to 100%.

Bots are a very different animal, and it would be difficult to implement this sort of restriction without a lot of work. Perhaps a vote for the same author multiple times in a short period could be seen as equivalent to self upvoting, and then the same or similar weighting rules could apply. This would affect a lot of users who vote for their friends often unless there is a way to distinguish between a bot account and a normal account.

Hi @dragosroua, I have expanded on this idea with a full proposal to update the voting model. I would love to know your thoughts and any shortfalls you may see in its implementations. It's a bit of long read, but hopefully doesn't have too many holes in it. It also covers the problem of delegated SP to gain upvotes via bots. It is here if you are interested:

https://steemit.com/steem/@bmj/proposed-hf-changes-to-bring-voting-back-in-line-with-its-intended-purpose-rewarding-quality

Thanks!!

dragosroua ,
I totally agree with a micropayment for an account. Because this will support Steem price. And having probably more active users that way. But I read your : where comes the money from to. And with every good will I have , I don't see any buying pressure on Steem, (except those who turn it in Steem Power) , for wrighting more valuable blogs by upvoting. But every reward is printed, every initial vote to. If anyone sees terrific ideas etc. in steemit's blogs. They maybe purchase Steem on an exchange and make a free account. Marketcap of Steem is on nr.17 (at moment of wrighting). With +/- 50.000 Sat. So the only good thing for Steemit is to attract "wealthy" steemians who wanna power up. And this has to be bigger than inflation. Without wanna do on assholism, I see a crypto copy of normal fiat monetary systeem. And sorry for using this awful word : is a Ponzi-scheme what is always able to pay out. But at what price ? cfr. dollar chart from 1900 till now. Can u convince me I'm wrong , I would appreciate it. (maybe during a chat session ?) grtz

I won't consider the stupidities that you listed as a bad thing. All it means is that there are rooms for improvements to make the platform better. Learning how to cooperate with each other is key in every community and organization.

If you look at my post closer, you'll see these "stupidities" manifests because lack of cooperation. Catch 22.

Only just started up yesterday, and Point 2 you made I have seen a few of these against my 2 measly posts Ive created. Was wondering actually how do you tell if the person is a bot or not? I seemed to get a few comments straight away and sort of seem a bit sheepish now responding to them all if they were bots (because thats what I did and followed, but non followed me back). Maybe someone could start a campaign "Watch for the BOT!" and and then admin mark a bots comments as the colour red automatically.

great post and valuable infos
thank you for the tips

Hi @amyy thanks for the comment as well. I saw you posted this less than 2 minutes after I published, so it means you're a very, very fast reader. I also saw you joined just 3 day ago.

So, do you have any insights about how this platform works?

This comment is as valuable as the article! :))

Great post, thanks for sharing. I think maybe there might be more efforts done to educate new users like having them go through a tutorial after signing up on the platform similar to game or educational tutorials when first starting. I know right now there are welcome bots that post some guides on a person's introfuction post. However, I feel we should be more direct in giving guidance quickly to new users.

The onboarding process of Steemit improved greatly once Steemit Inc started accepting suggestions from other witnesses outside their initial BitShares/Steemit seed, and @timcliff is a great example. But I think there's a lot more room to improvement here, it's obvious that just content is not enough. Maybe combining content with app logic (creating some sort of a fast "ladder" you have to climb by performing a few tasks over one or two weeks) will be a more effective approach.

Solid post. Thanks. What do you propose be done about point 5 that is actually feasible? Bearing in mind the realities of selfishly motivated people, interested in getting rich/making an easy living posting on Steemit etc.

I mean, there is not a lot of value in describing the situation we wish to see if it's neither realistic nor practical (and I'm not suggesting this is what you've done) - I'm asking what pragmatic approach can be done to solve the problem you've described.

I've said from the beginning of my time on Steemit (less than two weeks) that the learning curve is seriously difficult for a newcomer to grasp. If you want the average person to 'get' it, then the economy and the workings of Steem, Steem Power, Voting Power, SBD etc need to be much easier to comprehend.

If we can reach some kind of 'pick up and play' state, where you can do what you observe others doing in a straightforward and intuitive way, then this system will explode and be the amazing thing so many of us see it having the potential to be. But I'm sometimes worried it's design is simply too technical for that, as it stands right now anyway, to truly compete with the likes of Reddit and Facebook.

Very keen to hear your thoughts here.

I think everything is very clearly explained in the available FAQ's, in the Steem wiki and in a lot of Steemit posts how this platform works. The information is not the problem. How information is presented is not the problem.

The laziness of many people is. Laziness to read and learn.

It's the same kind of laziness that makes people accept those long-ass Terms & Conditions contracts without reading them whenever they join the Facebooks & Googles of the internet and then are surprised after a few years that those platforms sell their data or do whatever nefarious things they do.

You can't educate people by force.

I find it a bit similar to what's happening now at a macro level with the crypto market as a whole. I've read many comments and posts in the past couple of days from people that are panicking because they clearly haven't taken the time to understand what blockchain technology is or why it is important. The kind of people that maybe have heard that bitcoin and cryptos are "hot" right now and they can make a lot of money in a short period of time without doing much. The kind of people that can't really differentiate between investments and speculations.

I find it a bit similar to what's happening now at a macro level with the crypto market as a whole. I've read many comments and posts in the past couple of days from people that are panicking because they clearly haven't taken the time to understand what blockchain technology is or why it is important. The kind of people that maybe have heard that bitcoin and cryptos are "hot" right now and they can make a lot of money in a short period of time without doing much. The kind of people that can't really differentiate between investments and speculations.

Word!

See, I am not arguing that all the available info is out there and readily accessible in FAQs etc. I know it is, for sure. And I don't deny that the problem is people's laziness.

My whole point is: Steem is not pick up and play accessible, and if we don't look at it pragmatically and from a lens of how the world is (people are lazy and want things easily) and instead focus on how the world should be (people find the resources, educate themselves, come to Steemit informed and ready) then we're not approaching this correctly.

That distinction between how we want it to be vs a realistic look at how people are, and how Steemit's onboarding may need to be amended as a result of this, is really what I am commenting on.

I don't think there is a way around this. I think the biggest learning curve is understanding how blockchain works. Once that is understood the rest is not that hard to pick up just by actively participating on the platform.

You are objective. Our desire is to ensure the sustenability of this platform.

Well to be very true my growth here is really slow I made few mistakes on my way but I followed some great people like you and now I am still learning and one thing for sure learning steem platform is a never ending journey umm I guess so!

One of your best, educational posts till now. It might come out of frustation but it is very constructive. Ignorance is the number one killer in here. Being a minnow I do confront the other minnows, with most of the times constructive comments. I tend to have some sarcastic comments also, but this only another way of education.

It might be a proper idea to pay for your account if you want a fast one, to start almost wihtout any power so the account creation shall be cheaper and somebody you know shall vouch or pay for you. Minning as some other suggest in here shall not be taken in account, as we want to attract the average content producer and not only the linux administrator with a super server.

I've also thought that more filters in the feed/browsing interfaces could be good. Such as only display content from users within certain reputation score ranges, only display content that's been voted (or as you're trying to get the term to be, "upsteemed"!) on x amount of times, hide resteems, only display content with x amount of comments or more, hide posts of less than 100 words, and so on.

Just ways to sift out some of the slew of low-effort posts via the front-end. I'm really hoping something comes along in a future update for both the front-end and the reward system that penalizes particularly poor content - such as downvote/flagging being free but requiring approval (unrealistic given the scale that Steemit will grow to and the amount of human effort required to do something like this). But just some way to raise the bar a little bit. I wrote an extensive reply on another post that had some of the same points. I'll link below as I'd love to hear your thoughts too:

https://steemit.com/steemit/@holoz0r/steemit-holoz0r-s-view-of-the-current-state-of-affairs#@badastroza/re-holoz0r-steemit-holoz0r-s-view-of-the-current-state-of-affairs-20170717t001038239z

This post has been ranked within the top 25 most undervalued posts in the second half of Jul 16. We estimate that this post is undervalued by $13.42 as compared to a scenario in which every voter had an equal say.

See the full rankings and details in The Daily Tribune: Jul 16 - Part II. You can also read about some of our methodology, data analysis and technical details in our initial post.

If you are the author and would prefer not to receive these comments, simply reply "Stop" to this comment.

Well said, well done, and yup sharing it out to folks!