Hardfork 18steemCreated with Sketch.

in #steemit3 years ago (edited)

The last month has been eventful here on Steemit by any measure.

There has been a lot of positivity

  • Many new users on the platform, which is great.
  • New marketing initiatives. I am very excited to see what @andrarchy and @zurvanic have up their sleeves.
  • The price of STEEM has been on a rollarcoaster but seems to have stabilised.
  • "The Experiment", which after an initial blast of negativity seemed to get accepted and was a positive contribution giving more power to the masses.

There were also some less positive developments

  • Dan leaving which has left us all sad but at least he seems to leave a post or two from time to time.
  • Continued high levels of people powering down. Only time will tell if it is a good thing that some early speculators are leaving.
  • The Standoff over Hardfork 17 which resulted in a late agreement/compromise which became known as Hardfork 18.

HF18 and its aftermath are what I am here today to discuss today.

What I have noticed since HF 18?

The payouts have gone a little crazy. They seem so low now, and it's still a mystery to me what happened to all the payouts that were due to be paid at the cutover. For new users it's worth noting that the reward pool (what's paid out each day) has been temporarily depleted and won't be full again for a few more weeks. After that time we should see better payouts each day. It will be back to its full level.

Payouts going back up to pre HF levels is quite reassuring but the following graphs are not so reassuring.
Can anyone provide some insight into what has happened here?, should we be concerned?

The raw data taken for these graphs is taken from Steemdata.com, by @furion. There may be issues following the HF but it appears to be up to date, despite the warning still on steemdata.com.

This data issue highlights another issue being experienced for developers. @heimendanger had a good rant about it here, SteemWhales Maintenance for HF18 - RANT INSIDE. Despite the frustration being felt by valuable members of the community we continue to get less than a satisfactory response from the Steemit core devs.

User Metrics

I am posting today to try to figure out what has happened since HF18. I have plotted some key metrics for each day since the 12th of March (the date of the @abit experiment), which I believe signify the state of health and strength of the Steemit Platform.

Unfortunately all these metrics are headed quickly in the wrong direction. We need to turn this Steem train around!


  • Votes are down
  • Comments and Posts are down
  • Active authors are down
    This drop seems to coincide with the HF. It is possible these metrics are resulting from data based on broken fields following the HF. I hope this is the case. Can anyone shed some light?

What is happening?

In order to investigate these trends a little further I have attempted to segment the data to see where the drop is coming from. I cannot show payout data as it is not flowing through dataset for me since the HF. (I am working on fixing that)

What I can show is how active the authors have been since the HF. I have grouped authors of different reputation levels together and plotted the number of authors who post per day in each group. The HF happened around the 30th of March. There are significant downward trends at all levels except for authors with very low reputations. i.e. the new users.

The data suggests authors have stopped posting. It is positive there are new users posting but worrying that so many high ranking authors are taking some time off.

Will they come back.....?

Thank you for reading this. I write on Steemit about Blockchain, Cryptocurrency, Travel and lots of random topics.


Posts are down because the rewards are down. People don't want to work for no money.

This brings forth a big problem with Steemit: it's not fun.

Only Steemit has to bribe people into using it; people use Facebook for free, Twitter, Reddit, etc. for free.

When Steemit doesn't pay, people stop using it.
This proves that people don't use Steemit as a social media, but as an ATM machine to make some quick cash.
This includes me. But. I would like for Steemit to be more than that.

At the moment the appeal that Steemit has that sets it apart from other platforms for me (besides the money) is the community and the interesting stuff people have been posting. If those people are leaving that is not good news.
What else does Steemit have to offer?
Steemit certainly isn't slick, user friendly and intuitive at the moment. If it loses its people it's losing its USP.
but ...
having said all that there is still great potential here. We just need a bit of momentum. People will not invest time in writing good, well researched content unless there is some reward. The drop in numbers we are seeing in the past week is the result of that. Maybe this means that Steemit will evolve into something else, maybe something better????

I struggle to find good content here, to be honest. It's so much easier to find good content on Facebook. Also, I don't see "well researched" content as valuable in and of itself; I also want fun content. Entertainment.

Not everything has to be about science.

Not everything has to be about science.

I agree and again this is part of the chicken and egg scenario. We need more volume, more critical mass here. @andrarchy it could be an interesting strand of the marketing campaign to focus on diversifying the content base. Give it broader appeal. Focus on bringing in a few writers such as

  • Sport writers bloggers
  • Movie critics bloggers

ALl you need is to spread wealth around with flatter reward curve.

Allowing people to create a feed of topics would also work wonders.

absolutely right. there are many social media with great contents (Medium, just not to nominate always Fb or Reddit). Steemit still exists because of the rewards.

This brings forth a big problem with Steemit: it's not fun.

How can it be made funner in your opinion?

Lots of small rewards. It should be like Reddit
Focused on only topics you care about paired with small rewards for adding something to the conversation.

You just said it! :D

The truth is, Steemit was introduced to most of us as "The Social Media that Pays". So if it stops paying, or only pays peanut amounts, most wouldn't want to waste their time here, because it has lost its original value to many.

For me it's the constantly moving goal posts that are the issue. If I am going to invest time in writing a post, I want to have some idea of what I will get for it.

As for the future and why we should stick with Steemit, yes you do still get paid, but without the ease of use and network that other sites have there is little incentive to try out Steemit for the first time. As you say if the carrot is "The Social Media that Pays" you would be very disappointed to go to the effort of setting up an account, figuring out what a posting key is and all the investment in time you need to put in just to get a payout on your first post of 1 cent.

The data suggests a worrying trend following HF18, maybe this is just the result of the bubble of optimism after the experiment, but we need to capitalise on any rush of activity and pivot off it. Will the stats recover in the coming weeks?

That's exactly the point. People don't want to invest much in what wouldn't pay either way. Other social media are used for fun. You post what you like without much discretion, but here on steemit, it's almost like a freelance job, and whole lot of discouraging downvotes to battle with. obviously, the rewards are the things keeping people here. so if it goes down, then it might be a threat to the success of steemit. There is apparently a need for more inventions on steemit.

thx for your data. but if you look at a period just a little longer you see that post and comments counts are about the same that they were before march 13.Cattura.JPG

What is interesting is that high Rep users (the longer term posters and users) have reduced their activity significantly in recent days.
Overall we are still down on pre experiment/HF levels in terms of posts and comments. New users/low reps must be picking up most of the drop of in terms of posts and activity.

This is an interesting trend and I just hope that after the reward pool is replenished we have not lost too many of the long term Steem users and the new users stick around.

Yes, daily there are some hundreds of new users, but probably, because of the low reward month, they create just a couple of posts and leave.
As for the old users, I think they are waiting to see how the new reward ruled will work with the pool filled again.
All this demonstrates that people are here mostly for the money (who can doubt about it, honestly?)

That graph appears to have cut off in early April sadly.

Thanks for that link. Some very interesting points.

Payouts going back up to pre HF levels is quite reassuring but the following graphs are not so reassuring.

We're still only at maybe 10-20% the pre-hard fork levels after two weeks. It seems like it will take quite a while to fully normalize. There's also loads of negativity over flags at the moment (both the experiment flags and random flags with no explanation).

The experiment flags will be redundant in the next hard fork. It seems like we need some system to be able to revoke the downvote ability of people abusing it. Perhaps you could spend/dedicate stake to counter someone else's ability to downvote but not counter their upvotes.

People have talked about only witnesses, or perhaps elected moderators having the ability to downvote. That could introduce accountability for bad downvotes, but it also means a special class of user who are allowed to downvote while ordinary users can't, even if they've never abused their vote.

I haven't been posting much since the HF and part of the reason is the low rewards. Compared to what it was before it's hardly worth while especially if its going to go up in the next few weeks when the reward pool is replenished.

I see the downvoting issue as heavily related to the inequitable distribution of STEEM at the moment and the lack of scale of the platform. I think this is an issue that would go away if STEEM was more evenly distributed and if there were significantly more volumes of users. In my opinion problems are manifesting themselves chicken and egg style until we achieve a critical mass.

If Steemit had much more scale, there would be a very strong incentive for people with a lot of Steem power to play nice, and it would be much harder to pick on someone as there would be so much more voting power at play. We need Scale! Driven by marketing! Lets get out of beta and see a roadmap for the release to the masses!!

It seems like we need some system to be able to revoke the downvote ability of people abusing it.

How can one abuse the downvote? It's just a vote. Can upvotes be abused?

Yes, both can be abused.

The purpose of a consensus network like Steem is to encourage the interests of individuals to be aligned with the interests of the network as a whole. If everyone behaves towards the benefit of the network, everyone benefits (tautologically). It is possible to behave in a way which undermines the success of the network, towards your own benefit in either the short or long term. If you can extract resources at the expense of the network without contibuting value to the network, that is considered selfish behaviour. Selfish behaviour is to be expected, but it is a failure of the consensus mechanism when selfish behaviour is not essentially the same as pro-social behaviour. But even when interests are successfully aligned, not everyone behaves in their own interest for a variety of reasons, and detrimental behaviour can persist.

Here's some obvious examples of abusive votes. If people were upvoting child porn and spam marketing to the top, that would be abusive. It would clearly be to the detriment of the network.

If a famous and popular celebrity joined Steemit, did an introduction post, only to be swamped with downvotes, those would be abusive votes. They are clearly to the detriment of the network. Would you disagree?

I agree that there are game theoretical reasons why if every individual makes the most optimal choice for themselves could lead to a sub optimal outcome for the entire group. An example would be if everybody only ever voted for their own posts and used up all remaining voting power to downvote others.

That being said, I don't think abuse is rampant from what I've observed. It is also inherently subjective here, what A deems to be abuse B may deem to be a legitimate use of power to downvote offensive material. Also, those who deem themselves to be unjustly downvoted are going to be disproportionately vocal.

On balance, the system of flagging is far from perfect, but it's probably not a high priority problem right now.

It is also inherently subjective here, what A deems to be abuse B may deem to be a legitimate use of power to downvote offensive material.

Every vote is subjective here. The solution is the same as it is for other votes, if you are countering someone's vote, it must cost you at least as much voting power/stake. If someone counter's someone else's vote, and it's the counter which is abusive, it should at least be at their own expense. And even when a counter-vote is abusive, the original vote can always be reinforced by someone else, until everyone involved has spent their voting power (and the sum of the remaining voting power essentially decides how the rewards are allocated).

I do disagree. If Steemit becomes communism where everyone has to work for the benefit of The Party, it will no longer be what I assumed it was supposed to be.

I guess I just value the individual more.

In your view, there is no such thing as an abusive vote?

No. All voting is arbitrary. If you choose to downvote everything I do from now on, that's your choice. You've invested either time or money to gain the STEEM Power you have, and you are entitled to use it any way you see fit.

If people want to counter you, they just need to do the same.

You essentially argued my current position when there was a user downvoting for the purposes of blackmail.

I'd point out, though, that all users here are stakeholders, and the health of Steemit is everybody's best interest. So, if something is considered harmful to the platform, there's a good reason to stop it.


I am an individualist as well. The rights of the individual are paramount, no individual should be coerced (with force) to act against their own interest. That does not mean that everything an individual does is good. Co-operation exists to maximise the benefit of multiple individuals. When actions benefit some to the detriment of others, it ceases to be cooperation, and ceases to be a net benefit to each participant (which discourages participation). Since Steem exists to benefit all Steem holders, the system needs to be designed to align the selfish interests of the individual with the interests of other participants. When actions go against everyone's interest, they need to be countered by the others who are financially incentivized to do so.

If people want to counter you, they just need to do the same.

We don't actually disagree here. All I am proposing is that the act of countering be made more straightforward. It shouldn't need to waste the degree of attention or blockchain resources as it does now.

In summary, when I say that something is 'abuse', I am saying that it has bad consequences, and it should be made easier to counter the effects. The ability to revoke someone's downvote with stake, is functionally the same as countering their votes on the blockchain, except that it consumes less resources.

I think we agree and disagree to a degree.

As a stakeholder, of course I want the platform to be successful so that my stake will be worth something someday.

The blackmail thing didn't make Steemit look very good.

But the downvote itself is a necessary tool in my opinion. I'd just reiterate something @beanz in fact has said: it's just a distribution tool, just like the upvote.

Without the downvote, people like the STEEM Guild members could just dominate the platform and the community could not combat that in any way.

And of course downvotes happen for personal reasons, but I guess the best way to counter that is to simply not piss off the wrong people.

I predicted this a while back; I said that a lack of marketing would ultimately kill Steemit. As great as @andrarchy is, the new marketing initiative misses the point completely. Talk of marketing the currency Steem without the website will not do the trick.

Also Steemit need to spend some money; social crowd marketing, SEO is bullshit. PPC is what works, Facebook advertising works; why do you think they made 11 billion from advertising last year? Do you think they would make that if it didn't work?


The currency is only valuable as long as the website exists. The website is only valuable as long as it had engaged users. The more the better, sure you need the infrastructure, apps etc for this to work but i agree with the there should also be focus on the user base. How do we grow this. How do we bring in new writers.

Maybe forget about established people but how about an idea of an academy, like sports clubs use to develop new and emerging talent?

This is a great analysis, thank you! I love to see this kind of thing. 😁

As for interpreting the cause, not to simple, but in general I would say the hard fork more than the experiment. If anything the experiment seemed to boost action.

I've stopped posting for two reasons:

  • I'm much busier than I thought I'd be now, by an order of magnitude!
  • The negativity of some users has kind of brought me down, as well as lack of engagement from those I had come to admire.

I will be getting back on the horse soon though 🤠

Thanks for the feedback :)

I have been busy this month too, but I do have a few things I want to write about in the coming weeks, when I get some time. And lots more analysis I want to do :)

Looking forward to reading your posts again soon too.

I think the instant feedback is gone and so posting has declined. It used to be fun to see your vote add money to a post (especially after the expirement). It was also fun to see how much traction your post was getting. I wish the reward data was there. I suspect

  1. rewards accrue over time. So some of your impact doesn't appear till well after you voted
  2. the 7 day window to vote magnifies the unfair rewards curve. It basically lets every account vote 7x as more in a payout window and magnifies the influence of high sp holders. 7 day reward should be paired with a reduction in daily voting power by 7 to get. Back to where we were.
    If you expanded your window to before the expirement ( which mainly relieved reward curve issues) I am curious if activity levels are now above or below pre expirement levels .

I was getting used to the instant feedback I liked that.

In terms of activity, It has declined since before the experiment. The graph that @paolobeneforti has posted in his comment gives an indication.
There was a spike of activity after the experiment began, which I was hoping was a result of the boosted incentives to participate on the network (more voting power and more evenly distributed rewards), but since then we are back down to below pre experiment levels. As I mentioned in another comment what is worrying about this for me is that although there seems to be new users posting (which is great) there are a lot of veterans stopping or pausing posting. Could be a coincidence, could be unrelated but the longer term trend will be important to watch.

For more stats since before the experiement and HF began check out an earlier post of mine Dolphins and Minnows get more influence...Continued

That was a great post.

Thanks reading. 👍

That is a great post. I have left some comments.

My sister thinks that the rewards are being distributed only to a limited few again. Wondering if you would do another one of these posts again to see it empirically?

The hardfork seemed to cause issues with the layout of the data feed I was using. I need to reengineer my programs to get it to produce these graphs.

The good news is I have found additional detail available from steemsql so I will be also be using that going forward. I hope to get some time this weekend to work on it.

The thing I have noticed recently is how the payouts seem to build up over the week now which is very different than the volatility that was there before exasperated with the downvoting.

Coin Marketplace

STEEM 0.17
TRX 0.03
JST 0.040
BTC 10732.56
ETH 353.85
USDT 1.00
SBD 0.96