Steemit is suffering from negative network effect

in #steemit7 years ago

You might be familiar with the term network effect. Usually it refers to positive outcome when a service becomes more and more beneficial to everybody when new users join in. But there can be also negative outcomes, which happen when the platform becomes worse with every new user.

Currently Steemit is suffering from negative network effect. This is because the user interface can't handle a big userbase. Every day we are seeing more new users coming in. And what they do? They post and comment. It adds up too quickly. That makes it harder for everybody to find content that they like the most. It's also harder for bloggers to get more followers.

If you have recently checked out the page for new posts, you know what I'm talking about. You need to go through a lot of posts to find even one truly interesting.

Just a while ago I made a test. I skimmed through about 50 newest posts. I couldn't find any worth of upvoting. Only one post was somewhat entertaining because it contained a funny video. But I had already seen the earlier and the post didn't contain much besides it, so I didn't bother to upvote. Needless to say, didn't find any authors to follow, either.

You might have also seen a lot of "nice article, please follow me" comments to posts. That's what happens when new users are desperately trying to get attention.

To make the situation worse, Steemit is planning to get ready to get millions of new users more to the platform. If that would happen without redesigning the whole UI/UX, Steemit.com would become almost unusable. That's why I disagree with the proposed hardfork 20. It just doesn't make any sense to focus on getting more users to the platform while it can't scale.

What we need, preferably as soon as possible, but at least before any mass signups of new users happen:

  • Customized feed. Now we can control our own feeds only by following other users. That's not much. Facebook has a good reason to control what their users see in their feeds. If everybody would see everything they follow, there would be too much noise and the user experience wouldn't be good. But the problem is that it's Facebook who is deciding what users see – not users themselves. With Steem we have a chance to let users be in control and customize their own feeds as they like.
  • Language support and filtering. If you see language that you don't understand, it's ultimate noise. There is no value. It just fills the page and steals your attention. That's why Steemit should implement a support for defining a language for posts. When a user publishes a new post, they can let everyone know what language it's written in so others can decide do they want to see it.
  • Communities. Hopefully coming in the near future. This feature will allow creation of real communities around certain topics and interests. Currently there is no option to follow a tag so this hasn't happened nearly as much as it could have. Comparable to groups in Facebook or subreddits in Reddit.
  • Notifications. The current notification system is unreliable and too simple. Users should have several other options for receiving notifications, like emails or notifications for mobile phone.

Once the user experience is better and we have shifted from negative to positive network effect, we can start to think about scaling the platform for millions of new users.

Sort:  

I've seen the problem grow bigger in just few months. I posted a few times around 5 months ago, then had a pause and started a week ago again.

It's crazy how different amount of new posts are coming all the time. It's very hard to find anything interesting and I feel there is a big difference in post quality. However, who would want go write quality posts when they can be missed by everybody.

Sad.

I've seen many poorly written posts such as posts where all the letters were bold, lousy punctuations, terrible spelling and sometimes even unnecessarily foul language.

That's sad and true.

My first reaction was "I hope you're not talking about me." :D

Haha. No, I was supporting you. Your post is top-notch!

Ha, thanks :D

Just had to be sure.

There is a lot of shitposting unfortunately.

I can't say it's a new thing, but it's way more common now as people are aiming for easy money. Or maybe we are grown old and boring.

I have been saying since I started here (ahem 8 or 9 days ago) that I'd really like to sort through the resteems to find the people I follow's original content! I hate sorting through the resteems!

I see that Steemit is still in beta, so solving a lot of these problems first before mass signup will help with retention. If a newbies signs up and find that it is not as convenienent then we have probably lost them forever. Not to mention possible post about bad experience on Steemit.com thereby discouraging others to join.

Good point. Increasing quality later is harder than increasing number of people.

Totally agree, lets just focus on quality at the moment and then we will be ready to handle quantity. good points :)

Steemit already has subcategories which act as a filter; for example, "homesteading"..

Adding more specific subjects might help, but in the end there's going to be a flood of screed on Steemit ... It just goes with the territory ....

So I disagree with you because growth trumps your complaint ... And your complaint can be fixed without sacrificing massive growth...

Finally, in this globalised world if your brand isn't growing it's dying. Think about that. I understand quality content is important, but most people aren't writers and they never will be so get used to it....

Steemit already has subcategories which act as a filter; for example, "homesteading"..

Filtering the content should be automatic. With current tags there aren't anything automated, users have to make an effort to find and see everything that is posted with a certain tag. Users can't even subscribe to a tag.

So I disagree with you because growth trumps your complaint ... And your complaint can be fixed without sacrificing massive growth...

On the contrary, I think it will be absolutely necessary to fix the UI/UX if we want to have millions of users. With the current UI the experience will be so bad that the system will become unusable and old users will leave and new users won't join.

"Filtering should be automatic "

Yes, that's what I mean. Fix that. Problem solved... Growth continues...

Yep , it is harder now than before ...

Doesn't Twitter (and many other social media platforms) have the same "problem" you're describing here? If a new twitter account tweets out to 0 followers, they shouldn't be confused as to why no one saw or interacted with their tweet. I'm not convinced there's really a problem here. This is just how social media works. Gaining followers takes time and effort. Writing an amazing blog post on some random blog doesn't mean people will find it. Building a following matters and that isn't effectively done with "follow me too!" comments. It takes time, engagement, and real effort.

I've been blogging for 13+ years while being mostly ignored. That experience has helped me write things here people enjoy which is leading to a growing following. I don't think there's a shortcut for that process.

I do think some small UI tweaks could be made which might help a little, but ultimately this platform deals with the same chicken and egg problem of every other social platform.

Q) "How do you get a lot of followers?"

A) "Consistently write a lot of great content over a long period of time."

Q) "Who decides if the content is 'great' and how do you ensure people actually see it?"

A) "'Great' is determined by a lot of readers who find your content, usually by following you."

Q) "So... how do you get them to follow you?"

A) "..."

Doesn't Twitter (and many other social media platforms) have the same "problem" you're describing here?

Nope. They don't show everything to everybody.

Getting followers isn't the only problem. The deeper problem is "how to find like-minded people and engage with them in a meaningful way". We don't need just minor UI tweaks but complete redesign.

I want to prioritize posts that are shown in my feed(s) based on rules that I set up myself. For example, "all posts with a certain tag", "all posts by a certain author", "posts that are commented by users I follow", etc.

I want to have notifications to my email so that I don't miss anything really important. I guess daily/weekly digest emails would be quite popular, too.

I want to get notified for discussions in my posts and in posts that I have commented. Currently I am missing a lot of interesting discussions because I just don't stumble upon them. I can't take part if they are practically invisible.

Currently it just takes too much time and effort to find and read all the good posts. If we get even more users on the platform, the situation will become horrible and I don't think I will spend as much time here anymore.

When signal-to-noise ratio is getting worse, the solution is to find ways to strengthen the signal automatically, not just force users to suffer from all that noise.

"Great content" is subjective. What is great for one is not necessarily great for another. The system needs to allow people with different tastes to find themselves, form communities and interact in a meaningful way. Now all content is just dumped for everybody to see.

When signal-to-noise ratio is getting worse, the solution is to find ways to strengthen the signal automatically

Very well said.

Now all content is just dumped for everybody to see.

But many social media platforms started that way. It took a while for Facebook to implement algorithmic filtering and for Twitter to follow with the same. I agree with the final goal, but I'm patient with the time needed to get there.

Thanks again for a detailed, valuable comment. I'm glad you found this reply (even with all of Steemit's shortcomings) worth replying to. :)

But many social media platforms started that way. It took a while for Facebook to implement algorithmic filtering and for Twitter to follow with the same. I agree with the final goal, but I'm patient with the time needed to get there.

I'm just trying to raise awareness of this problem to make sure it will get prioritized properly.

Based on the latest hardfork proposal, Steemit Inc wants to prioritize getting as much new users as they can. I don't think that's very good plan. It won't go well for anybody, not for the old or the new users.

The first priority should be to develop the UI/UX to the level that can handle millions of new users so that everybody will be happy. After that has been taken care of, we can start to talk about user acquisition.

I appreciate your concern, but that doesn't appear to be the path other successful examples took. Growth is key, more so then every person having a perfect experience. If growth took off (even without improved UI), Steemit signups would be crippled. Seems to me that's a higher priority over UI. But guess that's the backend programmer side of me talking. :)

I'm looking at this from the business perspective. It doesn't make much sense to try to get us much people to join as possible, if we know that their experience will suck. The users retention rate will be low and the brand of Steemit will suffer.

It would be much better to focus on the user experience first and let new users buy their own accounts. That would make the whole ecosystem much more valuable faster than focusing only user signups.

let new users buy their own accounts

Have to disagree with you there. Facebook and Twitter didn't grow to millions of users by requiring an initial investment. I think requiring an up-front investment would kill growth.

The value of connected systems often comes from the network effect and who can get it faster. Craig's List isn't a great UX/UI, but they have the network effect lead. Steemit needs exponential growth in order to have a chance at overtaking the incumbents. Yes, some will sign up, have a bad experience, and leave. It's a numbers game. How many will we lose based on how many will we keep? That should be how we prioritize things. If we get the network effect value, those who left will eventually come back, hopefully to a better UI by then.

Facebook and Twitter didn't grow to millions of users by requiring an initial investment.

Steem is totally different ecosystem. First of all, this is a blockchain and there is a cost for creating a new account. FB and Twitter can create as much accounts as they like, basically without any cost. We can't.

I think requiring an up-front investment would kill growth.

Just look at how much money people are throwing in the blockchain world. A Steem account is valuable cryptoproperty, it's not only a login name to a social media site. We should let our users know this, rather than acting like a Steem account doesn't have any value at all.

And of course I understand how important it is to have way to get accounts for those who, for some reason, don't want to or can't buy an account. In an earlier post I proposed that SBD should be replaced with Steem account token (one SAT gives a right to create one account) and users could earn them as rewards. That would make the system more viral and decentralized.

The value of connected systems often comes from the network effect and who can get it faster.

Read the OP again. ;-)

I know very well the network effect. But it's too simplistic to think that only thing that matters is the number of users. It can turn to a negative thing, too. That leads to chaos.

Will the mining of steem with miners reduce the overall rewards? If not, is the inflation rate going up in the process?

I'm not exactly sure about the current system, but I think block producers have their own rewards which are not connected to post and curation rewards.

If you mean POW-mining, that's not happening anymore in Steem. This is pure DPOS blockchain now.

Great post and all four of your ideas need to be implemented. I can't imagine how chaotic Steemit will be with a couple million users. Would be nice to be able to direct message a user as well.

Hi @samupaha thanks for this most useful article. I didn't know anything about Hardfork 20. I've only been on steemit a week but I have already identified most of the issues you mention with scaling - particularly around the 'noise' issue.
How do you find useful content. Definitely need to deal with language filtering. Communities would be most useful, and feed customisation. There is some great content on steemit but it is hellishly hard to find it!
Thanks for the headsup on the new proposed HF20 developments.