Steemit Enhancement Proposal - Extend The Voting Period Over 30 Days By Implementing A "Hosting" Fee

in #steemit8 years ago

steemit

As it is right now, the design of Steemit doesn’t allow posts older than 30 days to receive rewards.

The Problem

This limitation creates the following problems:

  • low affinity for existing users: it gives only a limited time window to interact with the content, by artificially creating the impression that the content which cannot be voted on is outdated or not relevant (or frustrating people that want to vote on those posts and can't)

  • limiting interaction for new users: although new users may arrive at the platform via search engines and they find something they like, they can’t see the possibility to vote, and many may even not know they have this possibility, hence, they don’t see the benefits of the platform

  • limited revenue for writers: long tail is a relevant source of income and it’s now completely ignored

The Proposed Solution

In order to solve this problem, I propose the following enhancement: a “hosting fee” at the post level.

It means that posts that are allowed to receive votes over the 30 days have to pay a hosting fee. Those posts should have the UI updated accordingly (there will be buttons for upvote, etc) and the authors should be credited as well.

The hosting fee should be split: 50% will be burned (sent to null), 50% redistributed to the rewards pool (which means it will increase the curation rewards pool as well). The hosting fee per post should be significant, but not exaggerated (in any direction, not too high, to discourage people to pay, nor too low, making it insignificant).

Hosting fee can be activated / deactivated globally (for all posts) or individually, for each post.

Example: Right now there are roughly 1000 posts written each day. Out of them, only 100 may be subject to this strategy (the rest are memes, announcements or other stuff that may not be relevant 3 years from now). If the hosting fee for those 100 posts will be 0.01 Steem / day, during one year the total spent for them will be 0.01 x 365 x 100 = 365. Multiplying it by 365 days (assuming each day the authors will produce, on average, 100 posts worth of hosting), that gives 133,225. Out of which 50% is burned 66.612.5) and the rest is allocated to the reward pool.

For an author, the cost per post, per year, will be 3.65. The potential reward can be between 2x up to 10x (based on my experience with long tail).

Benefits

  • increased affinity for the platform - more people will be exposed to voting and that will increase engagement
  • increasing potential reward - long tail posts will tend to perform better
  • inflation limiting factor - half of this hosting fee will be burned which will take out from the market a significant amount of STEEM (think of this burning process like paying now for promotion, the STEEM you pay right now goes to null, and you get exposure, with the “hosting fee” you send STEEM to null in order to access the voting feature).

Please leave a comment stating your opinion on that. If you agree with this proposal, please resteem it so we can reach a critical mass.


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

Sort:  

I actually love this idea because much of my content is what I might consider "evergreen posts". But I think I've probably conditioned myself to let the 30+ day posts go because frankly, no one even clicks on yesterday's blogs unless someone else resteems it. I've resigned myself to directing people to those "archived" posts when appropriate and consider the accrued vests as my "residual income"

Would I pay to have those posts active? Certainly! But it would be nice to have a profile page that highlights MY favorites for incoming newbies.

Im a big fan of this idea. I've heard similar concerns from people on the 30 day limit for payouts. I think it's a reasonable approach even if numbers are tweaked as needed.

To some degree we'll need better searchability on the site for many users to dig it up and/or a need for search behavior to goto Google or custom search (which can be done. )

I'd easily see this leading to being a part of a paid vip account in the future. Make free packages for casual/ new/more basic users with a paid tier for more aspiring established authors with perks of so many "long term" payouts available, some spiffy visual vip designator/badge or whatever other benefits...ideally for fiat or SD. While i feel keeping things free for basic account, why not offer subscription benefits those those looking for that.

I wish adsense could just be inserted when payouts ended currently, but don't think it's possible for data to get out from steemit for that right now (i might be mistaken on this limitation, assuming from page view counters from external sites not being possible. )

Thanks for the support, and interesting points about that freemium model. It might be regulated as freemium per se, or it might be just self-regulated, i.e. people who want to pay for getting the long term benefit will do so, while those not so keen on that (or who are just writing timely content, nothing wrong with that) will choose not to pay. As for inserting AdSense into Steemit I don't think it's a good idea, regardless of the technical part. It just doesn't seem aligned with the site principles.

I can see not wanting adsense included. Only reason I mentioned is since I'd heard of it being discussed. I'm not stuck on that idea.

A great idea @dragonsroua !! great to think out of the box on this !! Thats the way !! Steem On !! upvoted and resteemed !!

I suspect something like this will be done eventually. But I think the strategy up until now has been to motivate Steemers to post as much as possible to get Steemit full of content, for Google ranking perhaps.

Would this cause people to post less, knowing they'd constantly be making Steem off old posts? Would more people take more time to polish up their posts?

Because there's a fee to expose your posts to the voting activity (and to revenue), I suspect people will be more careful about what they write and about how much they pay and for which posts. Eventually, the quality of content will go up and that can't be wrong.

That is actually a good point. I think this idea should be further elaborated.
Many content is not outdated after 30 days, some even gets more valuable.

Let's spread the word, then, and see if there's interest from the owners / devs too.

Good Idea.
I'd said something similar earlier although I called it a Legacy Pool.
It, or something like it definitely needs to be done.

Yeap, thanks for the comment.

I'm a wanna be author. Steemit fits my writing style (episodic).
It pays, which is better than no pay.
But the 'long tail' or 'backlist' would be nice.

Interesting. I like the idea of people posting for free, but paying to extend it sounds cool.

My understanding that it's limited to 30 days is because the way the graph database behind the scenes stores all the active posts in ram. So they limited it for technical reasons.

Not necessarily. Voting will not influence the post itself. Technically it can be done, or that's what I understood.

"Voting will not influence the post itself"? I thought the whole point of Steem Power and voting was what made posts be worth somthing? Now I'm confused lol.

It's from a technical point of view. Voting will not modify the post, like editing it will do. Voting will just allocate a portion of the daily reward pool to the post author.

Oh I see what you mean. Modifying the actual post itself. Yeah I don't understand why everything is stored in ram for active posts and their votes. I used databases such as MySQL, SQLite and MongoDB before and they seem to utilize disk. Memory usually for indexes and temp things.

I know I was talking to someone about this on dev channel on Steemit Chat a few months ago. Whoever answered said it's because of the graph database. I was wanting to know why they said witnesses required ram would double at some point.

It's a great proposal. I'll bring up the four concerns that I am aware of though:

  1. If the posts can be voted on, they need to be held in memory by the witnesses and steem nodes. The more content that needs to be held in memory, the more expensive it will be for witnesses and other entities to run servers to power the blockchain. The new more efficient graphene blockchain that they are planning to roll out may help to address this issue though.
  2. The rewards pool is a fixed amount of money to pay out, which has to be divided across all the posts that are receiving payment. If older posts are receiving large payments, that is less money to go to the 24/h and 30/d posts.
  3. At least currently, most posts are not getting any attention after the first 24 hour payment period. I assume the # of posts that would get significant upvotes after 30 days would be even less than that.
  4. Users are only supposed to be able to vote on a post once. If you vote on a post in the 24 hour period, it won't let you vote on it again in the 30 day period. This is for obvious reasons. If we allowed voting indefinitely, malicious whales would be able to exploit the system by voting on "puppet" posts, then powering down and powering up a new account, then voting on them again. They could in theory milk a historic post for new rewards indefinitely if nobody noticed what they were doing. (I saw this brought up by @smooth in regards to the shorter power down they are proposing - which would make this even easier to do.)

Assuming they can find ways to address these items, I think what you are proposing would be a great way to implement it!

Thanks @timcliff, appreciate the time taken to peruse this. I'll take them one at a time.

  1. Technically speaking, is very doable. They can keep only the reference to the post in memory and the actual post on a separate storage, strongly linked by some keys. I don't see any reason to keep blog posts in the blockchain (maybe just to show that you can). I see reasons to keep smart contracts in the blockchain, but blog posts aren't smart contracts. Also, by taking out the actual content from the blockchain and keeping only references, I expect the storage needs will be 100x times less.
  2. In this scenario, the reward pool si constantly enlarged by the contribution of people who are paying to have their posts votable. So it will actually be more and more money on the table.
  3. I tend to disagree. Long-tail is a significant source of revenue for many bloggers.
  4. This is already happening, there are cohorts of bots voting. My proposal suggests to have rewards only for authors, not for curators, as it is right now after the 24 hours period. It makes sense to pay for curation if you discover content, but when the content is old, you don't have to pay for curation, you only reward the content with your vote.

So it's kinda simper. What's your take?

  1. The blockchain code is not my area of expertise, so I don't really know if what you are suggesting is viable. There is a discussion going on in this GitHub issue with the developers that are working on a solution. You could offer your suggestion there: https://github.com/steemit/steem/issues/339
  2. If people are making a lot from historic posts, then in order for them not to be a drain on the reward pool, more money would need to be taken in by the "people who are paying to have their posts votable". I don't know if the fees would cover the rewards. If they did, then one could argue it probably wouldn't be worth it for the authors, since that would mean they are paying more in fees than they were getting in rewards.
  3. Fair point. With the current viewing/voting patterns of the users older posts are not getting much attention, but that could change.
  4. True, there are already ways that users can game the system. This would just open up another attack vector that the community would need to consider.
  1. Thanks for the link on github, I knew about that discussion, I will open a new issue.
  2. That's exactly the idea, people will become very conscious about what they can make by having their posts votable and it will very soon become self-regulating. I think that a relatively small percentage of users will have their posts votable over 30 days, but that will still be significant. I did a few simulations yesterday, it looks interesting.
  3. :)
  4. Yes, implementation should be as tight as possible. But as I see it right now, it's quite ok.

Thanks for the contribution to the debate, again.

This post has been linked to from another place on Steem.

Learn more about and upvote to support linkback bot v0.5. Flag this comment if you don't want the bot to continue posting linkbacks for your posts.

Built by @ontofractal

why the hosting fee?

Whats wrong with just voting whenever?

Many reasons:

  • high computing cost of keeping the articles in memory indefinitely
  • increasing the reward pool
  • burning steem for advertising
  • if you pay for something, you control it, if you don't, you have no control over it

This makes all perfect sense to me.

However I am thinking of this from a user perspective. They want to create content and then get rewarded when it is consumed. The hosting fee makes it a bit complicated. I think it is critical to create a better experience for users on steem vs facebook, reddit or youtube. Currently while you can get paid here it is hard to find content and to get rewarded for content long term.

In my personal experience, every time people pay for something, the value of that something increases. Every time they get something for free, they tend to diminish that something. Not to mention that in sites like Facebook or Twitter, the fact that you don't pay for using them makes you, as a user, the actual merchandise that is sold to advertisers (which, honestly, won't be a terrible problem, if you will get something - alas, you don't get anything, Facebook gets it). So, I stand by my words: this hosting fee will create more value for everybody. But I'm always ready to accept that I'm wrong, if reality proves it so.

Coin Marketplace

STEEM 0.19
TRX 0.14
JST 0.029
BTC 65771.81
ETH 3174.77
USDT 1.00
SBD 2.61