Like many of you, I'm a fairly recent arrival to the platform. Over the last week, I've spent quite a bit of time playing with the various features of the site, and trying to get a handle on how the site and reward structure works.

And although I'm still very enthusiastic about the future of Steemit, I have a significant concern about how the rewards are structured here, and what that means for the future of Steemit.

Follow me, and I'll explain...

The core value of social media

The core value of any Digg/reddit style social news aggregator is in leveraging the efforts of a large group of people both to source (submit) and filter (upvote) content for relevancy. As input, content posters both find and create new content. Content curators help filter the wheat from the chaff to ensure that the most relevant content gets the most visibility. The end result is that the cream of the crop rises to the top -- and the front page of the site ends up with a stream of timely, relevant, highly-recommended content, providing huge value to end-users.

These site are popular precisely because the ability to crowdsource the sourcing and the filtering of the content allows for far more depth and breadth than classically editorialized sites can produce.

The life-span of content submission

When I submit this post, it will go into the new posts list, where all posts are sorted by submission time. After submission, all posts have a narrow window of opportunity to get noticed, which is time-bound by the number of other submissions being made. During this window of opportunity, if a given post accures enough upvotes, it will move to the "next level", where it is given additional visibility and the opportunity for more upvotes.

We can divide the site into three levels:

• New -- this is where all content starts. The average quality of submissions is low, and the quantity of posts is high. It's hard to get noticed.
• Active/Popular/Hot -- this is the second level. The average quality of submissions is better, and the quantity of posts is moderate.
• Trending -- this is the holy grail, since trending is the front page of the site, which means the content here gets the most views. That means the potential for rewards for content that appears on this page increases significantly.

Rewarding participation

Previous incarnations of such social media sites (Digg/reddit) have rewarded users with Karma for participating in the site beyond just viewing. However, Karma doesn't have much value other than as a signifier of social status and personal investment on that given site. Steemit takes things to the next level by embedding a cryptocurrency layer into the site and giving out crypto rewards for various kinds of participation.

By making the rewards themselves more relevant to end-users, the end-result should be a higher level of effort put into the site, and a higher level of output accordingly.

As documented here, there are three common kinds of participation rewards that all users can take advantage of:

• Commitment rewards -- Steem rewards users simply for being on the site. According to this page, "Anytime less than 90% of STEEM is vesting, the network gradually transfers value from liquid STEEM to vesting STEEM via interest payments". Currently, the interest payments seem to be just north of 1% per day.
• Posting rewards -- Users who post content or comments get rewarded for bringing content to the site. The rewards for doing so are highly variable -- I've seen a ton of well thought through content not get rewarded at all, and I've seen content that takes 5 seconds to post a link accrue thousands of dollars in rewards. So this is kind of like a slot machine -- you may get rewarded, or you may not, but putting in more time and effort into what you submit at least increases your odds. And the rewards here can be quite good, as we've seen on the front page of /trending, where makeup posts can earn north of $30k.
• Curation rewards -- Users who are early upvoters of content or comments gets rewarded for helping determine what's relevant. However, the overall reward for being one of the first upvoters on a moderately popular post appear to be quite poor. As an experiment, I spent two hours per day over two days reading through posts on /new and upvoting the ones I thought had the best quality, or would most likely be popular. Although most of these never broke out of /new (and were subsequently ignored), a few did. The total curation rewards for my time totaled ~0.01 Steem Power per day. It's unclear to me if this is scaled based on your current Steem Power or not (if so, this is equivalent to 0.005% of my total Steam Power).

Herein lies the rub -- if I can collect 1%+ interest per day for doing nothing, or 1.005% per day for spending two hours digging through low quality submissions, which am I logically going to choose?

The answer is obvious.

Why curators are important, and the risk we incur by not rewarding them enough

Early curators have the job of ensuring that the content that gets the most views is as relevant and interesting as possible, and they directly enable the content submitters to accrue rewards. They are the second most important users on the site (behind the content submitters), as they directly impact the quality of the content that reaches the front pages of the site.

So what happens if we don't reward the content curators enough? The (law of large numbers)[https://en.wikipedia.org/wiki/Law_of_large_numbers] can be helpful here in offering guidance. Put succinctly, the law of large numbers says that in any given experiment, the more data points we have, the more likely those data points will trend towards the expected value (the average). Or, put another way, the smaller number of data points we have, the more likely we are to have outliers as results.

In terms of content curation, if the number of content curators is reduced, then we'll see more outliers (lower quality content) being upvoted out of /new and onto the other destinations, reducing the overall quality of the site. Furthermore, with less upvotes, there is a lower barrier to entry for manipulation -- either bots or small groups of users colluding to upvote their own content.

How can we resolve this problem?

There are two easy tweaks we can make to help solve for this:

1. Divert some of the commitment interest rewards to help fund content curators. If I got a 0.5% interest reward for doing nothing, with an additional 0.5% reward for spending a couple of hours curating content, I'd be much more inclined to spend some of my time curating content.
2. Allow content curators to share a little more in the success of the content submitters. After all, the content curators are directly responsible for determining what gets seen or not -- it's not unfair for them to share in the rewards of the content they helped promote.

Conclusion / TLDR

A healthy early content curation community is vital to the long term success of the site. However, the current reward system favors inactivity and content submission over curation because the rewards for early content curation are insignificant compared to the rewards for commitment and posting content.