There is a daily pool of funds that is distributed across posts, comments, votes, etc. See @inertia's comment in this post for an example of the daily pool as it currently looks.
Your steem power determines your voting power.
When you up vote something at 100% it applies 100% of your steem power as your vote. Each vote reduces how many votes you can cast for the day. You can extend this by not always voting 100%. This % regenerates over time. This is to stop people from simply gaming the system by posting a bunch and up voting their own stuff continually until they drained the pool. Each successive vote gets weaker unless they let it regenerate by not voting for awhile. If you are careful with your personal vote this really should not be anything you need to worry too much about. I vote on a lot of comments so me personally I tend to vote comments 15% to 20% and in some occasions I'll vote more if I really like the comment. I tend to vote 100% on posts. Some people are a lot more focused on monitoring their votes than I am.
So your steem power and all the other steem powers are totaled. That is applied to that reward pool to determine the POTENTIAL EARNINGS of the post. So the value can fluctuate as more steem power is voted on around the platform on other things. It can also fluctuate depending upon the market value of steem.
So the down vote currently does the opposite. It acts as negative steem power and in a sense can reduce how much of the pool that post is using and thus frees up some to be spread against all other posts, comments, votes.
The crux of the issue here and why we have these debates is this:
You work at raising your steem power so you can better reward posts you like.
So you and 10 other people decide you like say FISH EYEBALLS as an appetizer and post recipes for it. You get the post up to $10 using your combined steem power. You guys really like your fish eyeballs.
Someone with more steem power currently can come along and say "eww... fish eyeballs, that's not worth $10" and down vote it. If their power is more than the 10 combined people that price can drop to $0.
The issue here is that some of us (including me) do not think it is a good thing to be subjectively deciding what other people are allowed to like and support with their steem power.
Yet, if it took time and effort to make those FISH EYEBALL posts how likely is the person to continue doing so after they've had funding removed, and they perceive they have been attacked? They often will get unhappy about it.
So many of us (myself included) have begun to state we don't believe decreasing payout for something because you think it is valued too high is actually good for the platform. It may have immediate short term benefits in spreading steem power around, yet the negative impact is long term, and while it can be argued it is not censorship as it is still on the blockchain, and the person can continue to post for free if they want, etc. It has a similar end result as that of censorship in many cases.
We want good content, yet we only reward it if it is content we agree with or like. It has been proven that OFTEN this down voting has absolutely nothing to do with quality of the post.
There have been alternative proposals that I'd be interested in seeing people TRY before they say it will not work without even trying.
@neoxian had an idea I am in favor of trying now as it requires no code changes.
If we TRY it out for a week or so and can measure the impacts then we'd have some backing evidence rather than assumptions and opinions. If it doesn't work then we can start learning from it and keep working to make things better. It is in beta, so now is the time to try things. I'm game to try anything as long as we are willing to take a step back if something doesn't work.
@neoxian said instead of down voting a post, why not find 10 other posts (or any other number) you think are deserving and up vote them. This will spread the pool around too and will not be a perceivable subjective attack against a specific topic, person, etc.
Sure overall payouts would be lower, but I suspect people would not notice this as much as they notice a down vote that some perceive as "stealing" payout, censorship, or an attack. Why risk these negative perceptions at all if we can avoid it and get close to the same effect?
It has been argued that this method is not as effective/efficient as a single down vote. This may be true if you are only worried about the short term goal of that day. If your long term goal is as mine which is to get more people to the platform and hopefully have them want to stay then I believe it will likely prove more EFFECTIVE than the down vote in that regard.
Hard to know without everyone agreeing to try it out for a week or so and letting us compare and learn.
EDIT: Added a word to correct an absolute.
One of the great challenges of (effective) peer-curated content is also about being able to discern the difference between "of value to the greater community" and "I personally disagree with this."
If I can't stand rats, and someone creates a brilliant illustrated essay on the miraculous beings that rats are... do I downvote it for being disgusted and not liking it, or upvote it for being really good content about something I can't stand? Curation is not always as simple as it seems...
The whole "downvoting on the grounds of excess rewards" remains a really weird fish for me to understand. The first thing that comes to mind is whether I-- by the same token-- get to pitch a fit when my piece I worked hard on for three hours is earning $0.00 after 12 hours? Obviously I can, but it's a pretty meaningless exercise.
Don't have any easy answers, though... although I'd submit that IF there's to be downvoting based on excess rewards, it should be both contextual and on a case-by-case basis. If someone works really hard and gets minimal rewards... and then finally publishes something that "takes off," only to have rewards chopped... that's pretty discouraging. On the other hand, someone who seems constantly able to sit at the top of "featured" without creating solid content might need to be reviewed.
I think one of @krnel's arguments is that if there is a TOO MUCH then let's define that. So that it is not okay for one guy to make $150 as long as he is writing about sports, and not okay for the guy writing about FISH EYEBALLS to make $10.
If we are going to call something TOO MUCH then perhaps that should be defined rather than arbitrary.
In talking to people that do down vote for this reason they usually cannot explain what TOO MUCH is. Really it came down to something about the topic, the person, who voted on the person, etc they don't like. That seems to be the common denominator for some of them (not all).
Like trends where if @dan voted on something they would Down Vote it and state it was due to being overrewarded even if @dan's vote was small.
I argue against this behavior, and it has never personally happen to me.
I've seen it hurt other people psychologically. I've seen them think the community didn't like them and that maybe they were not doing as good as they thought they should. I've seen people leave. I've seen negative PR from it outside of steemit.
All of these seem like very bad costs if the supposed reason is to protect the reward pool. There has also been quite a bit of hypocrisy from some of the people claiming to be protecting the reward pool.
One of them would down vote people they didn't like with reward pool as an excuse only to be caught up voting their own comments to $8+ in multiple cases.
Ultimately though my interest is growing the platform. If we can avoid negative PR by simply approaching things different then I don't know why we wouldn't try that.
We should be focused on long term goals rather than letting short term reward pool policing hurt those long term goals. They may work out great for the day in question, but long term results of negative PR could make any monetary value you saved from that policing be dwarfed by the long term loss.
I think it's just too long for a layman. :)
Likely true. Yet nothing is stopping someone else from answering them. I waited and no one responded. I am not very good at NOT being verbose. :) In other words I suck at twitter. ;)