This article is a continuation of the series of materials (1 - Statistics, 2 - Good Scorum, 3 - Bad Scorum(in Russian)) about my impressions of using Scorum during the first month of life on the platform. This article was not planned in advance, the idea of its creation appeared just a few days ago. I warn the material is large and requires attention while reading. I believe that the community will be interested in this, and you will find time to understand it.

So, what do I mean when I talk about the centralization threat. We all know that Scorum is a decentralized platform, and all users can participate in its management. In this case their influence is limited by the size of Scorum Power and voice power. The potential threat of centralization is that owners of a large number SP begin actively do certain actions, for example, flags and upvotes. This leads to the fact that one or more users have a significant impact on the entire system. You can now tell me "Well, what are you talking about here, there will never be such case." But in fact the threat from the potential level began to turn into a real one.

Problem 1 - Bots

Once in the comments to this post was found a bot that someone tested (see the branch after the @binoalien comment: "You got a 0.33% upvote from @binoalien courtesy of @binoaliendev!").

Let's figure out who these votebots are. Votebots are robots that scan the blockchain for new materials and vote for them. Their main goal is to increase the passive income of SP owners. The scheme is:

1. Somebody creates a bot account
2. Users who wish to earn a passive income delegate to the bot their SP
3. Bot works, putting upvotes and earns SP
4. Then the income received by the bot is distributed among those who delegated SP to the bot. The distribution mechanism may be different, and for Scorum it may be concluded in the conversion of earned SP into SCR.

Another type of work may be - paid upvotes, a bot having a large SP can sell its apvotes to authors who want to "promote" their posts.

Both schemes are harmful to the community, in first scheme bot upvotes articles randomly, and in the second case the advantage is for uhhonest authors.

Problem 2 - Whales

Bots is one trouble. There is the second, let's call it whales. Whales are users who have a large SP and therefore have a serious impact on the life of the platform. As majority of you knows once the user @williams appeared at Scorum with SP about 300.000 and began to upvote comments with all his voice power. As a result, such comments began to receive amounts to pay about 1000 SP. It is not honest when a person gets 1000 SP for 5 words, and the author of a serious post gets 100 SP. The community responded almost immediately - the user @balancer appeared with SP equal to 280.000. He began to "balance" @williams' actions with flags to posts and comments reducing the amount of payout. It would seem that the problem is solved, but in fact is not. I think that the problem became bigger, if earlier we had one whale, placing upvotes, but now there are two of them. Yes, at the moment the balancing works almost perfectly and it's fine, but what if @balancer goes crazy? Or what if these two will agree and actually monopolize the platform because they have 15% of the total SP! What if they find a third or fourth? And if they all just upvote each other? In addition to theoretically unfair remuneration distribution, the presence of such consortia discourages new users who are simply lose any influence with their 5 SPs on the balance.

So what should we do?

The conclusion from this story for me is - Scorum should not give the community such an important issue as the possibility of centralization or, more accurately, oligopolization of the platform. This problem should be tried to solve in the rules of the blockchain. In addition, besides fighting centralization, we should keep in mind the goal of keeping Scorum attractive for new users in the long term.

Now a new user Scorum gets 5 SP as a registration bonus and can increase them with their active actions on the platform: write articles, comment, upvote posts. At the same time, the newbie's impact compared to a whale having 300,000 SP is 60,000 times less. For example, a user who has 1000 SP and 100% voice power adds the author about 4.8 SP. If you have 300.000 SP, then your upvote will add the author almost 1500 SP, and other authers will lose some SP for their posts as a result of redistribution. The conclusion that follows from this paragraph is that the difference in influence between users from different layers should be reduced, and users should be interested in the growth of SP.

There are many theoretical options for solving this problem. If you are interested, write some lines in the comments, and I will separately describe the analysis of alternatives, but I chose two tools of all researched:
1. The ban on SP delegation. This will avoid the consolidation of "capitals" for speculative reasons. In addition, it will improve the sense of the blockchaim as P2P-platform.
2. Reducing gaps by using a power function (power - less than 1) in determining the effect of SP size on the author's remuneration. I mean that while calculating author's renumeration size not absolute value of SP should ve used, but calculated or relative SP size.

If the first point is clear, the second one needs explanations. Now the upvoters SP effect on the payout size is linearly dependent on each other, which results in the same difference. I suggest that we shouldn't take the absolute value of SP for calculating the amount of renumeration, but the relatibe number of the SP in power that is less than one. The table below shows the relative SP values for the certain values of the power.

The bottom line is how it is now. The top row with a power of 0.5 is the relative SP values obtained by extracting the square root of the SP value. As you can see, the difference between a beginner and a person with the category "Pro" (1000 SP) is now 200 times, when using the degree 0.5 it will be only 14 times, and between the beginner and the whale is not 60,000 times, but 244 times. In my opinion, the value of the power of 0.5 is quite strict and will not encourage users to accumulate SP after reaching a certain level. It seems to me that the optimal power value is somewhere in the range of 0.7-0.8. This will increase the decentralization of the platform, increase the interest of newcomers in joining Scorum, and at the same time will encourage users to increase SP.

So, write comments and:

a) do not hesitate to ask questions about the content of the article, because it turned out to be difficult in places
b) express your proposals regarding the management of the risk of centralization of Scorum
c) tell me whether the history of the whales appearence (williams and balancer) on the platform will be interesting to you. I looked at some blocks in chain, and there are some interesting moments, but I did not dig deep and did not include information in this article, because didn't want to overload post.

Scorum / centralization threat

Scorum. Centralization threat.

Scorum. Centralization threat.

Scorum. Centralization threat.