New to Busy?

Steemit and the Fractal Society


2 years agoBusy13 min read

This post evolved from a discussion with @smooth hosted in the comments section of the controversial "Operation Clean Trending" post from @heimindanger.

The discussion was triggered by what appeared to some as "hijacking" of the Trending page by posts reaching that position through extensive use of paid votes from bots. The position of the author and a group of other people was that because the rewards seem exaggerated, people should flag those posts in order to decrease the size of their rewards.

A similar position is set out in "The Heck is going on with Steem" by @guyfawkes4-20

I totally agree with the analysis and diagnosis of the problem. However I strongly disagree with the conclusion reached by both authors and their proposed remedy.

That the Trending page is (metaphorically) "sick" is hard to dispute. I do share the view that it has been hijacked and it's being manipulated with the use of voting bots in order to entice people to "jump on the bandwagon" and upvote the promoted posts, then follow the featured authors and maybe also resteem those posts for even more audience and upvotes.

What I disagree with is the proposed "cure" to the "disease" afflicting the Trending page.

According to these authors, anyone who, after reading a post, deems, in good faith, that the reward the post gathered is higher than what it should have been (in anyone's subjective opinion), should flag (= downvote) those posts. Id est each one of us should use (some of) our voting power in order to decrease the payout, thereby returning a part of the rewards for that post to the reward pool.

I am afraid that, if people start implementing those ideas, the steem community will take a path of censorship and internal strife that could have grave consequences for its medium-term development.

For those not aware of this important discussion, I recommend you read at least one of those articles and browse the comments that it attracted in order to make your own opinion.

I am convinced this debate is central to the working of the steem blockchain system, now and in the future.


I will begin developing two theses below:

  1. Blockchain based systems (such as steem) allow the emergence and evolution of ad-hoc societies (you may also call them communities). Being populated by the same species (humans) as the real societies, these virtual mini-societies reproduce, at a smaller scale, in a "lopsided by design" manner, and much faster, the evolution of the human society, not unlike the way a fractal pattern grows.
  2. Because the human societies have endured millenia of all kinds of shocks and developments, it is quite likely that what we came up with in the real world is, within the boundaries of our knowledge, about as good as it gets (sorry for the utopists believing that communism or cryptoanarchism would be better). Blockchain and decentralized systems cannot and do not suddenly render irrelevant 7000 years of refinement in human social constructs.

What they do though is provide us with an incredibly fertile ground for experimentation, in order to discover, learn and make social progress thanks to new beneficial rules or institutions which we might possibly uncover. They offer incredible social and economic laboratories.


Therefore, the faster we realize how far we got already and reintroduce inside the blockchain system the best our current societies have come up with (after 7000 years of experimentation), the sooner we will be able to set out to discover what remained hidden in the real world for lack of a proper "socio-economic laboratory"

source @lishu

"Quis custodiet ipsos custodes"

In order to avoid a boring academic approach, I'll use @smooth's comment as a prop:

Yes that's exactly right and people should be (and must be) skeptical of each other. In a decentralized system the answer to "Quis custodiet ipsos custodes?" (Who Guards the Guards?) is "each other". Our voting decisions are all open to scrutiny and disagreement by each other. That is the only way this system can work.

As its latin expression indicates, the question "Quis custodiet ipsos custodes" is several millenia old and is still actual whenever a group of people have to organize.

I read @smooth's position as "if we want the system to stay decentralized, then we should each assume the role of custode and question each other's voting decisions.

Step back for a second to reflect on that "shot from the hip" proposal. It starts by flying in the face of "division of labor" and "specialization" which are the foundation stones of the progress of humankind.

It also assumes that we are all equally qualified to judge whether a reward of, say, $200 for a text is "higher than it should have been" regardless of whether we are
a. a young crypto-millionaire and Steem whale from the US East (or West) Coast or
b. a young guy from Aceh, Indonesia or Caracas, Venezuela, or Lagos, Nigeria

Human societies have discovered prior to the establishment of what you call today proper civilizations what a poor advice that is. Proper civilizations, from the ancient Egyptians to the ancient Chinese had already figured out that having everybody policing everybody else was an unworkable system, prone to be consumed into a spiral of self-destruction.


Later we realized that having the most powerful (the "whales" of each historical civilization) do the policing according to their judgement and whims was not an ideal system either. And all the way in-between, various societies have tried out a great number of other combinations.

"Who guards the guardians" is still a central question to every human society / community and the bothersome truth is that ... it has no simple answer. The quick answer "everyone should do that in a decentralized system" only underscores a serious weakness of decentralized social systems.

Vulnerability is intrinsic to human societies

The theory goes that human society started out decentralized, and it evolved from there. Because we are not all born with the same abilities, it was rapidly obvious that with two guys, a brainy and a brawny,

  • having each make half the decisions and perform half the physical work
    led to markedly worse outcomes than
  • letting the brainy make the decisions and the brawny do the physical work.

That was called "division of labour", allowed "specialization" and was pivotal to the development of human society. Of course division of labour and specialization provided no free lunch: soon "the brainy" realized that they can abuse the trust of "the brawny"...

So the society took matters from there and developed mechanisms that included all the dimensions of what it means to be human and gradually progressed from "free-for-alls" to tribes and clans, to kingdoms, empires and republics.

Today, the most advanced societies, those that allow their citizens

  • the highest level of personal expression ("freedom")
  • the highest level of satisfaction of human needs (physiological, security, belonging, respect, self-betterment)

are mostly modelled on republics, even when they nominally have a (mostly decorative) monarch and call themselves "monarchies".

And also that conversely, other models of society have mostly been absorbed or still linger as living proof that, even if our modern republics are far from perfect, nothing else we've tried till now works nearly as well.

Yet republics are still vulnerable to abuse of trust by those in positions of power. If such vulnerability needs to be severly reduced or even eliminated, going back to decentralized is akin to "throwing the baby with the bathwater".

The problem with trust

We, humankind, made a great deal of progress in the past 7000 years and accomplished a lot. What we still haven't managed to solve is the unwanted side effect deriving from division of labour.

Even after 7000 years of living together and helping each other progress, some people, usually calling themselves "the brainy", keep finding ways to abuse their specific roles in society, to abuse the trust of the others, that they usually call "the brawny" (or sometimes "the thick ones").

This was basically the situation of the world in 2008 when Satoshi published his bitcoin paper. In the big fractal development of humankind, we started again, thanks to bitcoin and other blockchain systems, from a decentralized situation and on this newly created crypto / blockchain branch we are rediscovering (at greatly increased speed) the same basic truths that our ancestors have discovered millenia ago.

Bitcoin as peer-to-peer electronic cash system is decentralized and that is a huge strength. The blockchain technology that underpins it has made possible more advanced software designs, such as Steem, that can function as the central communication backbone of an ad-hoc society, which includes a currency but is A LOT MORE than a mere cryptocurrency system.

Steem adds around the digital currency a system of human social interaction that goes beyond the strictly monetary ones. Steem builds upon the concept of peer-to-peer electronic cash to create an ad-hoc mini society. We are way past the simple system of transfering value and deep into the complexities of multi-dimensional human interaction, economic AND social.

And when looking at a virtual society, 7000 years of social evolution tell us, with all their weight, that if decentralized means no division of labour then decentralization is an inferior organizational scheme that is better replaced.


What I find thoroughly exciting is that now, for the first time in human history, we have, in the blockchain, a powerful tool to address that deeeply wounding side-effect of division of labor and of the subsequent need to rely on and trust others - namely the risk of some of those others abusing your trust.

Economic and social action

I would like here to get back to another of @smooth's comments where he says:

There is no 'free money' here. It all comes from investors who buy in. Unless limits are put on those who want to take that money and put it straight into their pockets, investors will turn off the lights and the party will be over.

Indeed, the money comes from the investors, I am one myself. It is thus essential that one asks: what are the investors buying ? What is the potential, what is the vision that the investors are willing to take a bet on ? Is it for STEEM to replace bitcoin as another, reputedly better "peer-to-peer electronic cash system" ? Is it even for Steemit to overtake and possibly dethrone Reddit as one might think was the main driver of its founders, since Reddit is being used as reference throughout the Steem whitepaper ? I bet the answer is "no" in both cases.

Steem has already proven that it can be much more than that, that it could go beyond classical "social media" and, by adding value-creating economic interactions between participants, it could claim the title of possibly first and at any rate best socio-economic medium In other words, a mini-society, specialized around content creation and curation.

"Economic actions and social relationships are deeply intertwined" - Karl Polanyi

This is why I am so afraid of this proposed solution: by focusing overwhelmingly on the economic aspects (the problem with the reward pool being manipulated), it risks breaking the delicate balance of social interactions. In other words, it might improve the monetary part of the machine at the cost of breaking the social part of it and therefore breaking the whole vision that could justify the valuations we all want Steem to achieve



I contend that for Steem to succeed at the scale we all dream of, we should all root for it, to fall not in the realm of pure monetary theory or even pure economics but rather in the larger scope of Economic Sociology (check the source of the above image for a quick introduction).

By being a system that, for the first time, allows people to experiment with alternative societies, by being a study topic and lab for economic sociology, Steem could be bigger than Facebook.

Modern economic sociology was defined by a seminal paper, "Economic Action and Social Structure: The Problem of Embeddedness", M. Granovetter, 1985. A two page-summary of this paper can be found here. This paper elaborates on the "embeddedness" concept which states that economic relations between individuals take place within existing social relations and are shaped by them, as well as by the surrounding social structures.

It is particularly relevant to this discussion that one of the shortcomings of previous models and theories was "how to account for a given level of malfeasance in the economic life of a community". With the "embeddedness" concept, Granovetter successfully introduced a new approach, stressing the role of concrete social relations and structures in generating trust and discouraging fraud and malfeasance.

I conclude that the first and most pressing issue is to incorporate in a new version of the whitepaper and then in the code the explicit drive and necessary tools for combating fraud and abuse. Combat that the current version of Steem openly discredits at page 15 of the whitepaper.

source: STEEM whitepaper, 2017 version

While eliminating all abuse is not realistically possible, it must be one of the most important goals. Because one of the main value-creation sources of Steem is the amount of trustful social interactions and social links that it will generate among its users

I also conclude that correcting the problem manifesting itself in a rotten Trending page should only be done holistically, through a well-considered change of the current system, rather than by using existing tools which are too coarse. The good that such tools will do to the economic health of the system will come at the price of serious harm to the embedding social structure.

Then I stop here for the time being to let readers reflect and debate what might be a somewhat provocative position to a lot of whales, many of whom have come as young bitcoin- and crypto-anarchists with a strong passion for breaking as clean as possible from the current social systems; and also with an exacerbated memory of the bad last 10 years.

10 years out of 7000 of painful human experiments with social constructs.

I could go on to argue at length why applying dogmatically the "decentralized" mantra of bitcoin to Steem, a system far more ambitious and complex than the former, is akin to condemning it to fail in its ambition. But I prefer to let 7000 years of social progress sink in and speak in my stead and invite all the ex-bitcoin-now-steem-whales to enjoy a quote from the Master:

"If you don't believe me or don't get it, I have no time to try to convince you, sorry!"

Satoshi Nakamoto


Sort byBest