There is a lot of talk lately about the “level” of centralization in Steem / Steemit. The general consensus seems to be that, in its current shape, the governance layer of Steemit is centralized. Which will contradict, somehow, the goals of the organization: we’re a crypto project, hence, we’re here to decentralize monolithic structures, to empower back the people, etc, etc.

The Unpleasant Wake Up Call

I may upset some of the prominent voices in the Steemit ecosystem (wouldn’t be the first time, mind you), but my opinion is that decentralization - in every crypto project, not only Steemit - is a myth. A goal that most projects strive towards (or claim they strive towards), but something that is impossible to be attained.

And it’s not about the technical solution. This is not PoW versus DPoS, because centralization occurs equally in PoW and DPoS (or whatever other governance algorithms you may find).

Have a look at Bitcoin, for instance. It’s the grand-grand-father of crypto, a PoW governance that aimed to eliminate the “middle man” and empower “everyday Joe” to be a part of a distributed, fully decentralized bank. Alas, everyday Joe is out of the game completely, unless he’s ready not only to spend a significant amount of money on expensive mining equipment, but at the same time to “vote” with his purchase for one of the mining hardware companies out there, empowering that company even more (by sending more revenue to it, and generating the potential of collusion in bigger, centralized clusters, for those using similar hardware). As you see, PoW sounded feasible in theory, yet somehow it borked.

We’ll see down the road what are the reasons Bitcoin is still considered “decentralized”, but for now let’s all try to keep our eyes opened and look at the facts and not at the theory. In theory, everything is nice and easy, but practice has this upsetting effect of being a pain in the ass.

Back to the Steemit ecosystem: here, the governance is enforced through DPoS, which means greater stake controls greater parts of the block generation. The distribution of this power to generate blocks (and be rewarded for that) is imbalanced, meaning there is a core of 21 witnesses, which are producing the majority of blocks, and the rest of the participants are functioning as backups. It so happens that the core 21 witnesses are a very centralized structure, gathering the votes of the majority of big stake holders. Not only that, but pseudo-anonymous actors, like @freedom, have a bigger say in this process, sometimes negating public votes from non-anonymous actors. I won’t go into more details about that, just wanted, again, to look at the facts.

What Is, In Fact, Decentralization?

Decentralization means, at least in crypto, distributing the trust that protects some assets on a much larger surface than usual. Assets may be existing values (like tokens already mined) or futures (like future rewards for block mining).

In a centralized ecosystem, the trust is delegated to a monolithic (and very often, opaque) actor, namely a bank. Decentralization aims at unweaving and redistributing this monolithic player, to give it a much finer granularity.

In theory, decentralization promises that one participant will have, at any given moment, equal chances with any other participants in the network.

In practice, though, whatever surface negates the potential of a middle man is considered decentralization, wether or not trust is equally divided amongst all participants. This is probably the most common mistake - or I should say wrong expectation - that populates the crypto space.

Back to our examples, Bitcoin is considered decentralized because the surface to which the trust is spread is considered safe. At the current level of centralization, a 51% attack is not possible. It may happen at some point, but for now we’re safe. It’s worth noting that in other Bitcoin clones (BCH, BSV, BCP, etc) this 51% attack is way more plausible (and probably happened more than once).

Incentives Aligned = Decentralization

So, centralization exists in all crypto currency projects, that’s a fact. And, to get to the core of it, centralization exists wherever there is a group of 2 people. One of them would want, at some point, to assume the leader position and to make the majority of calls. If you don’t agree with that, then it means you’ve never been married. Or at least witnessed marriage from within, somehow.

So, the problem is not how equally people are participating in the trust spread, but how aligned their incentives are.

If their incentives (or expectations, or goals) are aligned, even if the structure is heavily centralized, then it will be accepted as functional. If their incentives are not aligned, then the structure will be deemed as not functional and amended until a new equilibrium is found.

What Are Your STEEM Incentives?

So, instead of trying to fix human nature with tech and complaining about the “horror of centralization creeping into Steemit” let’s just accept it as a fact and shift our focus towards incentives.

This structure (any structure, in fact) will work if people have their incentives aligned. That means - among other things - that the leader role is becoming more of a managerial position, a position that will just supervise whatever incentive-aligned members will decide upon. There will be centralization, though.

So, the questions are: what exactly Steemit INC wants from the Steemit ecosystem? What exactly want witnesses? What exactly want users (the ones who are actually contributing content)? What dApp creators want?

I think the current friction that we’re experienced can be solved not with complaints, hashtags to stop power down or start power down, but with a simple, yet transparent process of aligning our incentives.

I will start with the answer to the last three question myself (feel free to add yours in the comments, or, if you want to make a dApp for it, go ahead, I’ll join you). I will also expect some answers to the first question from @ned.

As a witness, I want to be able to run a node with decent costs, with audited code, to have a say in any hardfork that comes around and make sure I can deliver 100% uptime for my contribution to the network.

As a content provider I want to be able to leverage my skills in finer structures (ability to create and manage communities).

As a dApp maker I want to be able to build stuff on a stable and efficient codebase.

---

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.

---

Wanna know when you're getting paid?

	I know the feeling. That's why I created steem.supply, an easy to use and accurate tool for calculating your Steemit rewards

It's free to use, but if you think this is a useful addition, I'd appreciate your witness vote.

Thank you!

Psst: new to Steemit? Start Here