I know the word "decentralization" is and always has been a fictional concept on the Steem blockchain. But I think it is still achievable (even now). We actually have more than one kind to consider:
- Decentralization in technology: code/development/devops
- Decentralization in utility: users/apps/community
- Decentralization in governance: stake/block production
These areas and sub-areas tend to tend overlap in certain ways. Technology enables utility. Governance checks technology. Utility creates demand for the other two. These are only a few ways they are connected to one another.
When we talk about decentralization, very often we are referring to stake distribution. And of course, that kind of decentralization has never been achieved on Steem.
Successful crypto projects tend to even deemphasize the importance of stake decentralization because once achieved, it's somewhat difficult to re-concentrate stake into one stakeholder without making a lot of waves.
Properly distributed stake is very important, no doubt. It provides the multiple wills that strengthens desired network effects. But it's not the only kind to focus on. What do the others look like? Does it matter if there's not distribution in these areas while distribution is lacking?
I think it can matter, actually. Just like there's a theoretical efficiency for thermal-dynamics as compared to actual efficiency. You measure your engine against the theoretical efficiency to see if you're improving a design over time.
Indeed, for a time, distribution was improving on Steem (although, perhaps debatable). But if it was, why was it improving? I believe it was because utility and governance decentralization was improving, for a time. Sometimes it was stagnant. Sometimes not too stagnant.
We've had a few setbacks in technology decentralization. We've always relied on one or two groups to maintain the code. Even running full RPC nodes is very limited.
Although it's our best aspect, we've even had a few setbacks in utility decentralization. This is where marketing helps because marketing highlights what problems the technology can solve.
My point is: When you improve one kind of decentralization, it tends to drag the other kinds of decentralization in the same improved direction, albeit slightly.
For example, if you create a new kind of utility for the platform with principles of decentralization baked in, in turn, it demands more decentralized technology.
The opposite is also true. If you undermine decentralization, it tends to drag the other kinds with it.
For example, if you undermine STEEM Power (which directly undermines governance), you tend to also undermine the technology and utility, which we saw all too clearly, recently.
So how do we achieve better decentralization? Just keep moving towards the many kinds. If there are setbacks in one kind of decentralization, don't give up. Just pick up where you left off and keep moving towards another.
If we are unable to improve decentralization in stake distribution for the moment, then that just means we focus on decentralization in the other areas for a while.
A good athlete doesn't give up on their training when they have a setback. They figure out what the setback was and come up with a new training regiment with those details in mind.