The problem here is, nobody is interested in keeping their BTC as they are mostly viewing it as another money-making opportunity (here: get rich quick scheme) and so, it's not possible for anyone here to impose the actual mentality on them that what is Bitcoin all about. About this centralization thing, it isn't limited but a cycle that cannot be broken now (considering that everyone is looking for "regulations" to take effect as they believe that putting everything in the hands of "Central Authorities" will get their Bitcoins' value to sky high levels, which is somehow true as well).
I see and it is sad, though I have reasons to be rather optimistic:
Firstly we need to understand that Bitcoin is not a joke to be narrated by clowns or opportunists. Craig wright has spent millions to change anything about bitcoin and look where he has ended, almost nowhere. Bitcoin has a solid conceptual design, forget about the code and implementation details, the underlying basic idea, producing money by work, securing and transferring it by work and doing it in a decentralized consensus based p2p network, it wouldn't be a feasible strategy for any party no matter how powerful/delusional to take over it.
Secondly we have Core devs, still officially committed to decentralization which is a good point to start from. Don't get me wrong, I'm not a fan of these guys neither I think coding is kinda magic on the contrary I don't GAS about technical part it is about a far more important thing: HISTORY.
I mean, just look at Gregory Maxwell, he is a living encyclopedia of bitcoin, he has been involved in almost every single important event in this ecosystem and although I don't like his super conservatism and passiveism against ASICs, pools, bootstrap/synchronization nightmare, on-chain scalability, ... (you say) and I'm absolutely against downward compatibility and too much soft forking which (is his holy grail btw), no doubts he has a lot to share because he is committed to decentralisation and privacy as far as we are discussing intentions and incentives and ethics. And it is almost the same for most of these guys. All we need is finding a non-hostile decent approach to this, a compromise.
I see potentials because I see good faith.
How is it a centralization cycle?
Keep it to ASICs or even CPU/GPU mining or even take it back to those times when USBs were used to mine BTC, everything can be controlled by big firms when they've got the money and power both to buy the equipment ...
Well, it is not how I understand the problem. We have this
power law distribution of wealth , you are right but PoW is resistant against it if we could have it implemented flawless, unfortunately it is not the case right now.
We have two majors flaws in bitcoin that need immediate fix:
1- Small memory footprint of the naively
direct way bitcoin uses sha2 in its proof of work and its consequence of being vulnerable to ASIC attacks which we are discussing here and should be improved indisputably.
2- Winner-takes-all incentive system that inevitably leads to
mining variance and
proximity premium flaws which are responsible for
pooling pressure and the situation we have with pools. I have been working on this issue since a while:
An analysis of Mining Variance and Proximity Premium flaws in Bitcoin Getting rid of pools: Proof of Collaborative Work Of course there are other problems: the disastrous situation with setting up/synchronizing full nodes from scratch, spv wallets, wallets not actively participating in security of the network, ... to name few, but I think the above two problems are BIG bad flaws that has escalated centralization threats to the dangerous levels we are experimenting nowadays.
Conclusively, my point could be summarized like this:
Despite power law distribution of wealth and its natural force toward centralization, an advanced implementation of bitcoin is not defeatable easily by means of simple factors like economy of scale. Centralization is mainly a consequence of flaws in detail design/implementation of core concepts and delayed improvements.