Evan Shapiro's "Bitcoin’s Stolen Revolution" https://www.coindesk.com/bitcoins-stolen-revolution is a great opinion piece. However, he doesn't actually describe what crypto is centralization-proof, or what protocols must be in a crypto to make it centralization-proof.
I wrote a short novel of crypto fiction on Hive about a plot for centralization. I liberally took parts of the story from Justin Sun's hostile take-over of Steem--which is why I published it on Hive--but there were notable differences. @awbvious/epilogue-this-is-crypto-y-all . To read the short novel, one should start with links at the top. Chapter 12 especially discusses the centralization plot.
I think centralization-proofing can be done in three parts:
1: Governance is one vote per individual, preventing abuse by vote concentration with one person. The amount of funds do not matter. The number of accounts do not matter. A person can have multiple accounts, or more funds than anyone else, but will still get one vote. To have a vote, you must be verified as a unique, sentient being capable of giving consent. Verification will cost a nominal fee paid with the crypto itself. If the person can somehow spoof uniqueness, they cannot be verified, and if spoofing is determined, all funds are forfeited. To fast track, an individual can pay to verify, otherwise free verification slots are occasionally available and paid for by taxes. Once verified that person can be taxed, but also that person can receive benefits, either by operating a benefits-creating device themselves or delegating to one elsewhere in the world.
2: Taxes are highest on the richest, preventing abuse by wealth concencentration. One of the problems with American politics is once Reagan gave tax cuts, the wealthy got wealthier, then they realized they needed to be in politics (even more) to keep those tax cuts, then gave themselves more tax cuts, then money got deeper into politics, etc. It was, and is, a terrible feedback loop. Of course, the threat will always be there, but by starting with high taxes, similarly motivated individuals to corrupt a government will be dissuaded and set their sights on other crypto currencies. Taxes will be used for verification of people for uniqueness so they can vote and devices for location so they can receive rewards, and any surplus will be distributed among the poorest.
3: Benefits are dependent on location and distance, preventing abuse by regional concentration. The closer a device that creates benefits is to other such devices, the less rewards it can create*. Verification can be GPS or similar technology and/or by relating a device spatially to other devices. But if the device allows spoofing location, it cannot receive benefits, and if spoofing is determined, all its previously created benefits are forfeited. Further, the more potentially mobile a device, the more frequent its verification. Like voting, a device can pay to fast-track its verification, otherwise free verification slots are occasionally available and paid for by taxes. While theoretically one country might be able to own and operate the most devices, it would still be difficult for it to then amass the most control via rewards. They would not be able to gain the most rewards unless they spread those devices all over the world, and it would likely be prohibitively expensive to maintain devices far outside of its borders.
*A note about the type of benefits created: Benefits could be Proof of Work mining and/or Proof of Stake staking and/or other methods, but the method(s) should not inherently increase centralization. Global warming and other environmental destructions hurt disproportionately certain groups, particularly the poor who become poorer, and thus have less funds to participate in this or any crypto. This indirectly causes centralization. Thus, at the very least, reward methods should be ecologically friendly or neutral, e.g. transactions should not be more energy-intensive than your average credit card transaction. Likewise, a system that reduces global warming would actually help decentralization. For example, a device must first verify it directly sequesters carbon and in an ecologically friendly way, then it must sequester its operating costs in carbon, after which any extra carbon captured becomes benefit. Location spreading and verification would increase effectiveness. Further, there should be less chance of abuse and "cobra effect," such as when companies in a previous sequestering program made a pollutant just so it could sequester its byproduct. Location spreading would prevent connecting all the devices to a single pollution maker. And connecting to new pollution makers built just for this purpose, before of after verification, can be deterred by verification rules for ecologically friendliness. Such rules would be similar to anti-spoofing rules for location and uniqueness, as one would be swapping one form of centralizing pollution for another, and spoofing would likewise lead to forfeiture. Regardless, what is ultimately important is the overall effect of any benefits method(s) should not inherently increase centralization.
