The decentralization problem

A few days ago, I presented to my class about the thing everyone just loves to hear about: NFTs. Specifically how things can go wrong with them. I gave a bit of background on blockchain technology and then went into what an NFT really is. When I was writing my presentation the first time around, I practiced it once and realized I had left out a crucial portion. What is the point of this technology?

I stuck a slide right after my overview that went into the point. I stated that the purpose of blockchains like those of Bitcoin and Ethereum is mainly to get rid of intermediaries, and that NFTs are for having immutable proof of ownership of something. Today, there are many different purposes, but I went mostly into the problem with an NFT being considered “immutable”.

The biggest inspiration for the talk was from Moxie Marlinspike’s blog post about his take on “Web3”. He criticized the current state of Ethereum with one major point and the consequences of it: people do not want to run their own servers.

People not wanting to run their own servers has been a problem for a lot of computing history, and is mostly solved by centralization. The cryptocurrency space is no different. Almost every major player in Web3 has some major forms of centralization because of the issues caused by serving data to the client.

Yesterday, this became more publicly noticed when the most popular ways that people interact with Ethereum (accidentally) shut out entire countries due to sanctions. Infura, which is the entity that most of MetaMask goes through by default, accidentally misconfigured settings while complying with US sanctions.

A decent chunk of Ethereum users did not know what to do in this situation, or how it was even possible for this to happen. This was evidence of what I am calling the decentralization problem: decentralized things tend towards centralization for convenience. People care about UX more than what’s going on behind the scenes!

Power users told others that everything is fine and to just “run your own node”. Well, as Marlinspike said in his post, ’even nerds do not want to run their own servers at this point’. It is hard to envision that average NFT or cryptocurrency users will want to run their own nodes any time soon.

This was the point of the presentation. If even a lot of the nerds aren’t running nodes, there are many less servers than clients. This brings us back to centralization. When all of the apps are using the same major servers for convenience and UX, can’t they control how people interact with the blockchain?

I showed Marlinspike’s custom NFT that displayed different content depending on location. I was most interested in what happened after OpenSea took his NFT down. It practically disappeared from everywhere!

Of course, an NFT taken down by OpenSea is still on the blockchain, but what good is the blockchain if it is only interacted with from a few specific locations? I don’t think that proof of ownership would matter to many people if they are unable to use it in the place that is most convenient for others to see. Nobody really cares about your Bored Ape ownership if you can’t easily show it off in all of your profiles.

Good news for “Web3” users

Things are currently being done to solve these issues, from competitors like LooksRare gaining popularity and tools for accessing data quicker on blockchains. Big issues have already happened before there are hundreds of millions of participants facing consequences.

• If there are hundreds of competitors with no single excessive power share, it wouldn’t be as devastating if your NFT is taken off of a platform.
• Users are becoming more aware that these systems have points of failure when they aren’t run fully peer to peer and rely on major servers.