On the subject of decentralization and the Web 3.0

I lately gave a quick talk on the subject of the Metaverse Hype that is currently running over all of us. 

The bottom line was that we at Artificial Rome do believe that this is a serious development and that we enjoy being part of the community that shapes this new environment.  However, we do see a lot of indicators for a hype that could become a bubble, which will once it bursts, delay this entire promising quest for several years, yet again. 

You can see the full talk here:  

In the aftermath, several discussions were sparked, which I want to engage with in more detail. The cornerstone of the dispute was the subject of decentralization. Which I broke down to merely the crypto backend, the storage of data on the blockchain, which is supposedly decentralized. I do not think that this is restricted to the backend, it becomes much more obvious when we look at the front end. But this will need a separate article.

The key figures in this argument are three Industry Leaders.

 Moxie Marlinspike is the founder of signal 

he started the discussion with this brilliant Article

Vitalik founder of Ethereum reacted here 

Dan Finlay the founder of metamask reacted here 

Thanks to @DavidLinderman and @KatschePlatz for bringing this to my attention.

But let me fill in the blanks here and review the full discussion around the rather central decentralization of the blockchain, which is supposed to be the core of the dawn of this super new Web 3.0.  It started with an article by @moxie the founder of signal. To grab the subject he actually wrote two dapps ( decentralized applications ) and describes his experiences rather brilliantly. He has two main arguments: One is that in the past open-source decentralized projects have been slow in their development process to keep up with centralized approaches.

The second argument is, that right now the Ethereum blockchain is too heavy and too difficult to interact with, so practically we rely on a centralized server to interact with it, which is the exact opposite of the entire idea. Most blockchains are very large files, that you can download and work with directly.

Right now this is what you need to do. The 7 step user experience starts with installing a command-line tool and from then it goes on. Try explaining that to your mother. Or father
Apart from the practical feasibility for a wider population, the file size means that neither a web browser nor a mobile phone can interact with the blockchain directly. So they will make calls to a server, which then interacts with the blockchain for you. This seems like a minor technical detail, that non-programmers might like to ignore, but this very matter turns the whole idea of a decentralized blockchain backend for the new Web upside down. 

When you have to call a specific single entity ( I think it's okay to call this a server ) This does mean that there will be (a) a security risk in case the server gets hacked. 

(b) The single entity controlling the server can abuse the centralized power given to them.

Regarding (a) the Security risk: 

If I'm in control of this one entity infuria I could override every NFT that is calling the server and write my name as the owner of the NFT then right into the blockchain. So thank you beeple happy to have you. 

This is what Vitalik the co-founder of Ethereum is mainly referring to in this scrolling intensive article. He states that there are a lot of measurements that you can take in order to make the calling of infuria, the central server, more secure. 

But as a true cryptograph, he is mainly referring to the security issues involved.

Here actually underlines that he does not think that any common  user will ever interact directly with the blockchain. So there it is, not even the founder of Etherium sees this Web3.0 future that your millennial-gaming-kid-marketing-advisor is babbling about.  

Regarding (b) the abuse of power.

This is currently the higher risk with the bigger impact on society. Whenever a technology runs on a server that is controlled by a single entity, this entity can abuse its power and manipulate the technology. 

This is why the idea of a decentralized Web3.0 is so enchanting. 

But the core question here is how much work or machinery does it eventually need to run parts of the blockchain locally for every user. The idea here is, that every user will become a server and thru this effort, it becomes a truly decentralized network without a single entity controlling it. 

Currently, the industry leaders do agree on the absence of decentralization. But there are different perspectives on whether or not it is possible in the future. 

Yet here we are already, hosting events in Decentraland throwing our Mana everywhere as if we have arrived in some kind of great future already.  This is what scares me about the hype. The total disregard of those technical details by the public. I know this does need more than two sentences two explain.

The problem is that those details matter in tech, a lot. They are the little backdoor the NSA uses or the microphone accidentally placed in googles NEST device in your home or the friendly people transcribing your conversations via Alexa. These details do have a huge impact on a democratic society. 

So the at first glance boring technical question on how much data do you have to download, how many programs do you have to install to run the Ethereum blockchain locally on your computer, this tiny question can be the core difference between us becoming a truly decentralized society in the Web 3.0 or whether we will continue in the centralized manner that we know.

I was especially baffled by moxies mentioning of metamask, which also does make calls to infuria. Metamask is a service that makes it easier for non-programmers to interact with dapps and the Ethereum blockchain. It positions itself as a gateway to the new web3.0. 

And yes I do feel majorly hypocritic right now since it is also used by us to run this very blog.

But let's dive deeper here: David Finlay Founder of Metamask and now Group Manager there wrote this lovely reply to moxies article. He states in there that Metamask has pursued ( past tense ) light client approaches with Muskatela.

A light client means you can be part of the blockchain without fully downloading it. So less hassle = more adoption. 

The thing is the Muskatla idea is from 2008 ( that is what pretty much the medieval times of the internet ). And it did not make it into the current build of Metamask, mentioning it seems a bit desperate here. 

This is pretty much like Porsche saying, "Hey there, we did build the first electric car in 1898, and then nothing after it until tesla came along." The difference is that Porsche is shipping their promise today, while metamask is not. 

Metamask is shipping these days with infuria, as David Finlay states himself. Which is a centralized server. Yes of course you can change Metamask to interface with your local nerd wallet, but realistically how high is the percentage of users doing that?

Interestingly this whole discussion and even the term Web3.0 blends chaotically and seamlessly with the idea of the Metaverse these days and seems to come to life with projects like decentraland and sanbox. But I do get the strong perspective and can't find the contradictory argument even when the founders and tech giants of the cryptography community are battling, that the decentralization of this new Web is merely a marketing gag.

So leaving everyone to their own opinion and we are happy to agree to disagree here: We at Artificial Rome are focusing more on the front-end perspective, in our opinion the actual revolution that is happening right now.

It's a brave new world surrounding game engines, Unreal 5, what Nvidia is doing, and even current systems like Roblox, core, Minecraft, and any world that can be handcrafted by its population. 

So yes let's go, but with an open eye ;-) 

Patrik de Jong

Artistic Director & Founder Artificial Rome GmbH