My impression of 'My first impressions of web3'

I read Moxie’s essay about web3 and while I agree with many of the points he mentioned about the current state. I think he missed some stuff and wanted to go through them.

The main point the essay was focusing on is that people don’t want to run their own servers and that the current state of the web3 platforms is very dependent on things like Infura or Opensea. While that’s totally true, the point he missed is that there’s a very big difference between, you don’t want to run your own server and not being able to run your server even if you want to, which what I believe is web 3 is about.

A few years ago the U.S added new restrictions that lead to blocking sites like Coursera, Udacity, Github, and many others for many countries. People who bought courses weren’t able to access the content they bought or the ones who finished the course weren’t able to access the certificate. But if the content was on IPFS and the certificate was an NFT that situation would’ve been totally different. Not only in education. Github was blocked in many of these countries and people sites were down. Many people I talked to didn’t move to Gitlab but decided to host their own git service. In both cases, if these platforms were built on top of web3 infrastructure, this would’ve been a totally different situation.

I think Moxie’s essay was focused so much on the fact that people don’t want to run their own servers forgetting that businesses might want to do so in many cases in many places in the world.

I understand that the restrictions topic is a controversial one but even in other cases. Recent example:

NPM has reverted to a previous version of the faker.js package and Github has suspended my access to all public and private projects. I have 100s of projects. #AaronSwartz

View image on Twitter

I am not saying what’s right and wrong but I am saying that this is so much control in the hand of these companies.

Balaji has mentioned this Paul Graham essay in 2009 about how Twitter was like a protocol but when Twitter decided to change their APIs policy many of these apps went down. But if Twitter was built on top of a web3 infrastructure, I bet people would run their own servers or just access it through another client. And I think if it was on a web3 infrastructure, Twitter wouldn't even do that because they know there’re alternatives. But what allowed Twitter to do this is that they knew people didn’t have any other option.

Another point was highlighted on how things are centralized around Opensea. The point he forgot to mention is that you have the option to host your nft store by yourself even if you're now on Opesea but you can’t do that for your Facebook account. Even Opensea was the earliest to the space, we’ve seen many platforms emerge like Foundations or Raible as the barrier to entry is much lower. Or even building things on top of the graph which was created on Opensea itself like in Showtime.

Same thing with exchanges. FTX started many years after Coinbase. Rainbow and WalletConnect started after Metamask. In web3 you could very easily switch your exchange, wallet, or nft store but in web2, you can’t.

About protocols, I think Varun has an interesting point about why products are much faster and protocols are much slower in web2.

email hasn't just been slow, it's basically static.

the real reason is not that it's hard, but that it is far more profitable to improve email clients than it is to improve email itself.

the right answer might be to find ways to reward people for making useful protocol changes

While Email has been static. in Ethereum, EIP-1559 from the proposal to shipping took around 2 years which you can consider a very important change.

I also think that the essay was so much focused on art NFTs forgetting about things like AMMs, stable coins, and ENS.

For sure there’re many things to do to fix the issues mentioned in the essay. Like pushing the state of zero knowledge proofs, projects like Ghost.org for NFT marketplaces making it so easy for people to host their own stores, build more protocols abstractions and opensource alternatives for Infura that are easy to run is highly important.