#6: The Technology for a New Internet

This is the longest essay in this series since it gets into the specific technical innovation that is creating a new internet. But to recap so far, the design of the new internet we are trying to achieve is to create a blockchain that serves as a global content database that is common to all social media apps.

This global content database can simply be understood as a database of “Who Said What”. It has a global user identity and wallet with which anyone can put content on this database by paying a small fee.

Now there are two competing approaches trying to achieve this new internet.

Modular Approach

This approach tries to take existing financial blockchains which do have an identity/ wallet and combine them with other databases on which you can store content.

These databases are either private databases (like AWS S3 or dropbox) or data blockchains (like Filecoin or Arweave). This is practically the design that is used for all NFT applications where the image is hosted on Arweave and just the link to that image is stored on some financial blockchain like Ethereum.

The key thing to notice in this model is that the database of “Who” and the database of “What” remain separate but are thinly linked via pointers from one to another as shown below.

The good thing about this model is that it provides “modularity” meaning the developers can flexibly choose the “who” and the “what”. Examples of current modular databases are:

	Protocol	Who	What
1	Farcaster	Optimism (ETH L2)	Community servers called “Hubs”
2	Lens	Earlier Polygon, Now ZkSync (Both ETH L2)	Arweave
3	Drip	Solana	Arweave

But the downside of this approach is that it becomes very fragmented and you achieve multiple distributed content databases instead of a global content database.

An alternative approach tries to solve this problem:

Integrated Approach

In the integerated approach in which all the information would be directly on a single system including the names and the content itself without needing links two separate systems. So, in the current example instead of links to another storage system, the wallet, name and the content is directly on same blockchain.

Pros and Cons of Each Approach

Now, that we have understood the basic concept behind modular versus integrated blockchains, its time to understand the pros and cons of each.

For modular blockchains, the advantages are:

1) Building on existing networks - A big advantage of modular content blockchains is that you usually start with existing big blockchains like Ethereum or Solana for identity. This has the advantage that, you are not starting from scratch and you don’t have to face the cold start problem. There are existing communities of millions of Ethereum / Solana holders who hold billions of dollars in aggregated capital. You can strap existing components and launch your product much more quickly just like mirror.xyz did in just a few months.

2) Extreme flexibility - Modular blockchains allow any type of combination of existing components and thus allow extreme flexibility in designing solutions. Contrast this to integrated blockchain where you have to standardize a lot of things like format in which a blog can be written, which is limiting.

3) No single point of failure - Given modular blockchains exist in many forms and use a combinaiton of many different systems, it prevents the possibility of the whole content ecosystem being brought down by attack on one big system. For e.g. if Arweave does down, even then Ethereum blockchain will remain just fine.

For integrated blockchains, the advantages are:

1) Ease of development - Given all the data of “who said what” in the world exists on a single chain, it is very easy for developers and content curators to build apps on top of this. For example, if a football lover social media has to be curated, we can just query this single database to say give me all the content related to football. Contrast this with modular blockchain where you can’t query this on Ethereum directly. You would have to go through all the links one by one, scrape them and then create your own database of who said what and only then you could query it.

2) User experience - The user experience is also significantly better on a single integrated system because there is a single log-in across all apps, a single wallet address, lower latency because only one system is involved and lower transaction cost. Contrast this with Ethereum for example where just getting a user name can cost 100s of dollars.

3) Higher scalability - An integrated system which has been built from ground up for one use case like social will always be more scalable than a general purpose blockchain like Ethereum which tries to do many things at the same time. We see this already in Solana versus Ethereum context.

To summarize, the advantages versus disadvantages of integrated versus modular blockchains, it looks like this:

Blockchain Type	Advantages	Disadvantages
Modular	- Existing networks leverage	- Development complexity
- Extreme flexibility	- Higher transaction costs
- No single point of failure	- Fragmented user experience
Integrated	- Ease of development	- Start a network from scratch
- Unified user experience	- Single point of failure
- Higher scalability	- Limited flexibility

Players of each approach

Up until now, we’ve discussed the theoretical constructs of modular versus integrated blockchains. Now let’s discuss the specific projects that are pursuing this. Coinbase ventures had published a good mapping of different players in this space. It looks like follows.

It shows various services providing different parts of the stack similar to what we discussed in modular approach. It involves:

	Functionality	Blockchain service
1	Financial Blockchain	L1 like Ethereum/ Solana or an L2 Polygon/zkSync
2	Storage Blockchain	Arweave / IPFS/ Livepeer
3	Naming Service	Ethereum naming service (ENS)
4	Wallets	Metamask/ Phantom
5	Social Graph Protocols	Farcaster/ Lens
6	Social Tokens	Rally

However, this diagram is too cluttered and doesn’t make any distinction between integrated versus modular construct.

There is another flaw that it also puts the “DeSo” blockchain in many of the blocks because the DeSo blockchain is following an integrated approach meaning it has all components like hosting/ identity/ authentication/ wallets/ social primitives/ wallets/ feeds all built in a single blockchain.

A better diagram should look more like below. In the modular approach, there are many different components that must be stitched together to make Web3 social work. In integrated, there is one system that has everything from wallet, storage, social tokens, feed, etc. DeSo is the only player right now which is taking an integrated approach to web3 social to make everything on a single blockchain.

Given the DeSo blockchain is such an outlier in it’s approach, we will dedicate the next essay to understanding more about it.