It's been one week since I launched my newsletter, and I have had ten different ideas for a post and none written as of today. I had committed to writing one 'perfect' post on the evolution of Mirror.xyz and why someone might 'collect' articles open on the web as NFT's and in some cases might pay for them as well, but let's save that for another day. As someone (Voltaire/Shakespeare/Confucius??) advised, Im going to settle for a 'good' enough post for today.
Todays post is inspired by Dror Poleg's entry in his daily newsletter(whew! daily newsletter) on the future of work. He talks about Kim and Kylie (bonus points for guessing their last names before you click on the link) and why they are worried about a tweak in the algorithm that decides what content their followers see. This seems to have been the result of recent efforts by Meta to respond to TikTok's popularity. It seems that Insta and FB are moving towards more algorithmic recommendations on what you might like, rather than emphasizing the content from people you follow, as that will likely improve their ad revenue. It seems like a race to the bottom among platforms to create the perfect algorithm to give you more frequent dopamine hits so you keep scrolling/clicking/watching. I do hope LinkedIn does not change its algorithm now, given that I just launched the newsletter there as well and all my followers automatically got pinged to subscribe to it :).
All may not be lost. There seem to be experiments where you may be able to own your social graph: your connections to people, groups and even your interactions with content (thumbs up, likes and more). One such experiment is the Lens Protocol, a play on the name of the Lens Culinaris plant category or Lentils as many of might know them from the kitchen pantry. You get a single identifier/handle(vishalsachdev.lens for me) which can be used across any app that builds on the social graph. Im still a beginner on the platform, so no elaborate social graph for me, but you can check out the graph for wagmi.lens (snapshot below, but you need to click on 'social graph' button on the link above to see a visual).
The data behind this graph is stored on the blockchain, which might get you thinking that it is slow and expensive and certainly not scaleable to handle millions of people and billions of interactions. That might be true, but I am not sure we need every network to scale to the level of twitter/fb/insta and others. There might be other value creation/capture models compared to the existing model of a centralized social graph monetized by advertising, where you/your data is the product but you don't have any control on your data or the algorithms that decide your experience/consumption on the platform. I don't have a coherent alternative to recommend here, and this might be another post(so many ideas!!!).
You might also think that this social graph is open and publicly visible and that might be a concern. I don't have an immediate answer to this concern, but I am guessing there is a solution using Zero Knowledge proofs.
Now for the benefits. Anyone can build apps on top of this data and you can see nine apps(as of July 2022) listed on the Lens website, though many more might exist. Think about multiple apps presenting information from the twitter social graph, in different ways, and if there is a market for the algorithms that you can use, you could choose the algorithm you want. I agree that this level of user due diligence seems unlikely, but perhaps algorithms with with some cryptographically secure attestations of trust/reliability/neutrality from enough members in the network might be a solution. Further, if these protocols are open source by default, as the lens protocol is, one can look under the hood as well.
There are other approaches such as the Farcaster network(in beta currently), which is "sufficiently decentralized", your identity is registered on chain but your activity (posts aka casts, follows, replies, likes, re-casts ) is stored off chain on a distributed network to avoid centralization risk. As a result, Farcaster seems to be reasonably fast, and I am guessing faster than recording all activity on-chain as is the model with the Lens Protocol. App development on top of Farcaster is permissionless, with some interesting apps such as Searchcaster, which allows you to search through all the posts on the platform. Another interesting use case is combining Farcaster with another protocol, BEB. Leverage your identity verified (by connecting to your Ethereum wallet, you can still be pseudonymous) in Farcaster, and build/participate in special interest communities (bebcaster.beb.xyz, https://winecaster.beb.xyz/ and more). Think Facebook groups, but built by a third party on top of Farcaster protocol. Many more at the Farcaster Notion page. Need to look under the hood at how the protocol works; here you go - https://github.com/farcasterxyz/protocol.
Perhaps there will be a continuum from centralized platforms such as FB/Twitter/Insta/ LinkedIn where you are the product, and don't control your data, to protocols such as the Lens Protocol where everything is on chain. This discussion on owning your social graph and data, is enriched by learning about Decentralized Identity(#DID) and the umbrella term of Decentralized Society(#DeSoc). If you want to dive deeper into #DeSoc, Vitalik Buterin and collaborators just released a thirty page concept on the Decentralized Society: Finding Web3's soul, or if you are short on time, read this explainer.
How's that for a first effort at writing in a newsletter? I know I probably missed a lot in the analysis, so do keep the comments coming. Until next time....