# Scattered thoughts on privacy

Blockchains in general unlocks new use-cases, but most of the usage comes because they are convenient, not because they are necessary.
If you are building an application, with VM blockchains you don't have to worry about setting up a server backend, about downtime, databases and you have a unified language with tons of tooling, templates and standards. Users have a better UX (with still a lot of room for improvements) with Sign-In with Ethereum and similar wallet integrations, interoperability, transparency. A lot of this can be done with web2, but with web3 it's just easier.

The amount of privacy that most of the users need is already provided by web2 apps, in the sense that friends and family can't see your bank transactions and your instagram DMs. Centralized exchanges can already be used for hidden-from-friends trading or to fund a new wallet. Actual cryptographically private transaction make sense in two scenarios:
1. you want to hide your activity from centralized actors like banks and governments
2. you want to use unique app on blockchains with at least hidden-from-friends privacy

Note that point 2 can also be achieved with web2 but with a lot of trust involved and regulatory issues, while decentralized solutions are already being released. This is also the point that I think will get more traction, and that's why I think it doesn't make too much sense to launch a new private L1. To expand on this, it's important to take advantage of the activity and assets of the most used blockchains with trust minimized bridges -- also privacy is something related to execution and settlement and not to consensus so it doesn't make much sense to launch a new blockchain that needs a new validator set, that's why I think building a smart contract rollup it's the best option. Privacy to be widely used has to be convenient, not usable just when it's necessary: most people don't care about privacy when using blockchains because it's too hard to use and it's most of the time not worth the effort and a rollup is more convenient than a new L1.

Application and transaction transparency is a feature that really distinguish blockchains from web2 apps and a fully private blockchain can't take advantage of it. Supporting both transparent and private transactions can be challenging, but a blockchain can provide transparent transactions at the base layer (or on a L2) and private ones on the layer above (L2 or L3).

I think that a good framework is thinking that the best scenario is very similar to Ethereum + a switch that can make any tx private and any deviation must be extensively justified (e.g. different VM because zk + EVM is difficult, different L1 because that feature it's crucial etc)

Other things to think about: liquidity fragmentation, L2->L1 private txs (batching) and delays, deposit amounts, ...