“Farcaster must remove all the bots” 

This is a worrying line of thinking I’m seeing more often.

The argument goes that it’s not enough to hide uninteresting content. It *must* be removed from the protocol. 

That is, someone must judge the worthiness of accounts and delete them from the protocol.

Technowatermelon. Elder Millenial. Building Farcaster. 

nf.td/varun

Any related thoughts about hub sharding or syncing data subsets?

Agreed that there should never be an explicit "ban" at the protocol level, but also hard to scale for smaller hub infra without any ability to prefer content of interest

thoughts on scaling: https://github.com/farcasterxyz/protocol/discussions/163

tl;dr no plans to shard hubs until much larger scale (several TB+)

ML, robotics, DeFi | building /propertyforecast