I think this is the problem for open source software. It's not a one size fit all. We all do complex software and you all know its complexity. Sharding, failov

I think this is the problem for open source software. It's not a one size fit all. We all do complex software and you all know its complexity. Sharding, failover, retry, back off, schema etc...everything in a high scale systems. The problem is user who used it but doesn't need other features feels frustrated. At the same time, user who used it at scale need feature others don't like? What can we do? I don't know. A good example is Sentry: https://blog.sentry.io/2019/05/14/sentry-9-1-and-upcoming-changes https://blog.sentry.io/2019/05/14/sentry-9-1-and-upcoming-ch... Say you are a small single person SaaS? are you going to install ZooKeeper, Kafka, Snuba...No. But they need it...