the primal caching server will be a huge bottleneck, afaiu all reads have to go through their relay, so you don’t get the benefit of load balancing across many different relays if one become unreliable
Has anyone figured out if/how/why nostr:npub12vkcxr0luzwp8e673v29eqjhrr7p9vqq8asav85swaepclllj09sylpugg caching collides with VPNs? Hours and sometimes days of slow or no loading/updates, while nostr:npub18m76awca3y37hkvuneavuw6pjj4525fw90necxmadrvjg0sdy6qsngq955 works uninterrupted.
Discussion
In my experience, it is *not* working out.
They can probably get there, twitter is proof of that. It will just take lots of redundant servers and scaling up their server architecture. It’s a completely different model than what damus is going for, we’ll see which one works out the best. It’s nice you can switch between them if there are issues on either side. nostr rocks !
The centralisation of a decentralised protocol now live on Primalbase