I ask because development has been non-existent for most of this year, particularly since February, since when almost all commits have been related to their P2P/Pinecone stuff.

Obviously I think P2P Matrix is really cool but… I thought their aim was to make Dendrite a Synapse replacement? It’s nowhere near that right now and it doesn’t look like they’re working towards that direction any more.


I was wondering something similar so I asked in the dendrite matrix room, seems the devs are just short on time atm.



Thanks for the link! Yeah I assumed they were busy with the other stuff, I’m more curious why Matrix thinks that’s the better route to go down now rather than actually finishing Dendrite and then working on P2P etc, especially considering iirc they’ve said most the difficult parts of the spec are already implemented in Dendrite.


It’s not in their best business interest to have an easy to use and low resource homeserver, at least not in the short run. Right now nearly all larger organizations using Matrix (such as Mozilla) do so by renting these services from NewVector Ltd.


That’s hardly the issue or the business strategy. Chances are Mozilla would still not self manage the service if Dendrite was the homeserver


Maybe this is more promising: https://gitlab.com/famedly/conduit


I actually looked into Conduit and considered using it rather than Dendrite because development is way more active (and I also prefer Rust to Go).

For now though, going off of the README, it looks like it’s still behind Dendrite in terms of stability. Hopefully they don’t, but if Matrix continue to not touch Dendrite and Conduit overtakes it in terms of stability, I’ll probably switch over.

An open network for secure, decentralized communication

  • 0 users online
  • 14 users / day
  • 17 users / week
  • 35 users / month
  • 68 users / 6 months
  • 835 subscribers
  • 72 Posts
  • Modlog