this post was submitted on 30 Oct 2025
103 points (99.0% liked)
Privacy
2957 readers
28 users here now
Icon base by Lorc under CC BY 3.0 with modifications to add a gradient
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this seems needlessly combative… prevailing opinions are exactly as signal says… think differently? great! let’s do it, talk about it, see how it goes, and when the solution has scaled in the real world to what it’s competing against then you can feel superior as the one that had the vision to see it
but scaling is hard, and distributed tech is hugely inefficient
there are so many unknowns
anyone can follow a random “getting stared with web framework X” guide to make a twitter clone… making a twitter clone that handles the throughput that twitter does, that takes legitimately hard computer science (fuck twitter, but it remains both a good and common example)
heck even lemmy has huge issues with sync at its current tiny scale when there’s any reasonable latency involved… i remember only months ago when aussie.zone was getting updates days late because of a 300ms latency to EU/US and lemmys sequential handling of outboxes (afaik)
How is it inefficient for a chat app? If anything, a distributed architecture is the ideal for this use case. It's only potentially a problem if you need to have huge group chats, which is definitely not the common use case for a chat app, but even then I think Delta Chat's optimized relays can handle that.
Delta chat uses existing email infrastructure, which has already proven its ability to scale. Nigerian princes probably send more emails per hour than the entire global Signal network.
I'm guessing inefficient in a sense that with distributed you need more computational power in total than with centralised
Idk what to say to this. Is it true? I don't know, and you probably don't either. That's a weird way to look at it and I doubt anyone has measured the power costs of the global email network.
It's also useless for decision-making. What matters is a question like "how much would it cost me to host a server and contribute to the network?". Even if the total global cost is billions of dollars, the network will continue to grow because nobody has to pay all of it.