Damus
Gzuuus · 4d
We've talked about this different times. If you expect to have an undefined number of requests during a session, it's more efficient and performant to use WS. Yes, you could batch and create a single ...
Pip the WoT guy profile picture
You raise some good points.

I would say in general the tradeoff is simplicity vs flexibility.
WS allow to make many small requests, where HTTP does it with batching. WS pays a bigger cost at the beginning, which may get compensated, depending on the usage. WS are more complex to deal with, almost all client devs I met complained about WS being unreliable, having issues with reconnection etc. Maybe poor libraries? idk

I don't get the point about getting rid of IPs, DNS etc. Relays are HTTP servers that hijack the connection to upgrade to WS, so they have to comply with all that. If you are hinting about the fact that a CVM provider could use someone else's relay to serve its customers, I would agree only partially. The latency alone would likely render this business not very efficient at serving its customers. It's a nice to have but I don't think it will be relevant in practice IMO.

> I wish it could be transport agnostic so I can implement it in Relatr.

Let's talk about it mate. We may design something even better!
1
Gzuuus · 3d
Trading resilience for performance, is a good tradeof depending on the case. I will check the spec so maybe we can come with a more interoperable standard that doesn't depend on an specific transport