Our current main point against distributed TURN servers is performance. Tom suggested a procedure where we find the nearest TURN servers to both peers (or between them, etc.)
After figuring out #37, amend the section in the SA document with our findings.
Our current main point against distributed TURN servers is performance. Tom suggested a procedure where we find the nearest TURN servers to both peers (or between them, etc.)
After figuring out #37, amend the section in the SA document with our findings.