Lawrence Forooghian

Ah, OK. There is also a C version, though. [Apple’s guidance](https://developer.apple.com/documentation/technotes/tn3151-choosing-the-right-networking-api#WebSocket) is to favour the Network framework: > Unless you have a specific reason to use URLSession, use Network framework...

If we drop support for iOS < 14, as is being proposed, we can then start using Swift inside the codebase anyway

Oops — GitHub got confused by my commit message for https://github.com/ably/ably-cocoa/commit/706f269d204645e6cd64d7ce8594e6032c8efc94 and thought that it fixed this issue. It doesn’t, reopening.

I think the idea of being able to specify a separate token to be used for a specific push type is a sensible one. > It will help to eliminate...

I think this might have been accidentally closed as part of Unito sync issues. Reopening.

Oh, I was definitely not aware of that! Is there a place I could find out more about this? `man 2 write` wasn't hugely illuminating.

Yeah, I understand what Paddy meant now. We can close this one.

Whilst working on #396, it's become clear that EventMachine’s TLS handling leaves a lot to be desired. Another reason to consider moving away from it.

Hey @mattheworiordan, just taking a look at this one whilst prioritising bugs. I see that we don't explicitly advertise support for Windows in this SDK. But nor do we explicitly...

I might well be misunderstanding something, but running locally on macOS it appears that the client is sending SNI, and yet I get the same test failure: Will investigate further.