>On Sun, May 22, 2011 at 5:10 PM, Dhruv Matani <dhruvbird at gmail.com> wrote: >> Exactly my point! So, keeping the 2 (request & response IDs) together >> doesn't really help simplify the code (since out-of-order >> re-construction needs to be performed in either case) and "keeping a >> queue of sent requests, and reading responses from sockets in the same >> order you made requests" won't quite be the right thing to do. There >> is still post-processing to be done in either situation. > > It doesn't simplify the code *for async clients*, like XHR. (It > doesn't complicate them, either.) Keeping a queue of requests is how > synchronous clients work, and it's a huge simplification. See jbosh, > for example.
(This discussion is purely under the assumption that ACKs are in use - my apologies if I didn't make this clear initially. I also realize that most clients (and servers) these days don't support BOSH level ACKs, so this discussion is slightly non pragmatic as well). With ACKs enabled would it still simplify things for synchronous clients? Regards, -Dhruv. -- -Dhruv Matani. http://dhruvbird.com/ "What's the simplest thing that could possibly work?" -- Ward Cunningham
