The swarm roadmap suggests that the orange paper series and a landing page is released on swarm.
There is also an internet gateway for swarm: http://swarm-gateways.net/bzz:/theswarm.eth/
But how to access content hosted on swarm using the bzz:// protocol natively?
Just to point out, you don't need the public gateway. When you have a swarm node running, you have your own gateway at http://localhost:8500/bzz:/.
(I'd post this as a comment but stackexchange won't let me. Apologies.)
UPDATE: swarm-guide at RTD http://swarm-guide.readthedocs.io/en/latest/usage.html (Work in progress, expect decent status by mid december 2017)
but native to which level?
so there three levels that support the bzz family of uri schemes
these fallback on the http proxy API or a roundtripper wrapping the proxy call.
this API in turn falls back to the "native" API that addresses content via manifest paths and hashes, and range queries or indexes. (All these levels are neatly generalised by IPFS's Merkle DAGs and the IPLD representation).
The manifest path offer a file system directory tree mapping,and all these are provided through the usual JSON RPC (via IPC, WS and HTTP) does this answer your question?
bzz protocol, all I have to do is to wait for an open browser release?