Re: [streams] First draft at tee algorithms, for critique (#302)

>  This should preserve backpressure and also has the "OOM" property discussed in #271 such that if the consumer for branch1 is slow and the consumer for branch2 is fast, then branch1 ends up with a lot of unconsumed chunks in its queue.

Wait, upon re-reading #271 I realized this is contradictory. What I meant to say is that if *both* `branch1` and `branch2` are not read from, then backpressure is applied to `stream`. Whereas, if `branch1` is consumed slowly and and `branch2` is consumed quickly, no backpressure is applied.

---
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/streams/pull/302#issuecomment-82939995