Louis Gray wrote earlier today about Noise River, a new service that taps into the FriendFeed API to deliver intuitive filtered content, along with some other extras.
To use, FriendFeed users sign in with their FriendFeed user name and FF key. NoiseRiver then asks uses to rate their likes or dislikes based around keywords and user names. The result: filtered content, graded based on your on specifications and likes.
Also thrown in for good measure, but very handy, is that ability to list FriendFeed entries you have commented on or liked. There have been plenty of times I've wanted to check a thread I've commented on the night before to see if it had further comments over night, with NoiseRiver this is all built in.
NoiseRiver isn't a FriendFeed replacement, and nor at they pretending otherwise (you get a bright notice telling you), but what it delivers is very handy indeed, particularly as the level of noise of FriendFeed grows as the Twitter refugees start rolling it. There's only one way this service would be better: if these features were built directly into FriendFeed itself. I'm that impressed with the ideas and functionality I'm going to make a prediction: that FriendFeed will acquire NoiseRiver, or even Feedego, the company behind it. Of course, they could just build these features themselves, but the one thing I continue to be impressed with is what great, humble, forgiving (well in my case ;-) ) fair, community minded people the FriendFeed team are. I just can't see them cloning this, but I can see them acquiring it.