I solved the "Original site" problem by building the original site into NewsBlur -- http://www.newsblur.com.
The other big issue with RSS is that there are too many stories with a low signal-to-noise ratio. I built in filtering and highlighting into NewsBlur to address that concern. And it's a completely separate backend from Google Reader.
And now the common refrain is that people use social channels (Twitter/FB/Tumblr) to find links and news. So I just built that into NewsBlur with shared stories. You can sign up to be a part of the private beta at http://dev.newsblur.com. I'll send out invites to anybody who signs up.
Consuming the web through RSS can be problematic for both publishers and readers. I'm addressing the big three issues - context, relevancy, and surfacing - with a strong commitment to both readers and publishers. Let me know what else you would expect to see in your ideal reading setup, and chances are, RSS offers the foundation to build it.