It would be a damn good way of identifying the individuals interests, no doubt about that

There appear to be two main premises at play here:

1.That as an RSS Feeds consumer I want articles fed to me to be influenced by the popularity of those posts with other like-minded people.

2.That I want to be connected to these like-minded people in some sort of social network.

Whilst I totally accept premise 1, I am not so sure that premise 2 holds true – at least not for me. I am not averse to social networks and I belong to a couple, but I do think that there is a tendency these days to create them just for the sake of doing so.

So the question is do I need to be connected to a person in a social network in order to have my feeds influenced by them? I don’t think so. myFeedz from Adobe Labs is an interesting rss reader that I enjoy using – now that they seem to have resolved the performance issues I initially experienced on signing up. Basically it lets you import an OPML file and automatically creates tags from your imported feeds. You can add more tags and feeds as you please. You can then filter articles by any given tag – articles taken not just from your feeds but also from feeds uploaded by other myFeedz users. All this without being ’socially’ connected to anyone.

The one feature that is missing as far as I have noticed is the ability to vote a post up or down like you can with digg. If my Feedz were to add this and assuming people would actually bother to rate posts, then I think they could be on to a winner. I could then choose to sort posts within a tag by popularity. This to me would be more valuable than being influenced by what a particular person or small set of people may or may not have found interesting. Different strokes I guess.

To expand on your “parsing current readers” idea – this was something I thought of but decided not to add just for clarity purposes.

The process of having user’s export their OPML file and then upload it to the application servers is a daunting one for many people. You really have to offer a compelling service for them to go through all of this work – or make it easy for them.

Two routes immediately come to mind: 1) a Firefox extension that will parse a user’s feeds each time they visit their feed reader and report this to the application server; 2) a backend on the application that allows the user to enter their login information for their feed reader and the application then goes out and does the work.

Number 1 requires more user interaction, but has many benefits. The user’s OPML file will always be current, everytime they visit their feed reader, the application retrieves any changes. It also provides a bit more trust to the user, since they don’t have to give us any login info. A $20 thrown at India could get this extension completed in a day or two.

Number 2 is more easily developed in-house, but lacks the “secure feeling” users expect. You can tell users all day long that you don’t store their data, but they are still typing in their username/password (scary). You also won’t get the consistent updates from their OPML file, since you won’t be storing their login info and hitting their feed reader every day or so for updates. Finally, you are adding a lot of backend processing to the application that you can pass off to the user’s with a simple extension.

Just a few ideas I’ve had in the 20 minutes between sending it to you and now. Thanks again Steve!