I find it interesting how often developers tend to reinvent because of looking at a problem from only one perspective. Today I read a blog post by Sean Gephardt called RSS and syndication Ideas? where he repeats two common misconception about RSS and syndication technologies. He wrote
What if I only want certain folks to has access to my RSS?
I could require the end user to signin to my site, then provide them access to my RSS feeds, but then they would be required to sign in everytime they tried to update thier view.
More specifically, how could a company track people that have subscribed to a particular RSS feed once they are viewing it in an aggregator? Obviously, if someone actually views the page referenced, then web site tracking applies, but some aggregators I've seen simply render the contents of the description, which if it contains a URL to somewhere, and the user clicks that link, the reader gets taken over to that URL, bypassing the orignal.
Since there is no security around RSS and aggregrators, and no way to prompt users for say, a Passport authentication, should RSS be used only for "public" information? Do you make people sign in once they try to access the “deeper” content? Do you keep the RSS content limited to help drive people to the “real“ content?
Am I missing something glaringly obvious?
Considering that fetching an RSS feed is simply fetching an XML document over the Web using HTTP and there are existing technologies for authenticating and encrypting HTTP requests, I'd have to say "Yes, you have missed something glaringly obvious Sean". In fact, not only can you authenticate and encrypt RSS feeds with the same authentication means used by the rest of the World Wide Web, aggregators like RSS Bandit already support this functionality. In fact, here is a list of aggregators that support private RSS feeds.
As for how to how to track readership of content in RSS feeds. A number of tools already support tracking such statistics using web bugs such as dasBlog and .TEXT. One could also utilize alternate approaches if the feeds are private feeds since one could assign a separate URL to each user.
All of this is stuff that already works on today's World Wide Web when interacting with HTML and HTTP. It is interesting that some people think that once you swap out HTML with XML, entire new approaches must be built from the ground up.