For every post in an RSS feed, most aggregators use a field called "guid", which is unique for each item. If you've read that item, they consider it read and don't mark it unread unless you specifically say so. Bloglines, on the other hand, appears to take the approach that if ANYTHING in the item has changed, it must be read again. So, the earlier footer message I had, pointing to my other sites caused everything to pop up unread whenever the message changed.
Well, yesterday I put in another change that is going to make that happen again and I apologize. However, it's there for a good reason. I keep seeing people claiming unattributed theft of content as their reason for only publishing partial feeds. You know the ones where they only give you a paragraph in the feed and require you to visit the site to get the rest of the article?
They see it as a choice between giving the full version, which is what most readers want and the excerpt because of the few places republishing their content elsewhere. To me, this just seems to be a scorched earth approach where something more subtle could work.
What I've added to my feeds is a single line at the bottom that shows what IP address made the request. I may have to go to something more unique if aggregators are using multiple IP's to request content. If you're seeing that via Bloglines or another service, please let me know. My intent over the next few weeks is to build a whitelist of the major services like Bloglines so that they don't see these messages at all.
For the rest of the web, I can then track where the request for the content that's being used inappropriately came from. The feed script will then check the requesting IP against both a whitelist and a blacklist. The whitelist will see no message at all, the blacklist will see a message warning readers that the feed is being used inappropriately and everyone else will just see the little note about requesting IP.
Then, all I have to do (or more accurately, people using the resulting code) is add the offending IP addresses to a file or database table and specify what those feeds should contain and the offending site goes away.
This technique has been used for those who steal images for a long time. Simply replace the real image with "This image was stolen from ABC site" and the problem is taken care of.
Incidentally, the IP address that Bloglines requested for my viewing is: 64.34.162.9. Does that match other Bloglines users' entry?
