One thing I like about FeedBurner is the little badge they can provide telling how many subscribers are getting their hands on your RSS feeds. Since I'm not planning to use this service, I've been thinking about coding something (for WordPress, of course) that could produce the same kind of statistics. Indeed, how many unknown and silent readers, who don't even come to your site, can you reach via RSS, that is an exciting question.
So, I went through some logging and digging in feed requests, to come out with a few stats, facts and questions…
- Some services like Technorati are fetching my 3 feeds in a row : RSS 0.91, RSS 2.0 and Atom. Isn't this a waste of their resources ? I thought they could only fetch one, say, the most verbose format. I really don't see the point of parsing less loquacious files.
- Pluck's bot (aka PluckFeedCrawler/2.0) is denied from accessing sites running Bad Behavior, because of the lack of 'Accept' HTTP header. I hope this will be fixed soon, since it looks like a rather nice service, almost as useable as Bloglines.
- People running a bot on fc-e1.feedcache.net are real dumbasses. Their bot sends absolutely no headers, not even a User Agent, and fetches my feed exactly 4 times in a row in 4 seconds, every fikin' hour. By the way, why isn't it blocked by Bad Behavior when sending no HTTP headers ?
- Someone running intraVnews on florent.netbios.fr is polling my RSS feed every 10 minutes. What a misuse of bandwitdh considering the frequency of my posts :)
- Obviously the RSS market is still growing, judging by the number of bots and services that crawl feeds when associated websites are not open yet : topicblogs.com, blogslive.com, zilbo.com, socialmarks.com, feedcache.net.
- Socialmarks/0.2 Beta comes from Socialmarks.com, a website which title is "Social Boomkaring". Wow, sounds cool. Let's boomkar, then.
- I don't understand why RSScache is regularly fetching my feed ? I'm not using their service.
- Some clients are unpolite jerks not clearly telling who they are and where they come from. How am I supposed to find about "Ken", "GM Panel" or, I liked this one, "SET USER AGENT" ?
- I've found about a few nice looking or promising clients,
and amongst them Lektora which I'll try when I have some time.(tried. It sucks)
- I've found about some websites I'm not going to trust. Although blogstreet.com and bulkfeeds.net are indexing my feed several times a day, I couldn't even find anything related to my site in their search engines. What a proof of efficiency.
- Some web-based readers have detailed user agent strings telling how many readers subscribed to your feed (Bloglines, IFeedYou, Rojo, NewsGator, Pluck, Feedster, Yahoo. This is smart, and I hope other similar services will do the same.
- I've tried several web-based readers, and no one is half as cool as Bloglines.
- … But I'm still waiting for an invite to try FeedLounge :Ãž
Now, as for the number of readers … If I take all requests, remove those from crawling bots and indexing services, and count the number of different pairs of IP + user agents, this give a number of … 99 different daily readers ??
Is there a flaw in my thinking, or do I really have one hundred RSS readers ? Sure, I'd be honored, but it seems way too much to me :P
Want to share or tweet this post? Please use this short URL: http://ozh.in/7p