The Top 50 Classical Blogs, Using 4 Different Methods
Some of you will probably have noticed that a couple of days ago ACD decided to release his own list of the 50 top classical music blogs (which incidentally should more properly be called the top 53, using his current dense-ranking method) based on the number of incoming links according to the god of the internet, Google. In the process he somewhat pissed off a few people, notably Scott over at Musical Perceptions who has been compiling his own rankings for the last couple of years (using the Technorati authority number) but whom ACD neglected to credit with prior art (there is a rather bitter exchange in the comments over at Musical Perceptions here). Also myself (though I’d say it’s irritated then pissed off) due to refusing to use the URL which everybody actually links to in their blogroll to determine the number of incoming links. However, it is his blog, his screening choices, his rankings list.
Well, it occurred to me that since I spend every day managing large chunks of data and automating stuff I could probably whip up a script which automatically calculates my own classical blog rankings, according to my own choices. So I did exactly that. There may be a couple of bugs still, if you find anything in error please let me know so I can patch it up. I was particularly interested in trying to perform this in the most scientific way possible, so I’ve used four distinct methods of ranking. There is the Google incoming links method (as ACD uses), the Technorati authority method (which Scott uses) but additionally I use Bloglines RSS feed subscribers (which Chris Foley did once) and a completely new method, which is the number of Google Reader subscribers.
Now, I’m considering this a bit of a beta-test since I probably missed a few blogs which should have been included (I tried to combine ACD’s list and the Musical Perceptions list, along with a bunch of others, the complete list I used is here, add yours if I’m missing it please!) and there might be a couple of mistakes. While the program I use obviously has a larger chance of having a bug then a human does (depending upon caffeine ingested), the huge advantage is that it only takes around five minutes to collect and analyze all the data for about 100 blogs. What I’d like is for people to suggest improvements, for example, should I combine the results together to produce a super-ranking? Should I combine the RSS feed data with the Technorati authority to give a “current readership” number? If people are interested I’ll polish it up and analyze it differently.
So without further ado, here are the “top” 50 blogs as of yesterday using four different ranking methods, the first number next to each blog is it’s rank, the number in parentheses is the total number of subscribers, number of incoming links, etc.