A scientific approach to blog ranking

Neville Hobson

•

28 October 2007

If you had to pick just 100 blogs that would provide you with all the content you need to stay informed on everything that interests you, which 100 would you pick?

And by which criteria would you determine that 100?

Some clever researchers at the Carnegie Mellon University in the USA have done the work for you with the publication of a ranked listing of 100 blogs.

This is not just any old list – it’s been arrived at through a complex scientific analysis involving outbreak detection, submodularity, node selections and sensor placements among many other things.

If that and the image above – taken from a presentation the researchers prepared – don’t give you a sense of the complexity of the research, how the abstract in the published research paper (PDF) begins ought to:

Given a water distribution network, where should we place sensors to quickly detect contaminants? Or, which blogs should we read to avoid missing important stories?

The 10-page paper, entitled “Cost-effective Outbreak Detection in Networks,” is a detailed reporting of what the research aimed to achieve, the methodologies employed and the findings that resulted.

Here’s the researchers’ rationale:

[…] Our goal is to select a small set of blogs which â€œcatchâ€ as many cascades (stories) as possible. A naive, intuitive solution would be to select the big, well-known blogs. However, these usually have a large number of posts, and are time-consuming to read. We show, that, perhaps counterintuitively, a more cost-effective solution can be obtained, by reading smaller, but higher quality, blogs, which our algorithm can find.

A very long tail approach.

I should mention at this point that my blog is in this list, at number 84. Actually, as far as I can tell (I haven’t checked every single blog), only one other UK blogger is in this US-focused list: Hugh MacLeod, at number 76.

So two Brits cut the mustard of blog indispensability :)

Here’s the complete list:

Make of it what you will!

10 responses to “A scientific approach to blog ranking”

Jed

29 October 2007

The analysis seems impressive, and certainly past my graduate days! But the relevance of a blog is closely related to what I want to see… This list seems USA biased to me.
Jessie Landis

30 October 2007

If anyone thinks this is important, the first glaring error is that No.58 hasn’t published anything since April.
Armin

30 October 2007

This is not just any old list

Actually, it is. The data is from 2006. Which in the high speed “blogosphere” I consider to be ancient.
Aloysius

30 October 2007

I was as surprised as anyone to find my blog at #77. I think it’s related to the fact that, during 2006, Catymology hosted a number of carnivals. I’m circulating the paper to some mathematically adept humans that I know, hoping they can explain this.
neville

1 November 2007

Well, I did write my post with my tongue a bit in my cheek. Maybe not far enough.

Some of the links in the list don’t work. Some blogs surprise me for being there, but I’m not the scientist.

As I said in my last sentence, make of it what you will. Which you’re all doing :)
Chris Edwards

5 November 2007

So, Neville, when you say your post was tongue-in-cheek, could you expand?

Did you think the CMU analysis was lacking in some way? Because that’s the only way I can read it as being tongue-in-cheek that these scientists weren’t as ‘clever’ as they made out. I wonder if you could share your insight.
neville

7 November 2007

Chris, the post is simply reporting on a different research method of ranking some blogs, compared to how you tend to see rankings.

I take little interest in blog rankings as a rule but this one caught my attention and my blog was in it. I’m certainly not crowing over being included in this list (I read your post); if I were, the post would have entirely been about that.
Armin

7 November 2007

My view on most blog rankings is quite simple: I think they’re trying to compare and rank things you can’t really compare and rank.

Most of them depend on number of links in Technorati or the number of visitors or similar metrics. But that’s assuming all blogs have the same audience and topics and are therefore comparable, which obviously they are not.

It’s a bit like taking a high powered sports car, a pickup truck for a farmer and a family/people carrier and trying to rank them to decide which is the best car. Entirely impossible, as that depends on the purpose of what the car is needed for, therefore the best car or a ranking as such can’t exist.
neville

7 November 2007

That’s what makes this relatively interesting, Armin – the lack of measurement consistency. Apples and lemons, etc.
Latest PowerPR blog index : NevilleHobson.com

13 November 2007

[…] I mentioned in relation to a recent post on another ranking of blogs, make of it what you […]