Beyond measuring Dell Hells, it is helpful for those who track blogs to measure relationships between them.
When dealing with a small set of blogs, it is easy to determine if and then how they’re related. However, with millions of blogs no person or organization has the resources to accurately track all of them. Thus, having a automated system to establish relationship is very helpful. That’s a gap that search engines fill for the Internet in general.
Relationships could form around a myriad of factors like: topic, geography, style, stance, etc. For the sake of simplicity, I’m going to focus on topic for this post.
There are many ways that blogs can align themselves with others around a topic.
Semantic analysis can determine if blog posts have a large set of common specialized words that tie them together. Data mining isn’t perfect though.
How about links? Bloggers can link to other sites that address the same issues that they do. However, linking is not standardized throughout the blogosphere. In many cases links will lead to sites which cover a wide variety of topics. At times bloggers don’t hyperlink to other sites, even if it is helpful. Many people simply posts a list of links that interest them while others will include links in their text. Then we have to ask: What’s a more meaningful link — one in the blogroll, in-text, or part of a list of several other links to sites discussion different topics?
If a blogger is generous with links, then tracking the sites linked to is useful. It makes sense to connect two blogs to each other if one links to the other in at least half of its posts. But as I discussed above, measuring this way is hard since linking habits differ greatly.
Bookmarking and tagging sites like del.icio.us are helpful when establishing blog relationships as web surfers classify blogs and posts by using keywords and writing their own headlines. As David Weinberger explained in a commentary piece “The Value of a Man-Made Mess, on the Internet” during NPR’s All Things Considered on June 11, 2007, people can categorize web content in a variety of ways. For instance, tags like “Africa,” “animal,” “pachyderm,” and “mammal” are all applicable to a blog post about elephants. Further, one can tease out blog relationships by looking at how people have tagged or bookmarked a blog and posts. Granted, such categorization is rarely standardized, but Weinberger argues that not necessarily bad either.
These are just some ways I can think of measuring and defining relationships between blogs. What are some other methods?
