Open Source and Blogs

One of the biggest learnings from the little mapping experiment I am working on is the power of community. To build the application I relied heavily on the hard work of others in the Open Source Community (many thanks to the teams involved with: HTMLParser, Graphviz, and Xerces).
Being able to share ideas with others in the blogging community such as Steven Dulaney and Ross Mayfield gives real-time feedback that then sparks new ideas. Even though between pressures of work and family I am a little pressed for time I still feel part of a larger community through blogging. Thanks all.

Navigation Links

Stephen Dulaney e-mailed me that he liked the navigation map on my blog. Well it is the season, so Stephen here is your own navigation map [feel free to cut and paste].


This is just a small example of how make use of the visualization of blog links. While creating nice graphics is fun there is the question to be asked - what is the use of visualization?

Jon Schull makes the point that the number of nodes quickly makes the problem very hard to manage. Just going a few levels deep from my home page I get the following image map that has 250+ nodes. It is clickable and it is fun to navigate around and get to other blogs in one click rather than navigation through several other links.

The value of mapping may not be in mapping the raw connections but lifting patterns out of the maps. Sounds like another project...

Changes to RSS to make mapping easier

As part of my mapping experiment I had to parse through HTML and determine by heuristics which link was the RSS feed for the particular URL etc. There were two problems with this:

• Determining if the link was a RSS file: naming conventions vary widely, any chance we could all agree to use .rss as the extension?


• Extend the RSS schema to include referrals as part the schema. This would allow very fast mapping of blog relationships.

These two changes would allow blogs to be machine mappable without the current crufty coding required to walk through HTML files. This should lead to lots of unintended consequences....

Mapping my Neighborhood

For a while I have been interested in different ways to look at information and how people interact to create new ideas. After talking about it for a while I decided it was time to actually do something about it. So over the holidays I hacked together some code to map the surrounding blog space from a starting URL. It is still a little buggy i.e. I get unexpected results sometime.

The code parses a starting URL builds a map matching links to URL feeds using several heuristics. The rational is that if you have an RSS feed you are likely a blogger therefore you are included in the map. This does exclude some notable sites and may include some non-bloggers but it gives a pretty good initial picture.

This is the first image map of my local neighborhood to a depth of three links removed.
Now that I have all this information at my finger tips the interesting thing is what can I do with it and what other relationships can I start to determine and visualize.