I just saw this lovely network map of the “product space” of exports (Atlas of Economic Complexity, via Infosthetics). The idea is to reveal what types of products are related to each other in the sense that there are common economic competencies need to produce them.
I’ve created a page to compare various tools for embedding a network diagram on the web. I used a small example set of researchers’ keyword associations. As always, browser compatibility is a big challenge. Most of the options I tested required a fair amount of web coding ability to get them working. I’d love to hear about other tools and ideas I may have missed.
I just gave a talk at a conference on New mechanisms for linking research & policy. I’m posting the pdf of the slides here, which include some links to various examples and a few resources. Thanks to Comunica, IDRC, Panos Caribbean and all of the conference participants. Since the slides have links, but no commentary, I’ll try to develop this into a full post later.
I just spent some time comparing various tools for publishing web visualizations of region-coded data. I looked at some Google tools, ManyEyes, and Tableau Public. This is what I learned using my demo example. Please let me know of any other tools I should test…
Guess who is pumping in the money to support CA Proposition 23 to roll back California’s global warming legislation?
Yup, mostly its the companies that would be regulated when the law goes into effect. I’ve been working for the past few weeks to build prop23.dirtyenergymoney.com, an interactive network chart of the funding flows. Its an adaptation of our previous dirty energy money site, but using campaign finance data from California, and from OpenSecrets.org for the Federal PACs that are contributing in California. I think it is interesting to see the multiple layers, how the various group funnel in money and influence. I hope to be able to do an expanded version which would also show some of the other strings that Koch is pulling, why Chevron isn’t investing in Prop 23 (it seems they are backing 25 & 26 instead), and the the backers of the various No on 23 campaigns.
Also, can anyone help me figure out what happened to the $5000 contribution from Western Petroleum Marketer’s Association? It was in an earlier version of the data, but now seems to be missing from the filings. It seems that Tesoro may have solicited support for Prop 23 in their meetings, and I’d like to be able to include them.
Update: We’ve added panning and zooming features on the graph. We’ve also done a second chart for CA Proposition 26: prop26.dirtyenergymoney.com
[download full map as pdf]
I’m interested in making political structures visible. Trying to put those half-realized connections and linkages into a tangible form–a map that we can point to. I recently located an online database of conservative funding relationships created by Media Matters Action Network. I was able to scrape the site and (with their permission) experiment with some network diagrams in pdf form to visualize the funding relationships among “angel investor” foundations and right-wing organizations.
… but only a few owning companies. Another great study and diagram of brand ownership relations by Philip Howard.
…To visualize the extent of pseudovariety in this industry we developed a cluster diagram to represent the number of soft drink brands and varieties found in the refrigerator cases of 94 Michigan retailers, along with their ownership connections.
Although, according to the Soda vs. Pop map, since the study was in Michigan, maybe it should be labeled “Pop” not “soft drinks”? ;-)
The Journal of Social Structure did a special online Visualization Symposium with peer-reviewed network visualizations. A good format, a great idea, and some nice viz examples. Hopefully next time there will be more submissions.
I thought that this map of overlapping topics between media outlets was a cool idea. The resulting network seems like a fairly undifferentiated core-periphery structure, which seems typical of a lot of topic-maps-as-networks I’ve seen. Does this reflect a property of media networks, or is this what topic maps look like? Maybe a threshold filter on the edges to then out weak (or strong) ties would reveal more subtle structure? I thought the blockmodel reduction of the network was a helpful summary. The model does seem to be backed by some substantial statistical work and
… “Results of the estimation indicate that both production volume and common ownership affect the topic overlap of news outlets.”…
My submission was cleaned-up interactive version of the bill-endorsement network from MapLight data with click-through to bill summaries. The layout was produced using SoNIA (now with variable node label sizes!) and the MDSJ library.
Michael Heaney and Fabio Rojas just released another great network map in a blog post. This one shows the co-mentions of topics (as coded by the researchers) appearing in the descriptions of panel discussions at the recently concluded 2010 US Social Forum in Detroit. The map functions as a coarse-grained representation of interconnectedness of the various topics, and presumably how important and relatively central they are to the activists and organizers participating in the forum.
A lovely organization co-membership network graph of the delegates to the 2008 party conventions. From Networking the Parties: A Comparative Study of Democratic and Republican Convention Delegates in 2008 (Pre-print. Seth Masket, Michael Heaney, Joanne Miller, and Dara Z. Strolovich). The authors surveyed the delegates at the major party conventions about which groups they were members of. Nodes are groups, links indicate delegates who named both groups. Red only nominated by Republican delegates, Blue only Democratic, Purple = Both. The image gives a nice quick overview of the relative positions of groups, the paper gives more detailed analysis.
Interesting but not surprising to see the Sierra Club / NRA relationship, and NAACP / Young Republicans membership as well. I’m curious what the single red node is between Amnesty International and Sierra Club… Would like to see a version with all the organization names. Does this image match your intuitive sense of the relative political positions of the organizations named?