Blogviz, by Manuel Lima is probably the tool that comes closest to what we want to do in our MemeMapper project. After reading Manuel Lima’s thesis and after trying out blogviz itself – for both I spent several days – I won a lot of insights and had sometimes the feeling of listening to an unknown relative, who thinks in a similar way as we do. Blogviz is definitely a benchmark for our own attempts, especially in the area of visualisation and interface design.

The visual language of blogviz is very clear and appealing, but the main problem that occurs seems to be that you don’t have a first sight experience – like you do have for example in Marcos Weskamps Social Circles; Some kind of immediate light bulb effect – in Marcos map you see the central persons at first glance. But maybe this is only a question of the map’s target group

Certainly a map’s design always depend on its readers and as Manuel focuses on people with detailed interest in diffusion processes like e.g network researchers, it is legitimate to use a interface that requires some time to learn how to use it. – Marcos on the other hand provides a tool that targets ordinary mailing list members, especially new ones, who can almost immediately detect the important people, the hubs.

In our case we want to target both groups: ordinary weblog users who need a first glance information AND expert users with specific interests (like researchers or marketing managers). For us it is obvious that we need at least two different interface solutions. I personally don’t believe that there is the one and only solution for all purposes. Manuel’s thesis by showing all steps of development and different prototypes gives a perfect insight in the variety of possible solutions. By choosing finally a diagram form that he derived from a train time table by Marey, Manuel focused on a specific” view” at data by excluding others. But the perfect view always depends on the people using the map. – silly example: for a car driver there’s not much sense using a map visualising population density. – As Manuel is interested in research it would be very interesting to know what researchers think of blogviz. Finally this should not only include user testing but also the direct involvement of researchers in the development of maps and data aggregation.

Maps do not have an absolute or objective meaning in the sense that they “map” reality 1:1. Even geographical maps use different projections methods, like e.g. Mercator projection which favours northern countries in size. (They simply become bigger and therefore more important) So maps should pragmatically be seen as extensions of our senses and our intellectual abilities. We are free to choose either this or that form as long as we make its specific view clear to users. Through usability testing and listening to our users we are able to work on a map’s usefulness as a tool for thinking and communicating thoughts. We increasingly enable people in comprehending the growing complexity of network processes. As people tend to become more and more distinctive in respect to their interests and thinking there seems to be an increasing demand in different views at the “same” domain. - but is it then really the same? However the decision for one map or another is always a kind of invitation to others to share a certain view on a phenomenon and to follow a certain trail of explanation. In that sense Blogviz is definitely a strong invitation and already provides very interesting results in respect to dissemination processes.

more info about Étienne-Jules Marey:
Wikipedia about Étienne-Jules Marey. Marey became particullarly famous for his movement studies, which seems to be a related topic to the visualisation of dissemination processes. Marey was a cinema pioneer; Obviously mapping diffussion processes could benefit a lot of moving imagery or animated maps - we also used animations in our first prototype.
The relations between diffusion, movement and animation would clearly be worth a deeper investigation. Maps do allways catch a moment in time, whereas we are interested in what happens in between:
here’s a tool (a “photopgraphic gun”) marey used to catch the “moments in between moments”:

source and large version at wikipedia

Related:
Manuel Lima’s visual complexity is a great ressource for all kinds of diagrams, maps and visualisations.

Small-World Networks show that even a few random links in a highly regular network may reduce the average distance between vertices dramatically. (The original paper by Duncan J. Watts and Steven Strogatz is very short and easy to understand: http://tam.cornell.edu/SS_nature_smallworld.pdf)

Strogatz/Watts introduce two different indicators to measure the degree of linkage within large networks:

1. Average Path Length which measures the average distance between two vertices (e.g. persons) in a given net (e.g. world population).
Suppose you want to send a precious present from Person A to B from one end of the world to the other, by avoiding regular mail services. You only trust friends, and friends of friends. In this case path length means the chain of friendship connections between A and B. If A is a friend of B the Path length is 1. If it is 100 than there are 99 friends between A and B who help in sending the present.
Some pairs of vertices (in our example persons) may need fewer links that A and B, some others may need more.
The Average Path Length is the average number of connections needed to link pairs of vertices in a given network. Thus the Average Path Length is a good indicator of the network’s overall ability to bridge long distances in it.
It’s a macro indicator and it doesn’t say much about linkage in smaller parts of the network..

2.
Additionally Duncan J. Watts and Steven Strogatz (1998) introduced the clustering coefficient to measure the interconnectedness at the “neighborhood level” of a network.. The clustering coefficient reaches its maximum of “1” when all possible links within in a neighbourhood are links indeed. (or in other words, within a neigbourhood all vertices are connected with each other). For details see: http://en.wikipedia.org/wiki/Clustering_coefficient

It is noteworthy that the clustering coefficient does not measure whether there’s a cluster or not. It is based on the assumption that there is already order in the network.. In the example of Strogatz and Watts it is a perfectly ordered ring of vertices each connected to it’s direct neighbours and the next but one.
Strogatz and Watt has demonstrated (see graph) that by introducing only a few random links in such a perfectly ordered structure the average path length decreases dramatically whereas the clustering coefficient remains almost the same (i.e very high). A few random links are enough to turn the network into a small-world that combines both: the ability to bridge long distances in short pathes (few amount of links) and a dense web of edges within a neighbourhood (high clustering)

If we tried to apply the clustering coefficient to an application like the MemeMapper it would imply that we needed to define the neighbourhood of Weblogs in advance. The clustering coefficient would allow us to measure if a predefined set of Weblogs (a predefined cluster of Weblogs) is highly clustered or not (in other words if there is dense web of edgdes between vertices or if they are rather loosely linked.)

Special algorithms like Mark Newman’s “Fast algorithm for detecting community structure in networks” can be used to detect clusters. Newman’s algorithm was used in Vizster, developed by Jeffrey Heer and Danah Boyd,.

For details see:
The orignal paper by Watts/Strogatz: Collective dynamics of ‘small-world’ networks
http://en.wikipedia.org/wiki/Small-world_network
http://en.wikipedia.org/wiki/Clustering_coefficient

I had a closer look at another classic of network visualisations: They Rule, by Josh On.
There are three factors that make this map so outstanding:
1. Data of company boards and directors - including links to all websites
2. Easy to use (some features are missing like a “back” button or a “select and delete” function
3. and probably the best of all: the possibility to save maps and show them to others.
This feature makes “they rule” a community based tool for the analysis of power and its networks. Everybody can uncover new connections among Politicians and Industry, save it and show it to the rest.

Barabasi - in his book “Linked” - mentioned a story about an actor called Bacon who was not well known during his career, but who became quite popular among network scientists. He’s a good example that in a small world (like hollywood) also less popular actors seem to be hubs; but bacon’s connectivity is more a attribute of the network and less an attribute of himself. There’s a nice visualisation of that story available at: http://www.netmapanalytics.com/demo.html.

Implicit Structure and the Dynamics of Blogspace was written byEytan Adar, Lada Adamic, Li Zhang, and Rajan Lukose, from HP Information Dynamics Lab .
Its a quite early paper (2004), and it seems as if its authors had started at more or less the same time (Spring 2003) as we did the first Blogosphere Map prototype. Whereas we were focusing on the aesthetics of diffusion mapping, the IDL focused clearly on its analytics. The work done is quite impressive as it poses for the first time the relevant questions:
How can we analyse infection pathes, when there’s no explicit information about how news (represented by an URL) travelled through the blogosphere? (because there are only a few “via” links) How can we infere Infection routes? How can we measure similarity between blogs in order to infer Infection routes.
The authors not only posed the right questions but also gave competent answers by formulating measuring methods like blog_similarity and iRank. It opens up a wide field of further research to be done like e.g. more investigation about the different weight of link_similarity of Weblogs versus text_similarity versus infection timing in respect to inferring infection routes. Probably also other methods can be found.
In any case the paper proved that there are methods to map the general collaborative structure of the blogosphere, by identifying general (i..e. more likely) trails of infection and it is possible to infer infection routes by embedding explicit links in those general trails of infection.
related:
K-means clustering
Wikipedia: Custeranalyse/k-means
K-means-demo explains the method quite obviously.
Kruskal-Wallis Test
TFIDF Scheme (deutsch),
Support Vector Machine (SVM) , try out
better introduction than wikipeda
LIBSVM — A Library for Support Vector Machines (used for this paper) Introduction for SVM-beginners by the creators of LIBSVM
Graphviz was used to generate graphs.
Zoomgraph

Vizster, developed by Jeffrey Heer and Danah Boyd, is certainly one of the best designed tools for visualising online social networks. By mapping the web of Friendster relationships Vizster proves that a visualisation can be both easy-to-use AND powerful at the same time. Its simplicity mainly derives from the fact, that users, mainly Friendster users, immediately understand the purpose of it: Looking on something very familiar, namely their own relations, from a new perspective, a bird’s perspective. Simply changing the perspective can be seen as a powerful means to initiate new cognitive processes and to amplify thinking.
I think a good map needs a shift of perspective, that on the one side needs to be rather radical (from worm’s eye to bird’s eye) but on the other side is immediately readable.
The sensation of discovering “something new” is not an attribute of the map: it is triggered by the map but takes place in the user’s mind. A good map like Vizster seems to bridge two previously unlinked nodes in our cognitive system with the result of perceiving something that we didn`t perceive before. Maps are in a way “new and attractive perceptions on the silver tablet”. – (Wow, I didn’t know that…)
In Vizster’s case the map bridges the everyday experience of having Friendster relations with their everyday ability to read maps. The egocentric view makes it even easier to understand the map’s visual grammar. A user discovers himself or herself in the middle of a web consisting out of his friends. He or she intuitively understands that the map is about him/her and his/her Friendster relations.
This immediate success is very important, because it invites to start a playful exploration of other aspects and features. Furthermore it provokes the formulation of “ad hoc interests”: Why is there a cluster? What do they have in common? Ah, they’re all students! Of the same University? Yes, indeed!
A good map like Vizster provides exactly that kind of functionality that is needed to follow your ad hoc interests. The intuitive learning of new functionality is therefore guided by ad-hoc interests and deduced by previous experiences (within Vizster and by interacting with similar maps). Vizster helps you to find your own way.
Designer often make the mistake that they want to invent something “new”, like “a revolutionary view at data”. But the revolution however takes place in the users mind and if the usage of a “new” map is not based on everyday experiences, it will fail. Good design picks user up where they are and invites them to explore something new. If the first step – understanding what the design is about, where to start… - is to big there is a high risk that you loose a lot of users.
Once the user started successfully using a map and already made first rediscoveries there’s a willingness to learn new functionality, like Vizster’s linkage view.

Related:
UCINet , “A comprehensive package for the analysis of social network data as well as other 1-mode and 2-mode data”,
JUNG, Java Universal Network/Graph Framework,
GUESS, is an exploratory data analysis and visualization tool for graphs and networks,
ContactMap, Organizing Communication in a Social Desktop,
The TouchGraph LiveJournal Browser allows one to visualize and explore social networks,
barnes-hut algorithm,
Fast algorithm for detecting community structure in networks, by Newman, M.E.J.,
Prefuse, A Java-based toolkit for building interactive information visualization applications. Vizster is driven by prefuse.