The history of network visualization has its roots in the 20th century. Whyte and Coleman both frequently used sociograms to visualize their data, although this was determined by their own artistic and analytic eye (Moody). Gephi is one of the most modern graph visualization applications available and certainly adds to the statistical analysis world. The importance of Gephi and visualization tools is directly from human interaction. It is this human interaction which allows the tool to “leverage the perceptual abilities” of a user to find new features within data (Bastian). This is a very important point on the success of Gephi as a digital tool. The information creates a new capacity for human interaction which is otherwise vacant by a brief summary of statistics (Moody). The design of many digital tools for textual representation only, works as a juxtapose for Gephi. A wide variety of designers were forced into providing tools and transforming prototypes to fit a wider audience (Mirel). However, Gephi, as this review will show, represents a friendly user interface with a sleek design and simple user controls.
The main objective of Gephi is to provide a interactive visualization and exploration platform for all kinds of networks and complex graphs. Gephi definitely succeeds in proving features which help this objective to be complete. The importance of understanding network theory will be discussed further, later on in the review. For now, a simple description, is that a graph is made of nodes,a person, place or thing, and edges,a relationship between nodes. There are for many aspects of Gephi which open up further and extend the knowledge. Developed modules are used to import, manipulate, spatialise, filter and export networks. Visualization uses 3D rendering through a computer’s graphics card to ensure the processor can focus on a variety of tasks. Nodes can be personalized to include images and no overlapping. Algorithms allow for the data to be molded by the user, this includes real time movement involving speed, gravity, repulsion, auto- stabilization, inertia and size adjustment. These algorithms are easily selected and work in real time which ensures that any user can benefit from the feature.The open source aspect of Gephi extends to the use of free plugins within the application. These help the digestion of large amounts of data, for example the Semantic Web Plugin which allows for specific Sparql, a query language, searches through large files including Dbpedia, a database of Wikipedia entries.
Technical problems with computer applications generally begin with the installation of a product. Gephi proves to be reasonably difficult to install as a result of its Java dependency. Windows 8 will have issues if the newest Java is installed and certain files will need to be altered and changed to ensure the correct running of Gephi. However, away from the installation issues Gephi is a very friendly application which has a plethora of tutorials and a quite simple user interface. The main benefit of using Gephi is that the program can be as complicated or simple as the user wishes. If a user is more comfortable with using spreadsheets such as Microsoft Excel then personal information can be placed into the application and made into a graph. A user can delve further into the information and use query language and huge data-sets to represent a larger corpus of texts. The most popular use of these data-sets for Gephi is Sparql, which was mentioned previously. The knowledge of coding does not need to be present beforehand, there are various sample queries and tutorials online which can help towards understanding the data further. This is an important point when discussing the impact that Gephi, as an application and Digital History tool, could have on history.
The impact on the historical community has not been very widespread. Gephi is a untapped resource which would prove very useful for historians if used to a greater extent. Harvard had a series of projects which made use of the program and had some very interesting results. The research team noted that looking at a matrix or relationships isn’t exactly new but the advent of digital technologies is spreading the idea across disciplines (Harvard). One specific example proved to be very compelling. The Inner Life of Empires includes five connection graphs which makes use of various Gephi plugins. One of the graphs can be seen below which represents the relationships between certain people and a full list of information, giving attributes and undirect links. This information is accessed when a node is clicked on and every node within the graph holds the same amount of data.
Two further popular projects which have used Gephi as a basis are the conjunction with the Google page-rank algorithm to find the most influential novels of the 19th century and the Republic of Letters. A closer look at the Republic of Letters shows the incredible scope which is available using Gephi. The website offers a list of letters and the relationship between some of the great minds in European history. Below is a graph showing the letters of Voltaire, there are various other visualizations of the data and allows for a user’s own method of inquiry. This is an important aspect of history as it grows towards a more universal public history mindset. The true beauty of Gephi is presented here and illustrates just how essential and key it could be for furthering history projects.
In conclusion, Gephi is a project which is being used sparingly in the historical world. If history is to keep track with the digital age then its need to find new and useful methods of dealing with information is important. An application like Gephi allows for a person to heighten their perceptual abilities using graphs and data visualizations (Bastian). Important conclusions could be found if the use of this application came to the forefront. However, Gephi will never reach its full potential in the subject of history unless effort is placed into learning about networks, graphs and a variety of digital tools. An education which, if present within a history degree, may help further the discipline into the modern world.
Moody, James, Daniel Mcfarland, and Skye Bender‐Demoll. “Dynamic Network Visualization.” American Journal of Sociology 110.4 (2005): 1206-241. Web.
Mirel, Barbara. “Building Network Visualization Tools to Facilitate Metacognition in Complex Analysis.” Leonardo 44.3 (2011): 248-49. Web.
Bastian, Mathieu, Sebastien Heymann, and Mathieu Jacomy. “An Open Source Software for Exploring and Manipulating Networks.”International AAAI Conference on Weblogs and Social Media (2009): n. pag. 2009. Web. 01 Dec. 2014.
Center for History and Economics. “Visualizing Historical Networks.”Visualizing Historical Networks. Harvard University, 20133. Web. 01 Dec. 2014.