bioGraph: an example of MutaGraph©.

interested in my life? here it goes:

A MutaGraph printout, and, top-left corner, a detail as seen in the MutaGraph client.


This is bioGraph. It is my bioGraph. Due to lack of useful data representation in standard curriculums (eg. tabled, autobiographic books etc.) i decided to write some software which provides for a more "natural", but also computation-friendly representation of someones life.
MutaGraph© is the result. The pictures are actually composed of screenshots and printouts of the first MutaGraph test-data set, my very own life.
The curriculum is now representated in so-called "Nodes", typed data objects, as "Fred" of type "Person" or "Hamburg" of type "Location". These Nodes are conenected by directed arrows, called "Edges" whose are typed like "resides". So the sentence "Fred resides in Hamburg" would be described by two Nodes and one Edge.
While those snapshots viewed here consists of about 800 nodes and 900 edges, the graph now currently exceeds 2200 nodes and 2700 edges.


Different views: nodes with spatial information tied to satellite map

Some details:
MutaGraph is written as transaction-secured client-server-architecture.
The server consist of one simple MySQL-Database (now ~80k for whole bioGraph) and a PHP-backend for client's communication.
The client can be run as java applet, java application or windows executable. It is partly based on Sun's graphing toolkit and on the (althoug heavily modified) TouchGraph engine from Alexander Shapiro. It allows grahpical editing and database querys, and contains optimized algorithms to lay out the graph in 2d-screen-space for viewing or printing. In fact, a full flat representation of a graph cannot be done without crossings for general graphs (if, graphs are called "planar") so even perfect layout may still show up disturbing crossings.
In my opinion, many more types of abitary data besides curricular data could be represented as graphs, and it makes sense every time texts, tables or simple relational databases can't cope. I hope more information-related workers will understand this in future, and use graph-based tools moreoften than today.


Another screenshot of the bioGraph inside MutaGraph client

An idea...
Next step could be "interconnecting". If some people, better friends, do their own biograph, than the multiple databases could be melted by some kind of interrogating process, in which the both owners collect nodes, which represent the same object (place, person etc.) These nodes will be melt, and the resulting graph will represent the life of two persons... Iteraring this over many people, a whole "life-web" of a community could be builded, showing up all relevant connections these people lifes have.

If you interested in these software's technologies or you want to build your own biograph, just state me so (), and force me to do a real public release of MutaGraph. At the moment, it is undocumented and not very simple nor self-explaining, so download here at your own risk. Please state also how firm you are with servers, PHP, MySQL etc. to make me not documenting more than needed...
or just check it out online: (Java 1.5 needed) open Mutagraph Exploder Java-Applet for loading a graph click inside the url-field of Mutagraph Exploder and press the return-key. show MutaGraph in IdeaGraph which also based on it

MutaGraph manual (german, html)
theoretical paper (german, pdf)
download MutaGraphExploder with sources for obduction - this is common software

explanations of graphs:
Wikipedia: graph (english)
Wikipedia: graph (deutsch)

MutaGraph ©Paul Geisler, 2003