It’s a little web-based application, and as it stands it is fun but only useful for very simple cases. Here’s an image with data I took from a paper by Scavarda, Bouzdine-Chameeva and others, about causal maps:
I made this with simple text like this:
“Cognitive maps” -> “Undirected graphs”
“Cognitive maps” -> “Directed graphs”
“Undirected graphs” -> “Knowledge maps”
“Undirected graphs” -> “Concept maps”
“Undirected graphs” -> “Mind maps” …. …. and so on.
When Diagramic makes graphs from plain text like that above, it will use one color and one shape. The automatic layout is fine with simple, regular graphs, but it needs manual adjustment for even the modest one above. As you can drag nodes wherever you want them and adjust the zoom to make best use of the fixed-sized box, it is easy to achieve a presentable result.
The biggest weakness is that to preserve the result, you must use the PrintScreen key on your keyboard and Word or some graphic-editing software to preserve and perhaps crop the image. There is no other obvious way to make a chart and embed it on your own site.
Diagramic can also accept data from spreadsheets, so as a component in a mash-up, it has more promise. Forbes.com have made a web component that feeds data to it, selected from its own database, to allow users to generate relationship diagrams dynamically. Here’s an example image taken from the Diagramic web site:
The web gadget is at the Forbes site. Tip: Start by selecting a network from the little combo box at the top centre.
Diagrams generated from Google spreadsheets will have 2 colours and 2 shapes – the first column in table is considered to contain the prime data. There’s no detail on the site about how the many colours in the Forbes diagram were achieved.
See you on Twitter or back here soon.