Free to use, simple social network mapping software?

January 2, 2008 2:34 AM

Looking for a simple, free (for commercial use) social network mapping / graphing tool

I have an immediate need to do some online community relationship mapping. It's not something that I'm likely to need to do frequently, and I don't need any significant degree of complexity or customisation, so I don't particularly want to pay for a software application like InFlow that emphasises added-value aspects.

Basically, I want to map the relationships between less than 100 blogs, using a simple binary cross reference based on the blogroll / recommended sites. I'd like to output a visual map that can be cosmetically tweaked (font, colours and group positioning) and exported as a jpeg or similar to be pasted into a larger report.

Obviously I've done enough research to know what sorts of academic pages / industry resources about these sorts of tools are available, but I'd like to get it right first try as I don't have much time for trial and error with this project. So, really I'm looking for recommendations issued on an informed basis (first-hand use or similar).

Cheers!

posted by bifter to technology (8 comments total)

Perhaps it would help to know particularly what level of functionality you're looking for. Are you really looking for a free tool specifically tailored to social network mapping? Are you expecting something you'd give a list of sites to that would do everything automatically? Or would a straight diagramming tool that would simply let you lay out a graph based upon the data be helpful?
posted by XMLicious at 4:35 AM on January 2, 2008

My functionality requirements are very low. I just need to create a network visualisation graphic that looks sufficiently smart to be included in commercial work, but I have too many groups in the network to make producing something manually practical. To give an idea of the sort of thing that I'm thinking of, I want something similar to InFlow, that is free to use and need not have all the bells and whistles.

I have the data sheet already, hopefully in a usable format (an Excel sheet cross-referencing groups with a simple binary notation for whether or not there is a link from one site to another). I can work on the data format if necessary.

A straight diagramming tool would be okay, as long as there is some ability to tweak positioning / format etc.
posted by bifter at 5:12 AM on January 2, 2008

I asked a similar question and GraphViz worked out nicely for me.
posted by divabat at 5:37 AM on January 2, 2008

Yeah, Graphviz is the thing. His link was broken, it's here.

I can't remember what Graphviz's output formats are but if it can create an .svg file you should see if you can import into Inkscape. That's a graphic design tool that would probably do a better job of converting to raster / jpeg and will allow you more stylistic finesse than Graphviz would. The tutorials under its help menu are fantastic.
posted by XMLicious at 5:55 AM on January 2, 2008

Haha, I'm a she, but thanks for catching my broken link. oops!
posted by divabat at 6:47 AM on January 2, 2008

Bah, lost the coin toss, I should have taken the few seconds to check your profile.
posted by XMLicious at 7:00 AM on January 2, 2008

Try GUESS or Pajek. Both excellent software tools for visualizing social networks.
posted by formless at 10:10 AM on January 2, 2008

I threw my lot in with GUESS in the end, as it seemed from the label on the tin to be more flexible with both inputs and outputs.

As it happened, it was a screaming horror for a non-techie like me to get to grips with (horrible documentation, and seems to assume intermediate ability in Python coding. Me = clueless on that stuff), and frighteningly inefficient to acquire and format the graph data in the right way (I guess automated harvesting / formatting is one thing that it's definitely worth investigating a commercial product for).

Anyway, after much anguish and bloodshed, I got there in the end. Learned some useful stuff and got a reasonable output just about in time. Thanks for all suggestions.
posted by bifter at 1:33 PM on January 17, 2008 [1 favorite has favorites]