A few months ago, I proposed a quite vague idea about a new “cartography” module for the upcoming LD23. Web Cartography is more and more used because of its curiously innovative and interesting aspect.
Now you may ask: “What’s the damn connection with Ludum Dare” ?
With the increasing popularity of the event, we see more and more game proposed for each LD session. Also, the initial idea was to realize a cartography of the submitted games.
To have a better visualization of the whole game submissions. Take a look to statistics in an original and interactive way.
- Which games are available for a specific platform? Multi-platform?
- Which games have more votes, coolness? (main nodes) => Imagine a visual helping tool for voting.
…and numerous other possibilities. (Why not something more realtime-oriented based on database snapshots?)
Proof of concept:
Using available public data and python scripts, I extracted and classified data concerning each game entries of a given Ludum Dare composition (platforms, ratings, creators,votes…). I’ve written a small web application displaying large directed graphs, generated from these data sets.
You can find my work over here: http://cboissiere.com/projects/ldviz/
Don’t be afraid by the messy aspect of those graphs, it’s mainly because of the huge size of the data sets. And don’t forget it’s still experimental =)
And of course, the source code is over here: https://github.com/cboissie/LD_Viz
Tell me more:
It’s basically two kind of graphs:
- WordCloud: We extract each words from all game titles. The words used together in a same title are linked to each other. For instance, if you click on the “TINY” node, you will see all the words that were used conjointly (like “WORLD”, or “PLANET”). The size of the node is proportional to the word occurrence.
- MultiPlatform: In this graph, games and their respective platforms are linked (Windows, OSX etc…). The size of a platform node is proportional to the number of game ported on this platform.
- You can change anytime the dataSet (between LD21,22 and 23) and the graph type.
- Zoom with the mouse wheel.
- Click on a node to see its immediate neighbors.
- The “Start algorithm” button apply a “Force Atlas 2″ algorithm to the current graph (see http://en.wikipedia.org/wiki/Force-based_algorithms_(graph_drawing)). You can stop the execution of the algorithm by clicking again on the same button. This algorithm will place the nodes in a more convenient way, give it a try!
Here’s a quick web app prototype for visualizing interactive graphs of game entries from old Ludum Dare compos. You can see two kind of graphs: Word cloud (most used words for a specific theme) and Multi-Platform (Game names associated with their respective platform(s)). http://cboissiere.com/projects/ldviz/
Feel free to contact me at clemzbox[at]gmail[dot]com or via Twitter.