Penguin March 18th, 2008
I came across this the other day and found it really intriguing.
Source: Moritz Stefaner’s Thesis on Tags
Elastic tag maps visualize the complex relationships emerging between tags (free-form keywords used to describe digital resources). The layout algorithm puts tags that are frequently used together closer together on a 2D plane.
I’ve always found tags, or human metadata, really interesting. Stefaner brings up some interesting points regarding the deficiencies and problems of traditional tag systems.
Just look at our tag cloud in our sidebar. We’re limited to only displaying a certain number of tags which precludes us from showing the long tail (the infrequent but more precise). It also doesn’t show our changing interests over time. Early on, I was blogging more on tech stuff, now, I’ve shifted to film and screenwriting. The tags don’t reflect that shift just yet.
Stefaner addresses those issues with two demos.
The first, is a 2D map of tags. It’s similar to a traditional tag cloud, but when you mouse-over a tag, it visually shows its ontology to other tags. Although the idea is good, it doesn’t seem very practical. One reason that tag clouds work, is that they can still be useful in a very small space. This system would require more real-estate.
The second is an emerging topics graph, seen on the right. This solves the problem showing changing interests, but it’s not useful. It’s just informational.
Definitely some interesting ideas and some good ideas.
Links:
+ Stefaner’s Thesis (PDF)
-Penguin
Related posts
Tags: Downloads, graphic design, Information design, Long-tail, PDFs, Tagging
) and 
