Tag clouds are a relatively new way of representing the frequency of keyword usage for a given data set. The set may consist of a group of tagged photos, a group of blog postings, or any body of text or tagged data objects, for that matter. Words/tags that are used more often are displayed using larger type in the tag cloud.
Flickr uses tag clouds to show the tags associated with images that have been uploaded to their site. Here’s a link to my personal Flickr tag cloud, and a link to the overall Flickr tag cloud.
There are a variety of techniques for creating and maintaining tag clouds. Currently I’m using a system provided by TagCloud.com. TC generates tagclouds based on RSS feeds from user specified sites. One thing that makes their system stand out is that they don’t just rely on categories or explicit tags. Instead, they analyze the text of the incoming feeds, and extract keywords. Though not perfect, it provides a richer set of tags than simply using categories and tags.
My cloud tag, shown below, shows the last 250 keywords from this blog, my new photoblog, luminosity, as well as the tags from my Flickr photos. You can click on the keywords to see a list of the postings where the the tags are used. And, that’s the secret, or the hope that tagcloud.com has in mind for providing this free service; the results page also displays low-key google ads. Seems a fair trade-off.