Tag Clouds and Word Clouds

A tag cloud is something very new (considering the age of the technologies that it builds upon) but very catchy. It ought to be so, since it is simple enough to program and has great visual appeal.

Let's say you decide to display the 50 most popular tags for your pics or your articles or for what people search through your website right on your homepage.
You can do so using a long list.
You can do so using a more elaborate graph that plots the tags on their count.
You are, now, not far from being kicked by both who look for information as well as those who fall for aesthetics.

Nobody wants too much of facts. My day isn't gonna be any better looking through long lists and learning about occurence of each term and their spatial representation. Selfish me only wants to see them being of some use to myself. The tags should promote activity, not have people correlating their usage statistics to build their thesis. And that is where Tag Clouds drive home the point. They present the information in more novel ways; in ways that would catch your eye and become a navigational aid. One can find them on several websites and being used in many presentations where too much detail mean popping a sleep pill in your audience's glasses. Read the Wikipedia article for more.

They are really IMPRESSIVE to see and EXCITING to implement!

An article titled "Building Tag Clouds" by Jim Bumgardner (O' Reilly) piqued my interest into this stuff (On a personal note, I'd had a rough day and anything connecting back to my geeky days and ways was only waiting to be consumed). So I decided to implement a Word Cloud - something that takes in text, and presents the words visually weighted by their frequencies (much akin to the occurence of a tag). The method of operation was simple: Throw in any text and the rest is logic.

You get a nice word cloud. Put in random text, and its fun. Put in your emotions and its introspection. Put in your old diary entries and its retrospection.

Now only if my website server weren't down will I upload the slick file.
See and use SAoS' Word Cloud Generator. It uses regular expressions to extract individual words. The frequencies were extrapolated to font sizes using a log function, which helps to balance the graph visually otherwise some entries would be HUGE while some very small. The log function was a revealation, and its just plain maths! Spread the link around.