
[1] - The user selects between viewing the whole database or just a sampling from a specific city, depending on their curiosity and if they're willing to put up with longer loading times.

[2] - The main interface loads and the viewer sees each posting is represented by a small symbol, in an ambient cloud. The user can uncheck the boxes at the left to hide entries from specific categories.

[3] - The user types in a word and presses enter. A small timer indicates the interface is loading.

[4] - The user's word emerges in the scene. Postings that contain the word flock towards it and highlight in color; those that don't fall to the bottom.

[5] - The user can also add another word by typing it into the text box and pressing enter. Again, postings flock towards words that they contain and highlight themselves. Postings hover in between if they contain both words, and highlight to an intermediate color. If they contain more of one word than the other, than their position reflects that bias.

[6] - If the user mouses over one of the words, the interface shows a small X that they can click on to remove it from the visualization.

[7] - If a user clicks on one of the posting icons, a more detailed view of the original post shows up.
-----------------------
Things I need help with:
-----------------------
- How can a user tag or highlight specific postings to keep track of?
- How are the positions of the user entered words determined? To more prevalent words come forward automatically? Does the user interact with or even specifically change them?
- My intention with adding color is to allow the postings to retain a reference to their user-defined groups. So if I add other views, say, geographic or demographic, the user can compare it back to this view. For example, if the user searches "yins" and then does a geographic view, they might find a concentration in the Pittsburgh.


