back to the main page


TolkienSearch - Concept Idea and UI Design

Middle-Earth joins Weather, Flights, and Stock prices for optimized Google Searches

TolkienSearch web view

As part of a class discussion, we considered various "specialized" searches that one might see in Google or other search engines for various queries such as stocks, weather, flight information, etc. For a brief homework assignment, we created our own specialized search. My specialized search delivers quick descriptions on elements from the writings of J.R.R. Tolkien.

This new specialized search for Google provides a more comprehensive search option for all things related to the writings of J.R.R. Tolkien. Currently, searching for character names, etc. from The Lord of the Rings, The Hobbit, and even The Silmarillion usually does return a specialized result, but it only pulls from Wikipedia. Therefore, the results are not customized and do not appear different from any other search with a Wikipedia result. There are many queries which have no Wikipedia entry (or the entry is not specific so it takes longer to find a quick description) because the character, place, or event is a small enough detail that it does not warrant a full Wikipedia entry; however, it is these detail elements that matter the most for these searches because minor details are more likely to require a detailed explanation (i.e. fewer people will search for a description of "Gandalf" than "Telperion" because Gandalf is more widely known).

This new system would pull it's data not only from Wikipedia, but also from "The Encyclopedia of Arda" (http://www.glyphweb.com/arda/), a well respected Tolkien reference site as well as the "Lord of the Rings Wiki" (lotr.wikia.com).

The interface allows for a quick summary as well as several photos (pulled from fan art as well as the listed sources) which can be scrolled horizontally in place within the result. The interface includes an excerpt from the above sources and could even have a custom combined description from multiple sources using natural language processing.

Or, if you prefer, you may view the above as a PDF.