This paper describes the current state of RUgle, a system for classifying and indexing papers made available on the World Wide Web, in a domain-independent and universal manner. By building RUgle with the most relaxed restrictions possible on the formatting of the documents it can process, we hope to create a system that can combine the best features of currently available closed library searches that are designed to facilitate academic research with the inclusive nature of general purpose search engines that continually crawl the web and add documents to their indexed database.
Eric G. Berkowitz, Mohamed Reda Elkhadiri