My background as a Systems Engineer combined with my phD studies in History of Religions (with a specialization in Tibetan Buddhism and a sub-specialization in Hinduism) has allowed me to be particularly qualified to develop a number of academic projects.
Tibetan Translation Tool
The Tibetan Translation tool takes Tibetan language passages, in Wylie transliteration or Tibetan script, and breaks them down into their component phrases and words, displaying its corresponding dictionary definitions from various sources. The tool was developed and implemented in its current state when I was studying at the University of Virginia. The Tibetan script input facility was built by Edward Garrett while working at the University of Virginia. It has been continuously running in THL servers since 2002, having gone through multiple updates throughout the years. It is the site that receives the most amount of traffic in THL.
Knowledge Maps
THL has invested considerable energy in building a library-wide system for managing semantic maps of various phenomena or areas of knowledge. Browsing collections through context-specific semantic maps offers distinct and valuable benefits that are complementary to searching. Browsing such semantic maps allows a user to explore an area of knowledge systematically and as a whole, rather than just finding granular discrete objects without a sense of the dynamic interconnections that make sense of such objects.
The key to our approach is "knowledge maps", our label for the hierarchies which formally represent specific topics such as Tibetan understandings of the types of animals, literary genres, Buddhist ritual forms, or languages. These maps not only express the hierarchical relationship between categories and subcategories, but also each category - a specific animal, literary genre, ritual type, or language - can be described by a scholar, or multiple scholars. Just as importantly, these knowledge maps are used as controlled vocabulary and tags in all other THL applications. Thus rather than each application - the place dictionary, or the image database, etc. - building and maintaining its own controlled vocabulary or ontology for a given subject matter, the knowledge maps globally maintain these for all the applications. In this way, the knowledge map then comes to function as a rich index of a large number of resources which are being indexed with them. For example, if you explore a knowledge map of Tibetan rituals, you will be able to potentially see for each type of ritual whatever videos, images, articles, web sites, dictionary terms, or other resources have been indexed as relating to that specific type of ritual. In this way, as scholars work in totally different projects, but all rely upon these common knowledge maps, a rich and complex understanding of the areas in question accumulates for end users.
The system will also work in the opposite direction, namely the name of any node of a knowledge map used to index a resource will be hyperlinked in that resource's metadata to that specific node within the knowlege map. Thus a user can be looking at an image that is tagged as pertaining to a "fire ritual", and then immediately proceed to explore a rich assemblage of scholarly resources on fire rituals.
The editorial interface was developed by myself under THL in Ruby on Rails. Currently Terms Dictionary continues to be actively developed in terms of software and content currently under the Newar Online Dictionary at the University of Toronto.