PanLex - The International Multilingual User Group (IMUG)

Imagine that your native language is not supported by your government, and you do not speak a more widely-known language. Interpreters are not available, and machine translation does not exist because of inadequate data or lack of interest. How might communication be possible when you need essential information or services in areas such as health, law, land rights, agriculture, education, disaster relief, and many others?

The mission of PanLex, a nonprofit project of The Long Now Foundation, is to overcome these language barriers to human rights, information, and opportunities. A possible model for addressing the problem is the book Where There Is No Doctor, which contains essential treatment information for people with limited access to medical professionals; the WHO calls it “arguably the most widely used public-health manual in the world”. PanLex is taking a parallel approach to solving communication problems by developing the world’s largest freely available lexical translation database, with particular emphasis on under-served language communities.

The PanLex database currently contains over 2,500 dictionaries, 5,700 languages, 25 million words, 1.3 billion directly attested translations, and billions more inferrable translations. This database will form the basis of products and services to address the growing communications divide worldwide.

PanLex is now actively looking for solutions we can offer to tech companies, international organizations, and small-language communities. Our data can be customized for localization efforts into specific languages and locales, or can be organized by subject area. We are looking to maximize our reach and impact in overcoming language barriers.

In this talk, the PanLex team will discuss: 
• how we acquire and analyze lexical data 
• challenges faced in growing and using the database 
• ongoing outreach to potential partners 

David Kamholz is PanLex Project Director. He has a Ph.D. in linguistics from the University of California, Berkeley. He has done fieldwork on languages of Indonesian New Guinea, developed the PanLex API, and is leading the current outreach effort. 

Benjamin Yang is PanLex Director of Technology. He has a BA in linguistics from the University of California, San Diego. He has assimilated over a hundred dictionaries into PanLex and worked on graphical interfaces to the database. 

Julie Anderson is PanLex Director of Programs. She has an MA in linguistics from the University of Hawaii. She has acquired hundreds of dictionaries for PanLex and has experience in nonprofit management and development.