Alan K. Melby
Overview of Scholarly Research Interests
My research interests focus on translation and include both a theoretical
branch and an applied branch. The theoretical branch is very long
term and investigates the differences between various types of language,
based on factors such as fluidity and generality, and examines their
possible connections with translation and human agency. The applied
branch is more immediate and attempts to foster the development
and use of data exchange standards, in hopes of increasing interoperability
between various translation tools.
Although it is not obvious at first glance, there are connections
between the theoretical and applied branches of my research other
than the fact that they both involve translation. For example, my
theoretical studies suggest that because of fundamental differences
between various types of language, machine translation will neither
fade away nor replace all human translators. This, together with
the commercial fact that there are several competing vendors of
translation technology products, leads to the conclusion that data
exchange standards will be needed to allow various products to interoperate.
Alan Melby's curriculum vita is available.
One type of data exchange standard that is needed is a standard
format for exchanging terminology databases (often called termbases)
between various translation technology products. MARTIF (MAchine-Readable
Terminology Interchange Format) is a draft international standard.
I am co-editor of the MARTIF standard and, as a member of the United
States delegation to ISO Technical Committee 37, have been given
a mandate to test MARTIF on real-world data by writing software
to automatically convert termbases from one format to another using
MARTIF as a pivot point and keeping information loss to a minimum.
In the BYU Translation Research Group, we have done preliminary
testing and have made adjustments to MARTIF based on deficiencies
identified during testing. Now we have additional test data from
IBM, Logos, Star Corporation, and several other sources. What remains
in order to keep my committment to ISO is to do further testing
and to report on our findings at the next meeting of ISO Technical
Committee 37, which will be held in Copenhagen in August, 1997.