WordCorr for Comparative Ling (FWD)

[Doug Cooper]

Date: Mon, 10 Mar 2003 09:57:44 -1000 (HST)
From: Yuphaphann Hoonchamlong <yuphapha at hawaii.edu>

Dear sealang-ers:
I am forwarding some news about a new software for doing comparative
linguistics called "WordCorr". The following is a talk announcement at UH
Linguistics, but you should find the info about WordCorr useful even if
you can't attend the talk.

Date: 03/11
Speaker: Dr. Joseph E. Grimes, UH Linguistics
Talk title: People and Machines in Comparative Linguistics

*****Abstract*****
It's exciting to trace the history of language families beyond where
history itself can take us. But getting the evidence together takes an
awfully long time.

We're doing all the boring stuff by computer. With funds from the
National Science Foundation and programming by a professional software
house, we've made a tool called WordCorr that is based on some simple
observations: Linguists are good at recognizing possible cognates,
segment matchings, and relevant environments. Linguists are not good
at tabulating everything and finding it again. Computers are good at
the very things that drive linguists nuts.

So we've designed WordCorr as a partnership between you, the linguist,
and the machine. You do what you're good at, and the computer
organizes what you've observed in ways you can come back to. It
doesn't lose data. It works so fast that you spend most of your time
thinking, not chasing down file slips. It also lets you view your data
under more than one hypothesis.

Today's presentation walks you through the main ideas, shown via this
week's test version of WordCorr. It's an almost-ready standalone
version you can download yourself from
http://WordCorr.SourceForge.net. We will have the fully working
version ready for field use by the end of this semester. You are
invited to a WordCorr workshop the week of May 19 to 23, 10 to 12 AM
each day in the Linguistics conference room (YH Note: at UH), bring your
own laptop.

Next fall we shift gears to do the Internet version. That will make it
possible for research teams whose members are located anywhere in the
world where there is an Internet connection to work collegially on
data bases of any size.