So where have all these efforts led us? As chair of the LISA Terminology SIG, let me make a bold claim:

Our efforts have not led us nearly far enough.

The language services industry is pretty much all talk and no action when it comes to really managing terminology within the localization process. Let's look at the facts.

We don’t use TBX, and our tools still don’t support it.

Thanks to OSCAR, we have a solid terminology markup standard, TBX, but we don't use it, and our tools still don't support it. We've presented our wish list of terminology functions to localization tools providers, and some of them have listened. But whether for reasons of lack of functionality, or lack of resources or personal initiative, most of us haven't made the big technology leap, and we're still using spreadsheets. That's just one step better than my grandmother's recipe cards. And such primitive approaches virtually guarantee that the so-called terminology that we’re supposedly managing for a translation project cannot be leveraged anywhere else.

Terminology is the biggest factor in poor translation quality.

We admit that terminology is the biggest factor in poor translation quality, yet we still don't pay for doing up-front terminology work for translation projects. (Editor’s Note: See the latest Translation Memory Survey.) Small and mid-sized localization companies rarely employ terminologists, and large ones reluctantly employ one or two, while governments of countries that you couldn't find on a map have fifty. We're so busy counting words and bragging about the exponential growth of our translation memories that we're sacrificing quality and ignoring the opportunities of a leading-edge information discipline.

The ontology and semantic web stakeholders will not wait for us.

The semantic web is taking shape, and its thought leaders are already developing terminology-related standards and processes (such as OWL and SKOS). Terminology is the DNA of knowledge, so any knowledge management activity, from ontology development to search engine tuning (or search engineering), is impacted. While TBX waits to be implemented, ISO TC 37 is churning out new standards for the extended language industry faster than you can say, "Bob's your uncle."

Don't kid yourself: These stakeholders will not wait for us. If the language services industry wants to hang on to its tenuous reputation as a leader in the field of terminology, we're going to have to start putting our words into action. That's precisely what the members of the LISA Terminology SIG have done.

We have proven that conversion to TBX can be virtually instantaneous and painless.

After spending several years developing best practices and promoting the virtues of TBX, we have decided to lead by example. With the help of Alan Melby from OSCAR and Marc Carmen, one of Dr. Melby's students from Brigham Young University, we converted samples of our own company terminology data from proprietary formats to TBX. And we're not talking about just any proprietary format; these are major terminology databases belonging to IBM, Medtronic, Oracle and SDL. We've shown that, through the use of automated routines, conversion to TBX is virtually instantaneous and painless.

The lack of standardized terminology formats and management procedures is possibly the greatest weakness in the localization industry today.

Why would we spend the time and effort doing this? Because, as experienced localization terminologists, we know that the lack of standardized terminology formats and management procedures is possibly the greatest weakness in the localization industry today. And we're tired of just talking about it.

You can link to the results of this ongoing project from the SIG Web page at http://www.lisa.org/term. While you're there, show your support by contributing a few terms and definitions to the Glossary Project, which is led by our new member, Lisa del Papa from Intentia. And, if you'd like to join the LISA Terminology SIG, send me an email at kara@ca.ibm.com.