Interview with Kara Warburton, PhD.
Author of "The Corporate Terminologist"

AET

1. HN (Henrik Nilsson, EAFT President): How is terminology work perceived today in the private sector? Has this view changed in recent years and, if so, in what way?


KW: The private sector lags behind the public sector when it comes to managing terminology. The public sector, i.e. governments and public institutions such as NGOs, started to develop termbases decades ago and they continue to operate the largest termbases in the world. The public sector has a mandate to serve society, and "managed" multilingual terminology supports that mandate.

When the private sector began using CAT (computer-assisted translation) software some years ago, the focus was on leveraging translation memories. After a number of years, however, some companies realized the limitations of their TM in handling sub-segment-level units. These companies started developing in-house termbases to fill that gap. Today, we see an increasing number of companies invest in terminology management. Unfortunately, though, that number is still too low. This, in spite of the reports that have been published and conference presentations held demonstrating the return on investment of managing terminology in a company. I attribute this low uptake to a lack of awareness among the decision makers in the upper echelons of a company. The organizational matrix structure so common in large enterprises makes it very difficult for "regular" employees to present innovative solutions to those who have the power to make investment decisions that benefit the company. That is the main obstacle in my opinion. Its a real shame, since by ignoring terminology as a form of data to be managed most companies are losing out on a great opportunity.

In recent years, I have observed that more companies are taking notice, especially when applications beyond translation are being deployed, such as controlled authoring, innovative content management solutions and search engine optimization (SEO).
 

2. HN: What do you think are the main challenges for the corporate terminologist today (e.g. compared to those of terminologists working in public authorities)?


KW: The corporate terminologist must focus on the needs of the company, whereas terminologists working in public organizations serve society at large. This driving purpose affects every aspect of the corporate terminology initiative, from deciding what "terms" to include in the termbase (the notion of "termhood" needs to be much broader than convention dictates), to designing the data model, and more. Everything the terminologist does is subject to return-on-investment scrutiny.

This means that you can throw much of the established theory and practice, which were largely based on the tradition of public-serving terminology work, out the window, at least, until it can be demonstrated that they serve the company's needs.

Due to the relative lack of training opportunities in this niche area, terminologists working in companies may not have a solid foundation in terminology management, and this presents more challenges. They end up putting a lot of faith -- too much I believe -- in existing so-called authoritative material, such as mainstream scholarly literature and even some international standards. This reliance on material that was not designed with commercial terminology management in mind can lead terminologists to make costly mistakes. Indeed, I have found that much of the established theory and practice of terminology as a discipline does not address modern corporate needs for linguistic data. That is why I wrote the book.

Since translation remains the primary use case for terminological resources, it is also common for companies to assign the job of terminologist to a translator. However, translators tend to be unaware of the applications where terminology data can be repurposed beyond translation itself. And repurposability across information systems has the greatest potential for commercial returns. Because of this repurposable nature, our understanding of even what a "term" is in this context needs to change. For today's language management needs, the notion, for example, that in order to be considered worthy of inclusion in a termbase a term candidate must be "domain-specific," is nonsense.So to help broaden this perspective, I like to use the term micro-content instead of terminology since the latter has so much historical baggage.

I therefore think the greatest challenge for corporate terminologists is actually two-fold: being able to recognize that they need NOT always follow established theory and practice, and developing a broader view of terminology as a form of repurposable micro-content.
 

3. HN: What special skills are needed for the corporate terminologist? What would be the best way to teach those skills – and which groups should take courses in terminology?


KW: The corporate terminologist needs to be highly skilled in data management. Terminology is comprised of lots of small, discrete, units of data.

This means that the terminologist should be computer savvy, and skilled in software tools for managing linguistic data, such as for term extraction, concordancing, file conversions, text manipulation (with regular expressions, xpath, etc.), as well as markup languages (XML) and their accompanying validation techniques. The terminologist should also be skilled in the tools where terminology data may be integrated: computer-assisted translation (CAT) tools, localization tools, controlled authoring (CA) tools, and SEO techniques and technologies, to mention a few. Terminology management in companies is a highly technical endeavour.

However, as I already mentioned, courses in terminology management are still lacking in many university programs that prepare  translators, technical writers, and other professional roles who should be involved in this field. Frankly when I see universities offering degrees in translation and there is not one course in terminology management -- and I do see them all too often -- I am shocked. And where courses do exist, all too often they are outdated, too focussed on redundant approaches, and lacking in technical aspects. It seems that most academic environments, like the historical foundations of this discipline, are out of touch with modern needs. Furthermore, once on the job, the terminologist may need to leverage terminological data in applications that were unforseen in the classroom. For all these reasons, corporate terminologists cannot count on academic institutions to teach them all the skills they will undoubtedly need. They should be able and prepared to teach themselves through research, tutorials, mentoring, and other non-conventional means. The corporate terminologist is an  autodidact.


4. HN: What do you think of the status of terminology work in private organizations today?


KW: I think I touched upon this question earlier. Let me just add that terminology work carried out in companies is still too focussed on the translation use case. The opportunities for leveraging terminology data for other purposes remain untapped.
 

5. HN: Why is this book especially important today? For whom?


KW: While there have been some previous ad-hoc discussions and papers about managing terminology in commercial settings, to my knowledge The Corporate Terminologist is the only book that is dedicated to this topic. I wrote it for anyone tasked with developing and managing terminology (i.e. micro-content) for a company, many of whom it should be noted would not be formally recognized with the title of "terminologist." I also wrote it for corporate executives who might "sponsor" the terminology initiative within the company. Repurposability and leverageability in modern digital applications should be key goals for corporate termbases. Prevailing theories and methodologies do not support those goals. This book proposes a new way to approach terminology management, one that does.
 

6. HN: In what way does The corporate terminologist differ from other terminology manuals and books on terminology?


KW: The book challenges some of the long-established precepts of terminology as a discipline -- precepts that are reflected, repeated, and engrained in most of the available body of literature in the field of terminology management. The conventional wisdom has been challenged before -- I am not the first to do so. But in this book I discuss those precepts from the perspective of a corporate terminologist. Fundamental questions that all corporate terminologists struggle with are answered. What works, and what doesn't work when managing micro-content in a company? How to get started, how to demonstrate ROI, how to get buy-in from the people who count, how to adopt the corpus-driven approach that is much needed today, and even, what is a "term" in this environment?
 

7. HN: The view on terminology theory and methods have changed during the years – how did you cope with this when writing your book?


KW: Well one could say that those views have NOT changed significantly over the years. That is largely the problem. Of course, there are a few notable exceptions -- scholars and practitioners who have diverged from the mainstream. Their viewpoints, which I discuss at length in the book, informed some of my own opinions.

I am, in fact, a classically trained terminologist of a certain advancing age. I completed an MA in Terminology at Université Laval (Quebec) in the early 1980's, where I was mentored by notable scholars and researchers of the Canadian school of terminology theory. It was long ago -- the computer was a large room and its output was reams of paper! This theoretical school of thought, if one can call it so, was largely modelled on the Wusterian General Theory of Terminology, with a slightly more corpus-based and lexical focus. We did terminology for terminology's sake. It was for a noble cause - promoting multilingualism and in particular, French in Canada. I was, therefore, an enthusiastic member of the community that advocates the conventions which, for modern commercial situations at least, I now challenge. Nearly thirdy years managing terminology in commercial settings have taught me what works and what doesn't. It was a difficult transformation, but one that I now embrace as a welcome change that opens new opportunities for terminologists.

Expressing my views in this book was nevertheless a bit discomforting because I know that some are controversial. I'm looking forward to engaging in further open discussion and debate that the book might trigger.
 

8. HN: You are an important part of ISO terminology work – does this come across in the book and if so, how?


KW: I have been a proud member of ISO TC37 for 20 years, holding various leadership positions including that of International Chair. This committee does exceptional work facilitating and promoting international standards not just for terminology, but also for translation, interpreting, and other parts of the language industry such as natural language processing technologies. At the same time I am quick to point out deficiencies in draft standards as they are being developed. One recurring theme of my comments is the tendency to repeat theories and methods of the past at the exclusion of more modern approaches that address commercial or other large-scale applications. In the book, I give recognition to TC37 standards that are useful for the corporate terminologist, but I also point out where others should be referred to "with a grain of salt."
 

9. HN: What kind of scientific research would this field need?


KW: Research in natural language processing. Terminology work of the future, and even of the present, needs to be heavily corpus-based. Research into how terms and other lexical units behave in corpora will benefit our profession. Term extraction immediately comes to mind; corporate terminologists need to master this process. I am very happy to see some of the more advanced term extraction tools leverage a reference corpus to aid in identifying domain-specific terms. But domain-specific terms are not the only lexical units that need to be "managed" nowadays. There are actually no guidelines about which lexical units a corporate terminologist should focus on -- and in fact this can vary from one organization to another. That missing piece of the puzzle is one reason why most term extraction tools are ineffective -- nobody really knows what they should extract. There needs to be a greater infusion of semantic, syntactic, morphological, statistical, and comparative knowledge in the tools that terminologists and other language workers use to develop and manage language resources. I am optimistic that research is already heading in this direction.