The stated purpose of the Kent conference was to evaluate how issues of language, multilingualism, translation, cultural difference, and language technology are affecting the world of international business. Beyond this goal, however, the organizers of the conference were aiming at bringing together an unusual mix of academics who train translators, localization companies who hire the graduates of translation programs, technical writers, and representatives of publishers.

Greg Shreve, founder and director of the IAL, characterized the conference as an attempt “to include the major segments of the language industry and get these groups talking to one another… Our major goals are to try to understand the language industry, who we are and what we do, and secondly, to try to find ways to manage our industry for maximum efficiency and effectiveness. There is no way we can understand the language industry, much less manage it, unless all of the major elements of the industry meet together and have a really serious conversation.”

The talent deficit

The language industry is one of the ten fastest-growing business sectors in the world today. With growth like this, it's not easy for companies to find the skilled personnel they need. Chris Langewis, who teaches computer applications for translation students at the Monterey Institute of International Studies, notes that “the combined output of the translation schools in the United States is somewhere around 250 students in any given year, and not all of them are trained to meet the technical challenges posed by industry.”

European universities turn out higher numbers of translators per se, but the technical deficits are for the most part just as great. Brigham Young University’s Arle Lommel cites the following figures from the DICE listing of high-tech job openings in the computer industry: there are 1,896 jobs listed under key words the internationalization or globalization, but the LISA Education Initiative Taskforce (LEIT) has only identified three courses in the world that are specifically devoted to internationalization. DICE lists 1,925 jobs under localization. Many of these jobs are for localization engineers, but others are for translators and project managers. Obviously the US translation schools are not turning out this many graduates in any given year. Lommel’s conclusion: “If DICE.com is a reliable indicator, demand for localization skills is high, but university-level programs are few.” This view conforms to LEIT’s findings in its recently conducted surveys.

The initial mandate of LEIT was to survey academic and non-academic programs that offer courseware and training for internationalizers and localizers, and to query market players to determine their needs with respect to major job profiles. The results of these surveys are accessible online in the form of user-friendly matrices (click on the “What is LEIT?” button at the LISA Web site at www.lisa.unige.ch). Interested parties are urged to participate in LEIT’s ongoing efforts to monitor the training needs of the industry by filling in the questionnaires available at that site.

LISA’s Alison Rowles noted that “The limited pool of qualified candidates for jobs in the localization sector has resulted in opportunistic, even predatory hiring practices that are hurting the industry. The LISA Code of Good Conduct and the Web site job postings program are designed to help address this situation. Coupled with the LEIT matrix, they are all designed to encourage collegiality and ready access to information on available positions as well as sources for “new blood” entering the market.”

LEIT is not the only group addressing training issues in the language-engineering sector. The Language Engineering for Translator Curricula (LETRAC) group is funded by the European Commission’s DG XIII, within the Telematics Application Programme of the Fourth Framework. The group consists of the translation departments of six European universities, all of which are also members of CIUTI, the Conférence Internationale Permanente d’Instituts Universitaires de Traducteurs et Interprètes. The fundamental difference between LEIT and the current LETRAC focus is that LEIT is worldwide and is looking at conditions for training both internationalizers and localizers, as well as business specialists involved in globalization. LETRAC is European and has limited its scope to training translators in language engineering applications.

LETRAC’s Ursula Reuther notes, “The profession of translator will undergo drastic changes in the near future. There is a crying need to develop and to integrate IT/LE elements into the training of translators.” Focusing specifically on translator training in Europe, LETRAC has also conducted surveys of trainers, students, and industry representatives in order to define professional profiles and to propose non-binding revisions to existing curricula. Reuther participated in the Kent conference and is committed to cooperation between LEIT and LETRAC. LEIT principals Susan Armstrong (University of Geneva/ISSCO) and Sue Ellen Wright (KSU/IAL) serve as advisors for LETRAC.

In the original LEIT timeline, the Kent Conference was scheduled as the kickoff point for Phase II of the project. In response to this obligation, LEIT researcher Arle Lommel proposed three options for further action:

Level 1

Maintain the matrices by monitoring academic and training programs in the industry.

In their current form, the matrices represent a valuable resource, but the information will deteriorate quickly if no effort is made to continually track developments in both the industry and in academia.

Level 2

Level 1 + model curricula, increased support from tools vendors, and “train-the-trainer” workshops.

This level of activity would involve LEIT members collaborating to agree upon a set of requirements designed to fill the needs of specified job profiles. The cost of this level would be approximately $20,000 per year.

Level 3

Levels 1 + 2 + develop full courseware.

The notion behind Level 3 was to create a “black box” package that would enable a knowledgeable trainer to use “canned” course preparations, lectures, exercises, etc. and to teach the course with relatively little extra preparation. This option would cost upwards of $250,000 and would result in a LISA-owned full courseware resource.

In a lively ad hoc meeting, educators and trainers from both academia and industry brainstormed a rough modus operandi for cooperation among LETRAC, LEIT and other groups involved in charting the course for localization and internationalization training in the next century. A strong link already exists between LEIT and LETRAC in that some LEIT members, together with other LISA representatives, serve as advisors to the LETRAC group. SLIG’s Reinhard Schäler is also a member of both LETRAC and LEIT.

New guidelines for projects funded under the EU Fifth Framework encourage the cooperation of EU and US researchers working on common projects, which suggests the feasibility of launching a joint project to create curricula and courseware modules designed to prepare students to fill the needs of the marketplace. The strong consensus coming out of the Kent meeting was that the various groups represented could work together to define specific skills areas required by industry and perhaps to develop individual courseware modules.

Nevertheless, it was widely felt that the creation of full-blown course materials (LEIT option Level 3) would be counter-productive given the great variety of programs and the different needs and resources available in different regions and venues. Further consensus favored a kind of “self-certification” on the part of programs that decide to conform to curricular guidelines as opposed to either individual or program-oriented certification, which could prove both costly and controversial.

Everyone is eager, however, to watch SLIG efforts with regard to the Certified Localization Professional (CLP) program currently under development in Ireland. (http://lrc.ucd.ie/SLIG/1998/SLIG98/clp.html)

Future synergies

As noted above, the original purpose of the Kent conference was to bring people together from all over the industry to establish a dialog that would reveal where training programs should go in their curricular development in the next five years. The organizers are relatively satisfied that this interchange has provided them with a springboard for action directed at several specific goals. The potential synergy that can be derived from the interaction of industrial partners and academic resources is essential at all levels of the process.

Not only do trainers need constant feedback and guidance from the commercial sector, they also need the support of software vendors to provide the tools to train the next generation of localization specialists. Reflecting on the inability of educational institutions to pay for these tools, Monterey’s Chris Langewis declares, “We have to convince software vendors of the benefits of providing training programs with the tools they need to prepare students. To have these “savvy” translators exposed to their products is of great value to the software producer, hence they have an incentive to supply goods under favorable conditions. Otherwise academic programs can't afford to do the job.”

Langewis’ position is echoed by Susan Armstrong’s comments in a recent LETRAC meeting: “Both IT and traditional teaching is necessary for training the next generation of internationalizers and localizers, but there is a lack of teaching material and equipment too. Software providers should be more supportive because universities need more free licenses to train their students adequately.”

LISA member tool providers will no doubt recall Armstrong and Wright at the Madrid meeting diligently soliciting their support for providing the software they need to do their jobs. Some software companies (e.g. Logos and TRADOS) have responded creatively and willingly, offering campus visits along with special campus-wide licensing agreements or beta-site arrangements.

Conference participants

The Kent conference was originally planned as a showcase for overall language engineering activities, but grew strongly in the localization direction after the initiation of the LEIT project. Consequently, the participant list for the conference strongly reflected LISA interests. Representatives from major publishers were there (Microsoft, IBM, Diebold, and Lucent Technologies), from tool providers (Logos and TRADOS), from prominent service providers (LMI, L10Nbridge, Translingua, Berlitz, Alpnet and Harvard Translations), as well as local freelance translators.

On the training side, university teachers were well represented, with speakers and participants coming from the KSU/IAL, the Monterey Institute for International Studies (MIIS), as well as universities in Binghamton, Iowa City, Vienna, Surrey, Cologne, and Saarbrücken. Industry trainers were present from Berlitz and Alpnet, along with industry analysts from Equipe Consortium, LISA, the LEIT initiative, and LETRAC (the Institute for Applied Information Sciences in Saarbrücken).

The industry press was represented by Language International’s Robert Sprung and Kristen Padden, and there was a steady flow of KSU/IAL students, plus students from the Universities of Binghamton, New York and Geneva, Switzerland.

Translation is a process, not an event

LISA’s Alison Rowles set the tone for a major facet of the conference by quoting Logos CEO Jens Thomas Lueck, “Trans-lation is a process, not an event” a concept that was echoed by LMI’s Tracey Feick with “Quality is a journey, not a destination.” (David Carr). Whether the discussion was project management, translation, localization, technical writing, or overall quality assurance, the importance of process and controlling it held a foreground position throughout the presentations.

The internationalization/document generation phase of the process was well covered by Robert Dianetti of RadCom, Inc., an Akron, Ohio-based company specializing in technical communications, and Diebold’s Corinne Moore, who described the efforts of her work group to introduce Controlled English: “Of course, the goal is to reduce costs, speed up turnaround times, and increase control of the process. CE reduces choices, but it also reduces ambiguity and the chance for error.”

Integrated systems and return on investment

Wright’s focus on integrated systems was reinforced by TRADOS’ Henri Broekmate: “In the future TRADOS will offer solutions that provide enterprise-wide applications for multilingual information creation and dissemination, integrating logistical and language engineering applications into a smooth workflow that spans the globe.” This position has its counterpart trend in Logos’ integrated technology-based translation solution, which combines terminology management, translation memory (TM), machine translation (MT), and related tools to create a seamless full-service localization environment.

Logos’ Scott Bennett addressed financial concerns: “Machine translation offers real return on investment, … but with a major investment. Machine translation is never plug-and-play. Significant customer support from the vendor is essential to achieve the Return-On-Investment (ROI) potential offered by the system.”

ILE’s Clove Lynch takes the discussion of ROI a step further: “Tool use in localization is on the rise and projected to keep rising. However, at least one study has shown that while translation tools are expected to produce better quality deliverables, they do not always reduce production costs for suppliers or their clients. This discrepancy must be reconciled, or at the very least quantified.” Lynch reviewed analytical and testing procedures designed to support a system of process control metrics aimed at quantifying ROI values associated with different tools and procedures.

Speaking on behalf of L10NBridge’s Rory Cowan, Katrina Teague analyzed the evolution of a natural business model and projected this analysis onto the localization industry. “Technology,” writes Cowan, “determines business models; business models determine industry structure, and the industry structure determines profitability.” The next few years will be critical for the localization industry: “Industries have strategic inflection points. Localization and translation are entering such a 5-7 year phase.”

Terminology

The importance of accurate, consistent terminology at all levels of the document production cycle, from product development to the technical documentation stage to localization and translation was a common theme throughout the conference. Margaret Rogers from the University of Surrey at Guildford stressed the significance of dynamic terminology in contextualized discourse. She described the generation of subject-specific text corpora using a sophisticated Internet search tool called SEBRIBA. Rogers’ text-oriented work in terminology points in the same direction as a prophetic (or at least provocative) comment from Equipe’s Rose Lockwood, who visualizes today’s translation memory systems yielding ground significantly as we see further movement in the direction of structured markup of texts.

Regardless of the environment in which terminology is gathered or documented, the quality of final documentation and the effectiveness of tools such as TM and MT depend to a great extent on the reliability of terminological resources. Vienna University’s Gerhard Budin provided an overview of efforts that are underway in ISO to develop a standard for terminology work groups active both inside and outside the standards movement.

A further area of language engineering where terminology plays a significant role is information management and retrieval on the World Wide Web. Marsha Lei Zeng of Kent’s School of Library Science and Information Science introduced her audience to a range of search strategies used by the various major search engines to access information in multilingual environments, particularly with respect to situations involving non-European character sets and the emerging UNICODE standard.

Plan for action

Since specific funding for any of the proposed Phase II levels is uncertain, the immediate plan of action for the participants in the Kent LEIT meeting is to begin to exchange information on curriculum and courseware planning. There will be a LEIT listserve in which any interested parties can participate.

Next fall’s LISA Forum in Monterey will offer a “train the trainers” workshop to help bring more skilled trainers on board. In the meantime, in America, Monterey will continue to teach project management and localization workshops. Kent will launch its project management and localization courses in fall 1999, drawing heavily on collaboration with industry partners to ensure the validity of approaches taken to solve real-world problems. For further information on the conference, check: http://appling.kent.edu/resources.htm.

---

Prof. Sue Ellen Wright
Kent State University
Institute for Applied Linguistics
Kent, OH 44242
USA
Tel +1 330 673 0043
Fax +1 330 673 0738
E-mail: swright@kent.edu