Frasciello

Michael J. Frasciello
Doctoral Student
Composition and Cultural Rhetoric
Syracuse University
E-mail: mfrascie@uc.syr.edu

Current popular localization practices of online courses fail to consider appropriate contextual elements, such as culture and language variety. These practices are inadequate because they address only structural components of the document. This narrow scope is derived from a limited understanding (on the part of the course authors, instructional designers, and course translators) of how the target audience will interpret the course text. Localized online courses that consider only structural criteria fail to address the more critical criteria of effective communication and learning, such as the rhetorical perspectives and cultural complexities of the target reader/learner. Without a broader understanding of the reader/learner and deeper treatment of these contextual criteria, localized course content is treated as a distinct and separate component of the online course. This paper explores the case study of an online technical training course to illustrate the consequences of localization practices that fail to consider contextual structures at all phases of the online course production process.

Localization is the act of modifying an information product to make the product usable for a specific target audience. Localization affects an online course on a structural level and a contextual level. Structural level localization is the process of translating and customizing the online course’s features, such as color schemes, screen layouts, graphics, etc., for a target audience. Contextual localization is the process of modifying the aesthetic appeal, logic, functionality, and information architecture of the course to conform to a target audience’s cultural complexities (Lingo Systems & LISA, 2000). The goal of the instruction is compromised when the localization activity considers only the structural level criteria of the online course.

Online courses are global products. As such, the instructional designers working on these courses need to consider the role of culture in how the courses will be used by learners in different linguistic communities, and how localization addresses cultural contexts. Effective localization is more than simple modifications to structural level components of an online course. Yet, increasing numbers of international companies specializing in technical information for global audiences are localizing their online training courses after production is complete; ignoring contextual level criteria (Raley, 2005). To create effective online course materials, instructional designers must be aware of the possible problems created by not considering contextual level criteria at the beginning of the writing process. Each problem introduced during the early stages of course development creates additional problems for targeted and non-targeted users of the course. This phenomenon is illustrated in the following case study of a translation project undertaken by Carrier Corporation.

Carrier contracted with Syrtis, a technical training development unit at Syracuse University, to produce an online course to prepare heating, ventilation and air-conditioning technicians (HVAC) to take the NATE certification exam. The NATE certification allows technicians to bid and work on certain types of HVAC projects and systems.

The course was originally produced for an (American) English-speaking audience. The needs analysis conducted by Syrtis during the instructional design phase of the project revealed that the target audience had, on average, a sixth grade reading level. The instructional designers working on the project, therefore, authored the content with the appropriate level of language complexity and sentence structures. Figure 1 below illustrates the typical style in which textual content was presented at the structural level throughout the course:

The course page captured in Figure 1 contained the following English language text:

After a one year deployment, Carrier decided to offer the NATE online course to Spanish-speaking technicians working in the United States, primarily of Mexican and Central American linguistic backgrounds. Syrtis estimated a cost for translating the course and proposed a localization plan to properly prepare the content for the target audiences. Carrier determined that the project costs were prohibitive and opted instead to use a machine translation service (and forgo localization) to convert the course for the Spanish-language target audiences.

Immediately after deploying the machine-translated version of the course, Carrier began receiving negative feedback from the target audiences. The most relevant feedback was that much of the text in the course was nonsensical—grammatically and structurally incorrect. Figure 2 below and the subsequent text illustrates the typical manner in which the meaning and instructional integrity of the course content (previously illustrated in Figure 1) was compromised by the machine-translation. The text in brackets is an English retranslation provided by the Spanish-speaking student technicians in the target audience (a translation of how the text was read). The italicized text indicates problematic terms or phrases.

Figure 2: Machine-translated content (International Spanish language version)

Faced with the failure of the machine-translated course to meet the instructional requirements of the target audience, Carrier decided to have the course translated a second time by a Spanish-speaking Carrier employee of Argentine-American decent. Prior to the second translation project, Syrtis reiterated the necessity to localize the content for the target audience, which did not include Spanish-speaking HVAC technicians with Argentine linguistic backgrounds. Carrier rebuffed the identified requirement (again based on cost) and completed the second translation without any formal structural- or contextual-level localization. Figure 3 and the accompanying text illustrate the results of the second translation project. The text in brackets is the text as previously rendered by the machine translation. The italicized text indicates problematic terms or phrases.

While the newly translated course proved more successful at transferring knowledge to the target audience, a usability analysis of the course (conducted by Syrtis) revealed that language-related obstacles to learning remained at the structural and contextual level of the content. Student technicians interviewed during the usability analysis indicated that many of the industry- and field-specific terms and phrases used in the course were incorrectly Spanishized—indicating that some degree of borrowed English was acceptable (if not necessary) for a more complete understanding of the course text on the part of the student technicians. The presence of mistranslated terms and phrases also illustrates how the prevalence of borrowed words in specific linguistic, geographic, and cultural contexts can be overlooked when localization is not performed. Table 1 below contains terms that were identified by the target audience (Spanish-speaking technicians primarily of Mexican and Central American linguistic backgrounds) as commonly used and understood technical terms rendered in English. It is important to note that all of the student technicians had backgrounds in heating/air conditioning or electrical engineering. The Argentine-American translator employed by Carrier did not have a technical background:

English Term Used by Spanish-language HVAC Technicians	Mistranslation by Argentine translator
Anode	Anodo
Dielectric	Dieléctrico
Diode	Diodo
Hertz	Hercio
Ohms	Ohmios
Red Pole	Polo rojo
Filament	Filamento
Hysteresis	Histéresis
Toroid	Toroido

Additional usability analysis (which focused on localization-related issues) revealed that the target audience spent a measurable amount of time (approximately 2 additional minutes per page) attempting to adapt the course text to their specific linguistic and work contexts. Localizing the course at the contextual level would have corrected the translator’s tendency to cast sentences using rhetorically inactive constructions (Bizzell, 1990). In discussions with the target audience, it was determined that some student technicians were internally recasting these sentences into more preferable action statements and condition statements—statements that are more common and acceptable in their linguistic communities. The following text illustrates the types of action and condition statements the student technicians were recasting from the course text:

The student technicians also indicated that active voice statements (in which the subject performs the action expressed in the verb) were preferable to the passive voice statements (in which the subject receives the action expressed in the verb), which were used excessively by the translator. The following text illustrates the types of passive statements in the course text the student technicians were recasting:

Further interviews with the student technicians also revealed that simple sentences of direct object patterns (subject + action verb + direct object) were preferable to the more linguistically problematic sentences cast by the translator.

Conclusions about the absence of localization from the Carrier online course need to be carefully considered because the usability analysis conducted by Syrtis was limited in a number of ways and involved a relatively small number of student technicians (15 total students; six in the initial phase of the analysis). However, the feedback of all student technicians observed and interviewed during the analysis confirmed the legitimacy for localization early in the course development process. The analysis also identified the different linguistic styles (from high- to low-context) that can reside within a broad linguistic category, such as Spanish. A more important implication of the case study is that the situation regarding context in cultures is too complex to make anecdotal generalizations. As the need to teach localization in Instructional Design, Development and Evaluation (IDDE) grows, so does the need to investigate the complexities of communicating information across cultures (Barber, 1998). The notion of high- and low-context within linguistic communities, although useful at a macro level, becomes problematic when applied to a single instance such as the Carrier online course. More appropriate for the purposes of this study, all usability interviews with the student technicians revealed a pragmatic, common-sense approach to the technical subject matter of the course. The remaining sections of this study attempt to capture such common-sense considerations of cultural context and structured localization processes in a formal call to action for instructional designers.

The Carrier online course case study illustrates a small set of the intelligibility obstacles created when online courses are not localized. More importantly, failure to consider localization at the beginning of an online course project masks the communication and instructional strategies embedded by the cultural perspectives of the course authors, developers, and instructional designers. In the case of the Carrier online course, none of the communication strategies implemented by the instructional designers were shared by either the Argentine translator or the Spanish-speaking student technicians. A more appropriate approach for the project would have been to consider the cultures of all potential users of the course at a structural and contextual level before producing the course.

As implied earlier, culture embedded in online courses is visible at the structural level of the course text, particularly in the choice of words and use of specific nomenclatures (Whitburn, 2001). To prepare for localization, instructional designers should first consider the communication event (from the structural perspective) that will occur between the course text and the reader/learner. Such a consideration reveals problematic terms that carry the same (or nearly the same) meaning in the language of the target audience; words with different associative fields, terminology, acronyms, colloquial expressions, and words with multiple meanings.
Similarly (and as revealed by the Carrier case study), instructional designers also need to be aware of the culture embedded at the contextual level of their online course texts; the constructions that affect the way in which readers use the text, what readers assume about textual symbols, and what linguistic and rhetorical features readers recognize. If not considered early in the project, such features limit the way concepts are understood or applied by reader/learners. Attitudes towards stylistic issues such as the length and complexities of sentences also reside at this level of the text. Because stylistic patterns differ from one linguistic community to another, instructional designers need to have insight into cultural differences as well as stylistic differences; helping them to write better in all contexts (Andrews, 1997).

Instructional designers need to understand the process of contextual level localization to anticipate the obstacles of ill-conceived translation (Esselink, 2000). As in the case of the Carrier online course, if the designers had been aware of a possible future translation, they could have worked with culturally neutral prose to prepare for localization. Culturally neutral prose would allow localizers to more readily create a text that is suitable for a broad range of readers (those who require declarative information and those who require only procedural information) (Hoft, 1995). Depending on the culture of the target audience, the instructional designers working on the Carrier course could have worked within a range of assumptions about how the course would be used, what the student technicians would want to know, how much assistance they would need, and whether they would need to use the course at all to pass their Spanish-language NATE certification exam.

The complex nature of culture (and how it is manifested in online course material) makes it impossible for instructional designers to address the cultural needs of all potential alternate-language reader/learners. However, the communication and instructional events can still be successful (as in the case of the final translated version of the Carrier online course) with even minimal knowledge of the cultural and linguistic backgrounds of potential reader/learners. Even though the Carrier course was not localized to the target culture and the prose was not culturally neutral, the student technicians were able to make the necessary linguistic adjustments to understand the message.

IDDE and localization are, by some measures, short-sighted and ill-conceived in regard to cultural contexts. Cultural contexts need to be seriously considered in localization practices to accommodate a wide range of audiences. Online courses are always situated in the cultural contexts of where they are designed, produced, tested, distributed, and used. Cultural factors cannot be extracted from communication and instruction, nor can they be removed from the localization process (Draga, 2001 and Bosley, 2001). Cultural factors behind communication and instructional conventions include common knowledge shared within a culture, the hierarchical structure of society and workplace, culturally specific rhetorical strategies, and cultural differences in processing information (Bizzell, 1990). In regard to textual communication and learning events in local contexts, translation, punctuation, and aesthetic appeal are representative of the common knowledge and values shared within a culture.

Internationalization and increased relevance of Global Englishes are changing the nature of local cultures. In the age of globalization, culture is a dynamic process in which meaning and identity flows across institutions and nations. But as a discipline and practice, IDDE fails to understand culture as a dynamic phenomenon locally and globally; preferring to work within a static model of culture (Draga, 2001 and Raley, 2005). This model neglects the dynamics among cultures that demand different perspectives and rhetorical strategies. As illustrated in the Carrier online course case study, this static model situates the instructional designer in a singular dominant culture with one broad linguistic community (Spanish), and assumes the culture will be dominant for all audiences (variants of Spanish). Within the static model, complex cultural knowledge is reduced to a set of generic cultural paradigms (Esselink, 2000 and Gribbons, 1997).

If asked to define culture in context, the writers, designers, and developers working on the Carrier online course case study most likely would have taken a narrow view of culture as a simple means of distinguishing among the different populations. And yet, even if the group had knowledge of future translation and location efforts (and knowledge of alternate language target audiences), they could not cast the course text using some set of generic markers that represent the diverse cultures in the world. No such markers or cultural formulas exist.

Culture is excluded from localization when the authoring process is removed from localization practices. A more realistic technique for instructional designers is to employ localization practices early in the content collection and authoring processes to capture the dynamic contextual knowledge that is often lost when localization is performed as an afterthought. Approaching culture as a dynamic phenomenon allows the instructional designer to maintain the goals of the communication and learning events through multiple language versions of the course. The designers are able to consider culture throughout the entire content authoring process, closely linking cultural contexts to content and localization. Thus, localization is no longer a separate activity from the authoring process. Likewise, the localization process is no longer considered a simple translation process or an activity that only addresses the deployment of the online course.

Useful localization processes regard culture as a dynamic phenomenon in which different linguistic ideologies exist. Instructional designers working through localization processes early in the project life-cycle do more than simply adjust existing information products on the structural level (Hoft, 1995). A completely integrated localization process exposes the course designers and developers to the most appropriate structural and contextual level communication strategies. The delivery of the online course is therefore not separated from the content, and the content is not separated from its situated context. The result is a broader-scoped rhetorical method that treats the course authoring process as a functional heuristic for localization (Kano, 1995 and Keniston, 2005).

Successful localization requires instructional designers to have a complete understanding of the structural and contextual level criteria of their online courses. Localization needs to be introduced early as an integrated activity of the course authoring process. The entire design and development team needs to use dynamic rhetorical and linguistic methods to think of communication and learning in terms that create a complete, integrated localization process—one that requires course authors and instructional designers to explore the best way to present information and to identify the most appropriate content for specific cultural and rhetorical contexts.

Andrews, D. C., and W. D. Andrews. (1997) Business communications (3rd. ed.). Needham Heights MA: Simon & Schuster.

Barber, W., and A. Badre. (1998). “Culturability: The Merging of Culture and Usability.” Proceedings of the Fourth Conference on Human Factors and the Web, Basking Ridge, New Jersey, June 1998.

Bizzell, P., and B. Herzberg (1990). The rhetorical tradition: Readings from classical time to the present. New York: St. Martins.

Bosley, B. S. (ed.). (2001). Global Contexts: Case Studies in International Technical Communication. MA: Allyn & Bacon.

Dragga, S. (2001). A Visit to the Forbidden City: A Sign of the Times. In Bosley, B. S. (Ed.), Global Contexts: Case Studies in International Technical Communication. (p.10-18). MA: Allyn & Bacon.

Esselink, B. (2000). A practical guide to localization. PA: John Benjamins Pub Co.

Gribbons, W. M. (1997). “Designing for the Global Community.” Proceedings of IEEE International Professional Communication Conference, Salt Lake City, UT, 1997, 261-273.

Hoft, N. L. (1995). International Technical Communication: How to Export Information About High Technology. NY: John Wiley & Sons.

Kano, N. (1995). Developing International Software for Windows 95 and Windows NT: A Handbook for International Software Design. Redmond, WA: Microsoft Press.

Keniston, Kenneth. Software localization, Notes on Technology & Culture. Accessed December 2005: http://www.mit.edu:8001/people/kken/keniston.htm

Lingo Systems & LISA. (2000). The guide to translation and localization: Preparing products for the global marketplace. Portland, OR: Lingo Systems.

Miller, C.R. (1979). “A humanistic rational for technical writing.” College English, 40, 610-617.

Raley, Rita. What Is Global English? Accessed December 2005: http://www.tc.umn.edu/~raley/research/global-English.html

Trillo, Néstor G. International Communication. Accessed December 2005: http://www2.soc.hawaii.edu/css/dept/com/resources/Intercultural/Intercultural.html

Whitburn, M. D. (2000). Rhetorical Scope and Performance: The Example of Technical Communication. CT: Ablex Publishing.

You are invited to join AE Extra staff!
Send your ideas and/or writing sample to the Editor-in-chief:
Elizabeth Haller
Kent State University (e-mail: editoraee@hotmail.com)

Academic Exchange Extra invites reader response to any writings in this issue--especially articles advancing the scholarly debate of issues raised.

Localization Practices for Online Courses: A Case Study for Contextual Considerations

Localization Practices for Online Courses:
A Case Study for Contextual Considerations