The Internet – when defined as a massive, uncontrolled database of information – is not a reliable source of information. Despite this, it has developed into the default terminology tool for most translators today. This study explores what criteria are used by translators to define website credibility and whether translation experience affects the criteria used. Three groups of differently experienced translators were asked to translate two technical texts from English to Swedish and document the websites they visited. The documentation was used as the basis for follow-up interviews, where the translators were asked to argue for their choices. The results show a significant similarity in application of criteria between student and experienced translators, indicating that other factors than experience are at play. Moreover, source text complexity does not affect the criteria used but rather to what extent research is performed. These results can be used to encourage further research into the information search behavior of translators.

Philosophiae Doctor - PhD / Working with Arab students pursuing a degree in English Language and Translation at the Taiz University, Republic of Yemen, has brought to the researcher‟s attention a number of errors or problems encountered in Arabic to English translation. This study aims to investigate the problems encountered by student translators (STs), novice translators (NTs) as well as more experienced translators (Ts) while translating from Arabic into English. The study starts with the assumption that Arabic and English belong to different families of languages and thus there is rarely a word-for-word equivalence in both languages. The present study is cross-sectional in nature. It is based on empirical data collected from several categories of translators. In other words, the data was collected from fourth-year students in the department of English and Translation in the Faculty of Arts, Taiz University, as well as five NTs who have previously graduated from this department and are currently working in a number of accredited translation offices in Taiz. The study also investigates the challenges faced by Ts. For this purpose, a novel, a tourist brochure, an editorial, and three academic abstracts all translated by established publishing houses and translation centres in and outside Yemen are examined. These texts are analyzed to determine to what extent the problems faced by STs and NTs reoccur in published translations produced by Ts. For its conceptual framework, the study adopts an eclectic approach that does not stick rigidly to a particular paradigm but rather draws upon multiple linguistic and translation theories. However, it is mainly based on Halliday‟s Systemic Functional Grammar (SFG) and the problems have been classified along his taxonomy of meaning metafunctions into ideational, interpersonal and textual. Extra-textual problems are also analyzed. Several SFG-based translation models such as Hatim and Mason‟s (1990) sociometic model, House‟s (1977, 1997) translation quality assessment model, Hervey et al. (1992) register analysis model and Baker‟s (1990) equivalence model are also employed in the study to help the researcher examine the problems encountered in Arabic-English translation within those four categories. In addition, Nord‟s functional model to translation which is based on Skopos theory is also taken into consideration although to a minimum extent. In addition to the analysis of translations produced by various categories of translators, the study uses several triangulation research tools such as questionnaire, Thinking Aloud Protocols (TAPs), retrospective interviews, and classroom observation. These tools are employed to assist the researcher to identify the possible causes for the problems the STs, NTs, and Ts experience from the perspective of the participants themselves. The current translation programme at Taiz University is also analyzed to determine to what extent it contributes to the poor performance of the student translators and would-be translators. The study concludes that STs, NTs and even Ts encounter several problems at the ideational, interpersonal and textual levels. They also encounter problems at the extra-textual stratum. The study attributes these problems to structural and cultural differences between the two languages, the reliance on the dictionary rather than the meaning in use of lexical items, the differences in the cohesion and coherence systems of Arabic and English, the negligence of the role of context in translation as well as unfamiliarity with text-typologies and genre conventions. In other words, participants follow a bottom-up approach in translation and come close to the source text translating it literally. This approach is very damaging because it ignores the fact that the three metafunctions might be realized differently in the two languages. Furthermore, the study concludes that the manner in which translation is taught at Taiz University as well as the syllabus contribute mainly to the lack of translation competence of the student translators and would-be translators. The programme is inadequate and it needs urgent review and improvements. The present syllabus does not keep abreast with the latest theoretical and practical developments in the discipline of translation as well as neighbouring disciplines such as contrastive linguistics, text-analysis, discourse analysis, corpus linguistics and the like. As for methodology, the study concludes that it is the transmissionist (teacher-centred) teaching approach rather than the transformational (learner-centred) which is commonly used in teaching translation. As a result, the read-and-translate approach dominates the scene and no tasks, activities, or projects are given to the STs. The study provides some recommendations, which if implemented, can be useful in enabling Yemeni and Arab universities to improve the competence among student translators in order to improve translation teaching at academic level. A major contribution of this study is the description and classification of translation problems in Arabic-English translation on the basis of meaning systems. Unlike traditional descriptive error analysis, which is widely used to analyze the translation product, SFG-based text analysis provides a systematic description of translation problems which allows a precise articulation of the nature of problems that would otherwise be explained simply as translations which “sound unnatural or awkward” (Kim 2008; Yallop 1999). As far as the researcher knows, no study in the Arab world has yet tackled translation problems from this perspective. Other studies have tackled deviated forms produced by students or translators using an error analysis technique rather than a holistic approach based on solid theoretical knowledge. In other words, while most other studies focused on specific „errors‟ and error analysis and ended at that, the present study does not only looks at „errors‟ as „difference‟ (from contrastive analysis) but rather from several perspectives. It is also more comprehensive by triangulating several sources of data and pooling them together for a more informed understanding.