No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

CD-ROM dictionaries are dictionaries on a compact disk read-only medium, which basically contain the same information as their paper counterparts. During the last ten years, a significant number of CD-ROM dictionaries have appeared on the market. Given that the content of CD-ROM dictionaries is almost identical to that of paper dictionaries, in what way do they differ from their traditional counterparts? Do CD-ROM dictionaries offer any specific advantages over paper dictionaries for translators? This thesis will address these and other questions. Before establishing the strengths and weaknesses of CD-ROM dictionaries, we had to examine their main characteristics. We first established various criteria for evaluating the characteristics of ten CD-ROM dictionaries. These criteria, in tabular form, were used to evaluate the dictionaries in terms of user-friendliness, presentation, content, search capabilities, text management options, and technical environment. However, these tables say little about users' consultation and search habits. We therefore administered a test to twelve subjects in order to gather data on the use of three CD-ROM and three paper dictionaries. We also asked six of these subjects to fill out a questionnaire in order to obtain their feedback on both types of dictionaries. Using the tables and test results, we were able to draw a comparison between CD-ROM and paper dictionaries and determine their advantages and disadvantages. This study shows that CD-ROM dictionaries have definite advantages over their paper counterparts: flexible consultation, easy handling, speedy navigation, powerful and varied search capabilities, and useful text management options. Nevertheless, they have some considerable drawbacks, especially regarding technical aspects. In conclusion, it is safe to say that CD-ROM dictionaries have the potential to increase translators' productivity as long as translators know how to take full advantage of the dictionaries' capabilities.

The aim of this thesis is twofold. First, on a theoretical level, it aims to examine knowledge from two standpoints: translation theory and artificial intelligence (AI). Second, on a more practical level, it aims to explore the applicability of research in the branch of AI known as knowledge engineering to the eventual development of knowledge-base systems for translators (KBSTs). The "opaqueness" of a text, or its resistance to understanding, often reflects a difference between the translator's knowledge profile and the knowledge profile that the author of the original text assumed for his reader. Consequently, knowledge maximization can be considered a viable translation strategy for countering opaqueness. When translators work without sufficient knowledge, they are forced to fall back on a "transcoding" or word-bound approach, which, although it sometimes produces acceptable results, is likely to produce an unidiomatic text and is much more prone to serious translation errors. The goal of a KBST, therefore, must be to provide enough knowledge to allow the translator to engage his "interpretative" or meaning-based mode of translating. A knowledge-management system developed at the University of Ottawa called CODE (Conceptually Oriented Description Environment) offers a knowledge acquisition and retrieval environment that can be adapted to the needs of translators. A knowledge base on stock-market options, called optionCODE, was developed by the present author using the system to explore the principle and problems of designing and using KBSTs. An informal experiment demonstrated that translators using this knowledge base as their sole source of knowledge to translate a text concerning options performed as well as, if not better than, a translator using traditional sources. (Abstract shortened by UMI.)

This thesis focuses on the notion of text as a project which implicitly underlies a certain orientation in translation studies, traceable through some of Berman's, Brisset's, Derrida's and Fitch's work. Its purpose is to offer a descriptive analysis of this textual perspective through a clarification of four correlated accounts of translation (Berman: 1985, Brisset: 1985, Derrida: 1986, Fitch: 1988). An article drawn from J. L. Austin's Philosophical Papers (Oxford, 1979) and its French translation (Aubert et Hacker, Paris, 1979) provide the descriptive framework for a three-stage analysis. Each stage focuses on one defining element common to the four converging approaches. Translation is thus described successively as a dynamic, supratextual, text-altering practice. Concurrently, each descriptive account of the translation phenomenon generates an analysis of the underlying concept of text, described respectively as a non-linear structure, as a project, as an action. The analysis is illustrated throughout by means of concrete applications drawn from Austin's article and its French translation.