SAE J2450 pdf download

SAE J2450 pdf download.Translation Quality Metric.
Whde reasonably complete definitions are given below to help the evaluator identify the major categories with some consistency. there may yet be room tor ambiguity and the final category selection may. in tact, be arbitrary. For example. the English word ‘repLace’ may be translaled Into French as either ‘replacer (to put back into place) or ‘remplacer’ (to replace with another item), according to the context of use. That is, the English verb replace’ is ambiguous with respect to French. Suppose that an English text describes a situation where a seat belt must be replaced in the event of an accident because the belt may be weakened by the lirsi accident and tail in a subsequent accident, If a French translation then uses the word replacer, an evaluator should mark this verb as an error. However, is it a ‘wrong term’ error, a ‘misspelling error or a ‘miscellaneous’ error? Without talking to the original translator, It Is fundamentally Impossible for an valuator to know If the translator mistakenly thought that ‘replacer’ and not ‘remplacer’ as the correct verb, or whether the translator simply mistyped and accidentally omitted the ‘m’,
In such cases where the primary category classification is ambiguous and there is no evidence as to which category an error belongs, then an evaluator should place it ioto the category that appears the earliest in the following list. This is completely arbitrary, but ii all evaluators follow this simple rule, then greater consistency of classification across evaluators lI result.
The secondary classification ot an error as either ‘serious’ or minor’ is always a judgment call by the evaluator, and necessarily so. While it is virtually impossible to define the notion ot a serious or minor error, as a general guideline if an error Is clearly serious in Its consequences for a technician or its effect on the meaning of the translation, then It should be marked as ‘serious’. If not, then it should be sub-classified as ‘minor’.lt is a natural and expected consequence that there will be cases when an evaluator hesitates in this choice. Whenever an evaluator hesitates, let us take It again as a general guideline to sub-classify the error as ‘serious’. This again is completely arbitrary, but it seems the safer alternative since safety may be an issue In a service text.
Once the primary and secondary classifications have been assigned, then a numeric weight between 1 and 5 is looked up in the definition and assigned to the error. 5 corresponds to the most severe error, while a 1 indicates an error with relatively small consequences. This assignment of weights is a mechanical process and there is no room for judgment. While sometimes this assignment of numenc weights will over-value the severity of an error, it will under-value it at other times. The underlying assumption of SAE J2450 is that these deviations will tend to cancel each other, I.e.. “regression to the mean” will normalize these deviations on a large evaluation text or with a large number of evaluation texts.
The SAE .J2450 categones are meant to be used to tag errors that are linguistic in nature. Thus, a translation deliverable that is free of any SAE J2450 errors may still be unacceptable to the client due to other problems, such as formatting errors.
ALso, SAE J2450 is designed only for the evaluation of translations of service information where the style, tone, register eic, of the target language deliverable is not as Important to the client as it may be for owners manuals or marketing literature. The target customer of the translation is the service technician. Therefore. SAE .J2450 deliberately avoids tagging “errors” of style.
The definitions as follows are generally syntactic in nature. That is, they depend upon the surface form of the translation deliverable and are generally divorced from the meaning. However, meaning is accommodated in the notion of a ‘serious’ versus a ‘minor’ occurrence of an error type.
The general concept is that once an error has been identified by category. the evaluator decides whether the effect on the translation of that error is severe or not. If severe, then the error Is classified as a ‘serious’ occurrence of that error type, and otherwise it is regarded as a ‘minor’ occurrence. This is necessarily a judgment call by the evaluator.