Mistranslations, idiom and humour

Subject: Mistranslations, idiom and humour
From: Susan Harkus <Susan -dot- Harkus -at- xt3 -dot- com -dot- au>
To: "TECHWR-L (E-mail)" <techwr-l -at- lists -dot- raycomm -dot- com>
Date: Fri, 29 Sep 2000 09:17:51 +1100

I caught the tail-end of the numerous postings on translation software and its inability to capture idiom.

I suspect from the tone of Barry Kieffer's mail that some of the contributors may not have used translation software in production and that the experience of most people would be the numerous translation websites.

Today's translation software enables users to create custom dictionaries that actually define the usage of words as well as identify an appropriate rendition in the target language. The software is coming of age and deserves far more credit than it is frequently given.

I do not suggest that translation software is sufficient to achieve publication-ready text but the technology certainly has the potential to produce a 90+% acceptable draft that can be brushed up and edited by a human translator.

For technical communication, translation is not even simply about getting the right language feel but about enabling the "source language" engineering experts to review the technical content of translated versions. For large projects, we are faced with the issue that one engineer is not capable of reviewing all the technical content in the source language ("Fred will have to check that for you. That's not my area."). A non-subjective draft translation reduces the likelihood of distorting meaning. Also, re-translating the translated version into the source language enables the document owners to cross-check the content and identify possible mis-translation areas.

We can laugh at the renditions of software translations that emerge from the simplistic use of translation software, but we would all be quite foolish not to want to reduce translation turnaround times by exploiting appropriate technologies. In the research I did earlier in the year, I estimated that it is possible to reduce translation time for an initial product by at least 33% and more likely by 50%, simply by using a combination of good writing techniques, translation software and human post-editing. I have to say I was astounded at the results but the same percentages were repeated round after round of testing.

When I was doing the research, which was primarily to identify how writing style could complement translation software, I passed a raw translated passage to a fellow technical communicator who had not kept up with the progress in translation technologies, and he was astounded at the result.

If any of us are faced with the challenge of delivering translated versions as well as source language versions, we cannot afford not to be aware of the translation tools at our disposal, and those tools include the translation memory tools that were mentioned earlier this year in a discussion about managing translated versions. Tools like Trados are the mainstay of corporations who have a heavy commitment to releasing localised versions of their software.

So we can laugh... as long as our laughter doesn't prevent us from seeing our options when we need options.
Susan Harkus
Sydney, OZ




Previous by Author: RE: Writing functional specifications
Next by Author: OT: Phoenix, Tempe, Mesa, Scottsdale, Arizona
Previous by Thread: RE: Best application to capture screen shots
Next by Thread: Thanks for burnout thread


What this post helpful? Share it with friends and colleagues:

Sponsored Ads


Sponsored Ads