yndigo

Computer Assisted Translation

There’s nothing new here for experienced CAT tool people, as this post is more geared toward giving translation customers a clear idea of what CAT tools can mean for them.

Computer Assisted Translation, or Computer Aided Translation refers to a broad range of tools from the very simple spell-checkers and word-count features to the more complicated Translation Memory Managers that store strings of text from originals and translations to be applied when the same text comes up again later. When people in the translation industry talk about Computer Assisted Translation, they are usually referring to the latter technology. A quick intro to these tools can be found on the popular translator website, Proz.

For starters, Computer Assisted Translation is not Machine Translation — a computer program that actually performs the translation — but a tool that assists human translators by obviating the need to retranslate the same (or nearly the same) sentence encountered earlier in the document or in similar documents in the past. The general consensus on Machine Translation is that it has not been perfected to the point that we can rely on the output for most applications. CAT tools on the other hand are in wide use and can reduce time and money and improve quality by ensuring consistency both in and across documents.

CAT tools, a few of the most popular being offered by Trados, SDL (now consolidated in one company, SDL), Atril, MetaTexis, and Wordfast, have also made the translation industry more efficient and competitive, meaning price reductions can be passed along to the end client.

Some detractors cite downsides however: a loss of cohesion when translating segment-by-segment; errors or poor translations perpetuated rather than corrected; tech-savvy translators favored over seasoned professionals, some of whom experience a loss of job satisfaction and leave the profession. A fair treatment of the pros and cons of CAT tools can be found in Dr. Joanna Drugan’s Power Point presentation.

Currently, not all translation jobs are well suited to this technology, and this is often the case for law firm jobs. The majority of documents requiring translation in this area are in hard copy, which, when scanned, can prove cumbersome and costly to convert to editable form, especially if the legibility is poor, so more time must be built into the project than legal projects typically afford. Also, ironically, the high volume and fast turnaround times in legal translation don’t lend themselves well because these jobs often have to be broken up into small pieces and the editing stage must begin even before the translation is complete, a practice that does not play to the strengths of CAT tools.

These factors have also meant that legal translators have not been required to adopt this technology as quickly as their counterparts in other disciplines. But as technology within law firms moves in the direction of e-discovery and native files, translators will need to adapt, too. In the meantime, customers should know that this technology exists and, under the right circumstances, can help them save time and money and promote consistency on repetitive or long-term projects.

Because of the strong feelings both pro and con, I invite other translators, legal or otherwise, to share their thoughts on CAT tools, too.