Translation tools definitely have some ergonomic advantages.

Obviously, translation tools have some great selling points, in as much as they allow the translator to work faster and more consistently, but they’re not always right for every translation and they’re not always right for every translator.

On the plus side, translation tools definitely have some ergonomic advantages. When you’re working with two documents, it can be very hard on your eyes constantly looking back and forth between the documents as you continually find and lose your place over and over again. In addition, the colored text fields used by many translation memory tools are much easier to work with than constantly looking at a glaring white screen with black type. And, of course, less typing is better for your hands. Translation memory segments require that you only look at one segment at a time, which is another advantage for people who read too fast; also, you’ll probably find that you work faster when using programs such as Word fast or AmegaT, even when there aren’t many repetitions in the document.

So, What Are the Disadvantages of Using Translation Memory Tools?

Some of the disadvantages are obvious, while others are more subtle. Match propagation can mean error propagation because you’re automatically applying a given translation throughout the project or document. It’s bad enough receiving a reference document that includes a serious error in translation, but it’s terrible to think that this same error is embedded in the client’s translation memory database, and thus being sent to every translator handling the client’s translation work.

It’s also a matter of opinion as to whether translation memory tools hinder good writing skills. There are advantages to the segment-by-segment approach, but smooth flowing writing is not encouraged throughout the document because you’re actually only working on one segment at a time. With some translation memory tools, the segments that have been translated and approved disappear from the screen, so it’s virtually impossible to create a cohesive text.

It’s also difficult to reverse the order of sentences or words when you’re using translation memory tools because you’ll end up with matches that don’t match up. There’s no way your translation memory tool can know that you restructured a paragraph for stylistic reasons – all it knows is how to match your source and target sentences. It assumes they mean the same thing.

Consider changing your workflow for translation jobs where you’re using a translation memory tool. Instead of using the translation memory tool’s interface for proofreading, try cleaning up, compiling the translation, then proofreading from the target-only document. Now you can input any changes into the translation memory tool’s interface. This will possibly take longer than proofreading directly in the translation memory tool, but if you’re proofreading for tone, flow, and style, it’s almost impossible (for some translators) to work with one segment at a time.