blog.johanka.net

Open Source for Translators (OmegaT)

Being the curious and creative beings that we are, we translators abhor mindless repetition and do as best to avoid routine tasks as we can. There is even a whole industry that caters to this need, translation memory tools software development (or CAT tools, which stands for Computer-Assisted Translation).

For those of you not familiar with CAT tools, they ought not to be confused with machine translation, a different beast entirely. Computer assisted translation works on the basis of giving the user two parallel texts - the original and its translation - placed side by side and split into user-defined segments, usually on a sentence level, but it could also be the whole paragraphs, or localization strings, which are just shorter pieces of text extracted from the software code and destined to be translated to another language. Whenever you work with texts similar to those that have already been translated or need reference material for the sake of terminological consistency, translation memory tools are your friend.

In the world of open source, the alternative to the more commercial applications is OmegaT, a multi-platform software package.

You can get OmegaT from its website. It is well-documented and I strongly encourage everyone to have a look at the user guide and other tutorials that are available at the project's website for free. The Linux version can be downloaded as a compressed tar file, which you can unpack like this

tar -xvf OmegaT*

Using the command

tar -tf OmegaT*

will show its content without unpacking.

Place the folder in a preferred location on your computer and follow the instructions in the Readme file that give you a step by step guide how to install and run this software package on your operating system. Under Linux, you can just type

./linux-install.sh <enter>

in the command line (while in the same folder), press Alt+F2 (or Alt+Fn+F2 when on a laptop) and type in the name of the program (omegat) to start running it.

Now that you have OmegaT running, click on Project and select New (Ctrl+Shift+N). A new window will pop up where you will be asked to specify the source and target languages of your translation project, whether you want sentence-level segmentation and a couple other options. You can also see where the source and target files, as well as glossary entries and translation memory (.tmx) files will be located. No need to add any files yet. Click OK at the bottom right. When the new translation project is successfully created, you will be asked to add the source files to translate and the translation memory (.tmx) parallel texts in two languages that you are going to use as your reference text for the sake of terminological consistency, saving yourself unnecessary keyboard strokes and reinventing translation solutions that you've already dealt with. The source file will then appear in the left pane of the application window. You can type your translations directly into each active segment. Suggested translation equivalents (based on earlier segments or similar entries stored in the .tmx files) will then appear on the right. You can move through individual segments by simply double-clicking on them or using the keyboard shortcut Ctrl+U.

There are, of course, differing views when it comes to CAT tools' usefulness and their being fit for a particular purpose. My view is that whenever a significant time and energy saving is involved it is definitely worth going for this tool. You don't pay anything, unless, of course, you want to express gratitude and support the project financially.

copy left Johanka Piskovská, 2020