What is it?
A two-step process used to ensure that previous translations are correctly matched to the original source in the translation memory (TM). First, the software matches existing translations with their original source. Second, a linguist confirms the matching of source to target.
Why is it important?
Alignment is a great way to retain previous translations that were not completed using a translation memory tool. Done correctly, alignment preserves previous language work and ports it into a reusable format.
Why does a technical communicator need to know this?
Suppose you are embarking on a translation effort with a team of professional linguists who will use a unified translation memory system. You have some translations already; that content is live, and your internal team is happy with it, but it wasn’t done using a translation memory (TM) tool. Can you salvage that work? The answer is yes, with alignment.
Alignment is a two-step process matching previous translations to original source, segment by segment. The texts are aligned by a TM tool first, and then by a linguist who confirms or corrects the matches the tool has proposed. This verification ensures that the source and target are correctly aligned and allows those previous translations to be reused as leverage for future translation work.
It’s important to note that alignment is not an exact science, and if the source and target don’t match exactly, not all content will align into the TM. Also, if there is any question of the quality of the translations, it’s smart to edit the translated content before you populate your translation memory with it.
Linguistic alignment can be expensive, depending on how much content you have and how closely the source and target content line up. It’s worth evaluating the Return on Investment (ROI) before embarking on any major effort. But if you’re happy with previous translation work that was done outside of TM, it can be worthwhile to align that content into a TM so you can continue to put it to use.