Dear All,
From time to time, there have been mentions in this weekly newsletter of a term that no doubt is familiar to you in general terms: translation memory (TM). But do you actually use it? And if you do, have you ever considered sharing yours with others who may be working with the same language pair?
The Best Practice this week takes up the use of TM in an online tool that you would really benefit from if it can deal with your usual language pair (and I would be very surprised if it cannot!). I refer to Matecat. Matecat is a free and open source online Computer Assisted Translation (CAT) tool. (https://matecat.com). You will need to register, perhaps through your Google or other account, but it will never cost you any money.
But first, just what is TM?
Translation memory is a database that stores previously translated text for future reuse. Entries in a translation memory are logged with their source language — the original text — as well as the translated version in your target language or languages. This means that if there is a sufficient match (usually regarded as around 75% or more), then it will be recognised and re-used.
Translation memories can store single words, phrases, sentences, and even entire paragraphs as distinct entries. Should any of these items appear again, the translation memory will automatically detect it and recommend the stored translation.
These are just text files, usually ending in .tmx, though there are other formats. But the tmx format is simple and therefore can be easily shared with others.
Do you always need TM?
The main purpose of a translation memory is to automatically reuse translations. So a TM will be of advantage to you if there is likely to be any repetition of what you have translated beforehand. If you have projects that have repetitive or closely similar content this tool will be perfect for making sure that you don't waste time on tasks that a tool can do for you. It’s important to note that texts are not always perfectly identical. Sometimes they are only similar, but can still be of use to the translator.
And since a tmx file is just a text file, it is not very ‘heavy’ (even when it contains hundreds of pages of content). It is also easily exchangeable, so that if you have a tmx file in a language pair that might be of value to someone else, you can easily give it to them as an email attachment, and then they can import it into Matecat as a private TM for their own use.
best_practice_28-Using Matecat.docx
