Consistency below the segment level (that is, consistency of expressions within segments) is achieved by inserting terminology directly from the dictionaries and by other useful features such as the concordance search and the dynamic linking function. On the other hand, when a segment occurs identically more than once in a project, consistency at the segment level can be easily maintained because it only needs to be translated the first time that the translator comes across it — the translation may be propagated to all other occurrences.
Transit NXT provides a special approach to translating repeated segments: the internal repetition mode. This approach allows the translator to translate all identical segments together in one go before translating the rest of the segments. You can translate one segment, then that translation is propagated to the other occurrences in the same group of repetitions, then you translate the second repetition in the project, that translation is propagated, and so on and so forth.
This is particularly useful when a project must be divided among several translators: if the internal repetitions (IR) are translated first by one only person, no segment-level inconsistencies will be introduced later by other translators.
Ok, let us see how the internal repetition mode is used to translate. First of all, open the language pairs:
Now, open all language pairs in global view (that is, in one single window). If you do not select and open all segments in one single window, the internal repetition mode will not work. You can select all language pairs by clicking once on the first one, then pressing Shift in the keyboard and then clicking the last one — the single window option will be chosen automatically for you.
After that, activate the internal repetition mode in the menu Processing | section Internal repetitions | button On/Off. Now it is active.
To go to the first internal repetition, you can use the button Next, in the same section. The cursor will be positioned in the first repeated segment that needs to be translated. You may now proceed as you normally would: look for fuzzy matches, then translate, and finally confirm the segment.
Upon confirming the segment, your translation will be propagated to the rest of the occurrences (from the second to the last one) within that group of internal repetitions, and after that the cursor will be positioned in the first occurrence of the next group of internal repetitions.
Depending on your matches options and on whether there are variants (more than one 100% match), the propagation will be automatic or supervised. Below you can see the recommended setting for exact matches (Insert 100% matches if no variants exist):
If you have been asked to check every segment or you prefer to have more control over your translation (at the expense of speed, of course), you can change that setting to Never and you will be able to supervise the propagation of the translation, checking every 100% match before accepting it.
When the cursor reaches the end of the last file, it means that you have translated all internal repetitions in the project, and you are done! At that point, it is time to switch off the internal repetition mode. It’s important to bear in mind that, should you need to apply any modification to a segment that is an internal repetition, you should apply the same change to the other repetitions within the same group, otherwise consistency would be lost. And that’s it for today.
As usual, any feedback about our tooltips will be highly welcome. Take care and see you in the next Tooltip. Special thanks to Karen Ellis for reviewing this post.
This blog has been moved to a new location, where you will be able to browse it and consult it just like you do now. Stay tuned!