Skip to content

For TM Managers

Jump to bottom
Celso-Scott edited this page Jun 19, 2020 · 3 revisions

For TM Managers

If you are managing a group of TM editors for another translation project, it should be easy to emulate the guidelines outlined here or adapt them to your own methodology. The pre-process and post-process scripts will need to be adapted according to the markup data you want to include in your final TMs. If you are familiar with python script you should be able to adjust the scripts as needed in the repository, but contact Celso ([email protected]) if you need help setting it up.

Also, it is likely you will need to train TM editors and review their work. I would recommend having them read the standards and try the alignment themselves following the tutorial. Most likely, you will need to give them occasional reviews and/or feedback in order to confirm that they are suitably following the guidelines according to your expectations. For this I would recommend collecting their submitted alignments and posting them to Google Sheets in which you can share with the editor as well as making comments and giving feedback.

Here is the specific process I use for uploading to Googlesheets:

  1. Collect the three files alignment files from the TM editor by email. The alignment is exported with “Alignment” → “Export”. This will create three .xml files: source, target, and alignment. These three .xml files are preferable to the .tmx file because you will need all three files to import it to InterText with “Alignment” → “Import” in order to make your own corrections. (make sure all three files are in the same folder and select the one with the "bo.en.xml" extension).

  2. Export the alignment from InterText with "Alignment" -> "Export As" -> New Line Aligned (One File).

  3. Using a text editor with regex search and replace (sublimetext, noteppad++, atom, etc...) remove all the xml tags from the text file using the following regex expressions (replace with nothing):

  1. Now the alignment will remain with segments separated by newlines and the source and target separated with tabs. If you select all and copy, then paste into a spread sheet, the alignment will be arranged with one segment per row, separating the source and target in two columns. You can now provide your comments and feedback in the sheet as well as offering examples in further columns or separate sheet tabs.
Clone this wiki locally