Hello! I am a student translator at Brigham Young University and have been experimenting with Language Cloud Term Bases. I have been attempting to create a multilingual term base for future COVID-related translations from en-US into several target languages. I have been using bilingual term bases compiled by Facebook and Google and combining the .csv files into one master spreadsheet. I have run into some problems and found some helpful tips and tricks so I thought I would share them here in case people have solutions or similar problems.

I have found that it is important to spell the name of the language in the identifying row exactly as it is spelled in the SDL Language Cloud system. If not, it has trouble recognizing it as a language. Another thing is to use the Glossary Converter that is made available by SDL (RWS). For those who haven't used it, this converts the .csv into a pair of files, an .xdt file and an .xmt file that automatically set up the structure of the term base and populates it with the terms.

I have had trouble with some languages not wanting to upload even after going through the conversion process, such as Tagalog and isiZulu. I'm not sure where to go from here, so if anyone has suggestions or things to check for then I am open!