All Uppercase Mismatch - Counting of "Hits"


#1

Dear Xbench team,

When “all uppercase mismatch” appears, there is a chance that some errors are not being detected.
For Example, the source has two times the word “SIM”. But SIM appears just one time in the target segment, therefore this error is not detected. Is there a chance to add a function that “counts” these words in source and target segment and will show the error if the number in the target segment is not equal to the number in the source segment?

Thanks for your help in advance.


#2

We will investigate it. Perhaps it is working as designed because counting instances might be prone to false positives.

Examples of cases prone to false positives:

  • For style reasons, the target language tends to use more pronouns than the original (to avoid repeating itself) or the other way around.

  • The translation needs to include parts of the the source text. A typical case are manuals or training materials for a software product that is in English where the translator is often instructed to put software item in English and include a translation in brackets.