There are many great new features in the latest release of Lingo from MadCap Software but my favorite addition is the Quality Assurance reporting feature. The feature allows you to choose a specific set of parameters you would like included in the analysis. The subsequent report actually lists hyperlinks to the affected file and lists a segment number for each error. This makes it very easy to find the issue and make an adjustment if necessary. The screen shot below shows all of the various options. A word of caution, you may need the assistance of a linguist to go through the report. The issues listed in the report are not necessarily all errors. I will go through each option below.
- Mismatching numbers in source and target – this is a great feature for finding numeric differences in translation segments. If your project has a long list of technical specifications with measurements this parameter may generate a lot of errors since the translator may have converted all of the measurements to metric units.
- Inconsistency in translation of repeated source segments – this feature will find all of the repeated segments in a project and show any segments that were translated inconsistently. I would resist the urge to just wipe out any differences between segments. We recently handled a project where the translator purposely translated headers or titles slightly differently than the same segment that occurred just below the header. Sharing the report with your translator will help you resolve any of these issues.
- Mismatching punctuation in source and target – this isn’t always applicable in translation because some languages handles punctuation differently but this is still a good parameter to use to locate potential issues.
- Incorrect use of project termbases – this is a great feature! Typically we will enter a clients translated software strings as a termbase (this functions like an automatic glossary that prompts the translator to use specific terminology as those terms come up in the project). This parameter will highlight any potential deviations from the termbase.
- Mismatching whitespace in source and target including double spaces – this is important when trying to find errant spacing especially leading and trailing spaces.
- Repeated target segments with varying source – this parameter will show you any occurrence of the same translation being used for two different source segments. Sort of like the reverse idea of the second bullet.
- Empty target segments – very important for finding segments that were accidentally skipped
- Untranlsated source text in target segments – I would use this one carefully, it will generate a lot of errors since in translation we typically leave things like company names, product names, proper nouns in English.
- Tag errors and warnings – this is very important in discovering issues with formatting tags
- Large variation in size of source and target segments – it is nice that you can set a percentage on this feature, remember languages like German expand a great deal so you might want to use a larger number (like 30%)
Each project will likely be best served by a different combination of these parameters. The best way to discover what set will best serve your project is to choose all of the parameters you are interested in, run the report and then decide which parameters highlight the most useful items for your project. Please keep in mind that you will likely need the assistance of your translator to determine how to properly address the issues.