Forum: MemoQ support
Topic: A few intermediate-level questions about the functionality
Poster: Stanislav Okhvat
Post title: Some answers
Hello,
>> 3- What are the differences between a word in:
>> a. the non-translatable list
>> b. the ignore list
>> c. the forbidden words list
Non-translatables are text units which are not translated from one language to another (e.g. "Apple"). Ignore list items are related to spellchecking, these are words that should be ignored by the spellchecker even if spelled incorrectly. If a term is marked as forbidden (by checking the box) in a term base, this translation may not be used if the source term appears in the source segment, and this is flagged by QA checks and in the list of hits on the right side.
>> 4- I quickly added several document pairs to LiveDocs and thought they were very well auto-aligned in general, with few mistakes, so I committed the (unreviewed) alignments to the TM. Later, I noticed there were actually many alignment mistakes. How should I proceed to correct these in the smartest manner? Is there a TM-editing function where I can delete the wrong entries? Or perhaps if I go and review the alignments and re-commit the fragments to the TM, the (now correct) alignments will automatically overwrite the bad ones? Or even start the TM over from scratch since it's so new and I won't be losing much? (those are a lot of question marks, I know, and I'm sorry)
Yes, memoQ has a TM editing feature. You can display TUs created between certain date/time instants and then remove them. In your case, however, I would recommend to recreate the TM (delete and create a new one), add the new one to the project and then update it from all the documents which you have translated recently. There is Confirm and Update command under Preparation tab for this - it simply writes all segments of certain status (e.g. Confirmed) to the TM without performing propagation of translations as the Confirm command does.
You can also realign and update the TM, but all the segments will be updated only if you don't split/join segments, because by splitting and joining you are effectively changing the context of the segments and source segment text itself.
>> 5- What is the best way to handle plurals in the term base? For example, consider the term "subject", which is translated in Portuguese as "participante da pesquisa". The plural, "subjects", becomes "participantes da pesquisa". Should I:
>> a. make separate term base entries for "subject" and "subjects", or
>> b. one entry as "subject*" ->"participante* da pesquisa", or
>> c. one entry as "subject|s" ->"participante|s da pesquisa", or
>> d. add the plurals with the little "plus" icon within the same term base entry, or
>> e. something else?
Use 50% Prefix (default) setting if the inflections of the term are formed by adding letters to the end of the base form. Use "Custom" + * or | symbols if inflections are formed by adding letters in the middle of the term or by changing letters. In the case of "subject" you use 50% Prefix setting, but for "participante da pesquisa" you need to set the text to "participante* da pesquisa" which will also change the Matching value to "Custom". This is because in the inflected forms letters are added in the middle of the entire term, not only at the end of the last word.
>> 6- What is the best way to handle capitalization in the term base? In the above example, "Subject(s)" is capitalized as "Participante(s) da Pesquisa". Do I make separate term base entries for the capitalized version of the word, or do I add them with the plus icon within the same entry?
The default case sensitivity setting in term bases is "Permissive". This means that memoQ will find a term if it contains the same letters as the base form or the upper-case equivalents of these letters. However, with Permissive setting, memoQ will not find "microsoft" if you set term text to "Microsoft", because the first "M" is already uppercase and is not allowed to be in lower case. If you use "No" (case-insensitive), memoQ will also find "microsoft" because case does not matter. In some cases, for example when you are supplied with big Excel termbases from clients, I would advise you to set default case sensitivity of the termbase to "No" instead of "Permissive", because such termbases may contain regular terms with uppercase letters instead of lowercase letters.
Best regards,
Stanislav Okhvat
TransTools – Useful tools for every translator