Correcting Wordfast and Wordbee XLIFF locale code error

If you export files from Wordbee platform and then try importing them into Wordfast Pro, the latter might report the following error: “Error converting .xlf. … Reason: Invalid locale code”.

Wordfast Pro error display
Wordfast Pro error display

If this happens, you will need to edit the files manually.

  1. Locate your XLIFF file and open it in a text editor. It has to be a plan-text editor to prevent file corruption. I recommend Notepad++ because it formats the XLIFF display.
  2. Search, at the top, for target-language=”sr-Latn-RS”. In my case, because I translate to Serbian, the value is sr-Latn-RS:
    <?xml version="1.0" encoding="utf-8"?><xliff xmlns:wb="http://www.wordbee.com/Xliff" version="1.2" wb:xliffId="foo" wb:sourceId="foo" wb:sourceName="foo" wb:translationsIncluded="1" xmlns="urn:oasis:names:tc:xliff:document:1.2"><file source-language="en-GB" target-language="sr-Latn-RS" original="foo.xlf" datatype="x-xliff" wb:fileId="foo" tool-id="Wordbee"><header><tool tool-id="Wordbee" tool-name="Wordbee" tool-company="Wordbee" /></header><body>
  3. Delete the part that identifies the language. Then, add a new value from ISO 639-2 page with the language codes. The snippet should now look like this: target-language=”sr”. Be careful not to delete other characters (=, “, <, >) to prevent an error.
  4. Reopen the XLIFF in your Wordfast Pro. IMPORTANT: Revert the process to upload to Wordbee (edit the files and insert the initial language code).

Wordfast Pro and Wordbee are examples of commercial CAT (Computer-Assisted Translation) software. Notepad++ is a free and versatile Windows plan-text editor with code highlighting (on Linux you can use Kate or Gedit).

Extending the use of Glossary in Wordfast Classic

For some types of translations translation memory (TM) and fuzzy matching are not very useful, but this does not mean that a CAT application itself is useless. On the contrary, you can work more efficiently with the selective use of the powerful glossary function in Wordfast Classic (WFC).

The author of this text has translated many pieces of news. Such content is usually unique and it is not possible to use TM to a satisfying degree: most of the sentences do not repeat. The use of TM is fairly limited. TM here functions as a reference, except in rare cases when background information is appended to a piece (two or three brief paragraphs that repeat occasionally).

TM Compensation, Fast Shortcuts

It seems that WFC if not so useful. After all, if the TM is of limited use, the CAT is helpless, just like a translator. Luckily, this is not true. The WFC has well organized glossaries, which are easy to use due to the shortcuts. That significantly compensates for the restricted usefulness of the translation memory. After putting the cursor in front of a source word (or selecting multiple words) and pressing CTRL + ALT + T once, and then repeating this in the translation segment, the entry window pops up and, after saving, the word/phrase is ready for use. When WFC recognises the term it highlights it, and then the translator can select it by pressing CTRL + ALT + LEFT/RIGHT ARROW, and place it into the target segment with CTRL + ALT + DOWN ARROW.

Overuse of the Glossary, What to Put in It

Here comes a very important question: what to put in the glossary to speed up the translation? Some translators stick to the meticulously selected words they look up in dictionaries, or they load glossaries provided by a translation agency. Of course, we can do this as well. The WFC allows four glossaries at the same time, so there is room for the glossaries acquired beforehand.

We are going to broaden the “range” of glossary and include not only unknown terms, but specifically – phrases and sentence parts/clauses. And here lies the combined power of the WFC shortcuts and glossaries. Have a look at some of the entries in our sample glossary (EN-SR):

company's representative – predstavnik kompanije 
democratic institutions – demokratske institucije
of democratic institutions – demokratskih institucija
with democratic institutions – sa demokratskim institucijama
in Romania – u Rumuniji
in September – u septembru
in several – u nekoliko
Ministry of Infrastructure – Ministarstvo za infrastrukturu
on a political level – na političkom nivou
to do all that was required – uraditi sve što je potrebno
Internet Corporation for Assigned Names and Numbers – Internet korporacija za dodeljene brojeve i imena
European Investment Bank – Evropska investiciona banka
European Bank for Reconstruction and Development – Evropska banka za rekonstrukciju i razvoj

As you can see, nothing particularly unknown is listed above. However, there are two interesting things about the examples. Firstly, it is easy to insert the glossary items in the translation by using shortcuts, so lengthy items are no more a problem. Secondly, by providing several examples of the same phrase, it is possible to cover most cases of a particular phrase. For example “democratic institutions” can be in the nominative case, but “of democratic institution” is in the genitive case. This use is now beyond the management of terminology and closer to the TM use. One might wonder what happens with synonyms. If WFC finds multiple entries for a glossary item, it will display all corresponding meanings from which you can choose by using a small popup window.

THE FINAL NOTE – SPEED

We can expand the meaning of the glossary, understood as a “termbase”, to the notion of “versatile text holder”. Thus, we can include not only unknown terms, but common ones as well. If “known” phrases are lengthy and in different cases in the target translation, this will speed up the translation. However, creating the glossary requires time. Although the inclusion of new items is very simple, it breaks the continuity in work – and could, in fact, prolong the finishing of the translation. From my experience adding items to such “extended glossaries” is best done sparingly during the translation, and in details after the work is finished.

UPDATE: I published this English to Serbian translation glossary on my  Proz.com page.

This text was initially published on ProZ.com Translation Article Knowledgebase (and written for it).