This is a question concerning the ISO language codes that are used in ELAN. I have observed that, if I set up a new content language, I can only select the 3-letter ISO-639-3 code. However, when converting the ELAN project to FLEx (flextext file), there is a problem, because FLEx uses the 2-letter codes from ISO-639-1, as is advised when tagging languages in codes. Whereas in the ELAN tier names this can be ameliorated, by using the appropriate 2-letter code in the tier name, so the tiers convert correctly, this does not work for the content languages.
Meaning that, when starting out with ELAN, it is not possible to add the correct content languages for the tiers (and, subsequently, the required fonts). For example, FLEx languages like en ‘English’ or ne ‘Nepali’ appear in the ELAN content language list only as eng and nep. Only when the ELAN project is converted to FLEx, and re-imported to ELAN, does the correct 2-letter code appear in the content languages list as en - en - en and ne - ne - ne, respectively. Only at that point in time, it is possible to add the content language / code to the tier.
Am I missing something here, or doing something wrong? Thanks for advice!
The situation is as you describe, so you’re not missing something.
Obviously this is no ideal; on the to-do list is an automatic mapping from 3-letter code to 2-letter code (which would fail or many 3-letter codes). Additionally (or alternatively) the restriction on the 3 fields in ELAN’s Edit List of Languages… window could be removed (this restriction is only there on the level of the user interface). This situation illustrates the trade-off between adhering to (emerging) standards and being pragmatic and flexible.
Dear Han, is there any way to ‘tweak’ an ELAN .eaf file so that it includes two-letter codes? I have tried but I can’t open an .eaf file anywhere else but in ELAN. And I can open the .etf and .pfsx files in Wordpad and add the two-letter codes, but then I can only save those as .txt files that won’t open/import in ELAN. Any ideas?
.eaf files are, just like .etf files, XML files and can be opened in, ideally, an XML-editor but also in most text editors. If you don’t have an XML editor, you may have something like TextPad, but even Notepad will do for simple edits.
If you want to update existing .eaf files, maybe the following workflow can be considered: after an import of a FLEx file, save it as a template (.etf) and maybe edit it in a text editor (or copy/paste the 2 letter LANGUAGE elements from this .etf to another .etf you already have). After that you could try the File->Multiple File Processing->Update Transcriptions with Template…
It has a “dry-run” option and in any way it’ll be best to have safe copies of the eaf files before applying the update.
I managed to tweak the .eaf file through making a new .etf with the required content language settings (before exporting to FLEx and back, that’s the whole idea) by using ZEdit.
I have another issue with adding new participants, but will create a new thread for that.
