Hi! I'm not overly familiar with the details of genereating and using .po files, so my apologies if this is a really dumb suggestion... :-) I'm trying to do some translations in preparation for 3.6, and one thing that struck me is that there are a number of strings originating from UNIMARC and MARC21-specific things like XSLTs and valuebuilders, that no Norwegian libraries will be using anyway. The opposite is also true, the overwhelming majority of Koha libraries will derive no benefit from having the NORMARC-specific strings tanslated into their language. Could it be possible to split out strings from XSLT and value builders into separate .po files, so there is one file for UNIMARC, one for MARC21 and one for NORMARC? (And yes, I know I could just skip the strings that are not relevant, but it would perhaps be "cleaner" to have them in separate files...) Best regards, Magnus Enger libriotech.no