Replies: 2 comments 2 replies
-
@tilmanb ROR's seed data is based on GRID, which had (a not consistently enforced) policy about using English-language names for the primary and varying degrees of label representation for other languages. At the request of many research organizations, ROR changed this policy to reflect language of organizational preference for the primary name and has been updating primary names (and labels, as appropriate) on a per-request basis. However, the current ROR schema does not support language tagging for any other fields beside label. We're looking at changing this in the v2.0 schema, which is currently slated for the end of this year. You can review the draft proposal document and provide feedback here. For now, your best bet is to use a library like langdetect, although I'll note that not every German research organization uses a German language name. It's not uncommon in the European research context for organizations, especially those with an international focus, to only use English (or other languages) for their names. |
Beta Was this translation helpful? Give feedback.
-
Yes, we also are very interested in adding a language code to the name field, @tilmanb, for precisely this reason! Do please add your feedback to the document Adam has linked if you can. |
Beta Was this translation helpful? Give feedback.
-
I see wildly differing styles in labeling of non-english entities.
https://api.ror.org/organizations/03hpxd290 has the name in german and an english label.
https://api.ror.org/organizations/002e7ss22 has the name in english and a german label.
https://api.ror.org/organizations/0291s1e82 has the name in german, an english alias and no labels.
How am I supposed to consistently get a german version of organisation names?
Is there a "best practice" document of sorts on how it's supposed to be?
Beta Was this translation helpful? Give feedback.
All reactions