Some of my edits have been motivated by looking at extracted Wiktionary data for inconsistencies, but I am in general interested in a large number of languages, recently mostly Czech.
Additionally I am using the Wiktionary data (mostly extracted by the great Wiktextract project to create open source language learning software, for example ebook reader dictionaries. I am a huge fan of other similar projects such as WordDumb.
List of project ideas (that I might get to in the future)
edit
- Automatically adding Czech comparatives where they exist (to the cs-adj line) (as determined by the Python wordfreq library)
- Checking different Wiktionaries (from the Wiktextract data) for inconsistencies (for example for Russian stress/Zaliznyak classification.)