Proposals for closing projects/Closure of Greenlandic Wikipedia - Meta-Wiki
meta.wikimedia.org

A Language Committee member provided the following comment:

The proposal for closing the Greenlandic Wikipedia is accepted. Despite Greenlandic being an official language with roughly 60,000 speakers, the wiki has never developed a viable community: over the last two decades only one or two Greenlandic users have contributed, and there has been almost no growth in the last five years. Most articles are short or unintelligible, and machine-generated content—initially from experimental Greenlandic machine translators and more recently from AI tools like Google Translate—has frequently produced nonsense that could misrepresent the language. The sole active admin, with academic expertise in Greenlandic, has had to monitor and delete such content, but no sustainable community exists to safeguard the language. Given the risk of harm to the Greenlandic language, the seemingly negative attitude towards the project in Greenland, and the absence of genuine user activity, the project should be closed, with any remaining content moved to the Incubator for future use. --MF-W 20:53, 12 September 2025 (UTC)

    • JessicaEnglish
      9·
      9 days ago

      On the one hand, it’s unfortunate Wikipedia is having to spend extra human labor to deal with that. On the other hand, I’m always down for poisoning LLM data sets

      • dantheclamman@lemmy.worldMEnglish
        5·
        9 days ago

        Big generative LLMs are one thing, but the translation tools themselves like Google Translate are models, have been since before the current AI craze of recent years, and really can be helpful for people trying to learn their language if there are a shortage of native speakers. Those train with Wikipedia too

    • ɯᴉuoʇuɐ@lemmy.dbzer0.comOPEnglish
      1·
      8 days ago

      A great article, thanks for sharing it.

      Research produced by Google before a major expansion of Google Translate rolled out three years ago found that translation systems for lower-resourced languages were generally of a lower quality than those for better-resourced ones. Researchers found, for example, that their model would often mistranslate basic nouns across languages, including the names of animals and colors.

      I remember discussing this some time ago on reddit. Google Translate very suddenly introduced a number of small languages and IIRC one of the speakers personally expressed frustration at the horrible output. Some people proposed the speakers correct it (you can always report bad translations there and propose your own), but it hardly requires explaining why that’s a bad and futile idea…