As for George Steiner's "The Hermeneutic Motion", I'm not really sure what to say about that... After reading it, it feels like it's one of those things where I technically understood the words he was using but couldn't understand a damn thing he said the whole way through.
As for the Google Translate article, honestly, I'm surprised by his description of how it actually works, as I honestly wasn't aware that it just looks for similar, already existing sentences out there to achieve its translation. Granted, this article was written 10 years ago so I'm certain it's changed significantly in its approach, but I think it's rather brilliant. Rather than try to calculate a translation based on a set of rules, just look for it's equivalent somewhere out there in the ether of the internet. It was such an obvious choice, I think, given that Google is at its core a company dedicated to search and retrieval of information. It is pretty amazing that even after 10 years of work, though, that Google Translate's performance is still rather poor, especially when translating languages like Japanese into English. Although, I guess it's really no surprise, as like Bellos said, "when it comes to real challenges in...translation, human beings have a hard time of it, too". After reading this, I'm actually quite interested in looking into how some of the other translation systems out there compare in their approach to Google Translate, especially ones (like Papago or DeepL) that seem to fair much better when translating East Asian languages like Japanese, Korean, and Chinese.
No comments:
Post a Comment