- Does anyone here have the know-how or motivation to set up an online translator?
- I realize it may not be the easiest thing to do, but given the organization and detail of both the lexicon and grammar, I think a valid effort could/should be made.
- I also realize that like Google Translate, Babelfish, and the other machine translators out there, it might not be 100% accurate and/or necessarily intuitive, but it might draw some attention to LFN and make it that much easier for people to learn.
- Bear in mind that a translation system has to be able to cope with the other language too (i.e. English, or whatever). Esperanto has an online translator, but all it produces is a word-by-word gloss – it doesn't attempt to create valid sentences. As a result, it's only a tiny step up from a dictionary.
- Google Translate is corpus-driven: it uses statistical models derived from large bilingual texts. That's the only machine translation strategy I've seen that produces anything close to an acceptable translation. Unfortunately, we don't have a large bilingual corpus of LFN text!
- People often mutter that LFN syntax and vocabulary are too ambiguous, relying as they sometimes do on a speaker's sensitivity to the context. Computers have trouble with the subtleties of context, although recent evidence suggests they're getting better.
- An LFN machine translator would be a fascinating project to work on, but I'm afraid it would take far more time than I currently have available. Simon
Community content is available under CC-BY-SA unless otherwise noted.