Sick of "Oops - we can't tell if this is correct" for new translations!

I find when undertaking a complex translation where fewer than 10 or so Duolinguists have contributed, and none have been rated, it is difficult to achieve a score of 50% or more. So having contributed my knowledge and congnition to help interpret the text, I get rated against some crass translation from the DuoBot - nul points! A number of early posters even seem to have changed some of their poorly scoring(?), translations to line up with the "best" translation - from the DuoBot! Come on DuoLingo - where do you think this trend is going to lead to??? There needs to be a better scoring model when there are not enough contributors for the DuoOwl to make a decision. Otherwise, no one is going to start a fresh translation. At the very least, the quality of your translations start heading out of the window until enough good DuoLinguists pile in to start reversing the trend. As was suggested elsewhere, your Skill level should be factored in when determining the "best" translation ratings and anyone with a good record of translation should automatically receive the benefit of the doubt in these circumstances. And/or maybe the DuoOwl should adjust the pass mark according to how many times the passage has already been translated and at least award half marks for any decent attempt.

June 30, 2012


I would agree with this problem of being the first or an early translator. It is particularly an issue if the text includes something which either contains idiomatic German, or would be best translated by an English idiom to make the translated version more readable: the dictionary only does the word-for-word and therefore comes up with the "You're first but this doesn't look right message" e.g. the literal translation would be "moisture in the air", but we'd actually say "humidity". I know I can be awarded bonus points if later translators rate my translation as "very good" ... but I still don't get the full value of the original points. Perhaps Duolingo could in some way "review" the non-scoring translations once there have been a body of other translators and retrospectively award the points?

July 1, 2012

@Kags - Thank you - an excellent example of what I'm talking about, and retrospective awards are a great idea - perhaps even doubled for the insult!. Although my main concern is that DuoLingo will NEVER recognise that a better translation exists, if everyone just follows the status quo to avoid losing points themselves. We might as well stick with the word-for-word translation howlers, if that is what is being encouraged by the scoring system.

July 1, 2012
Learn a language in just 5 minutes a day. For free.