I want to apologize for what I did.
I would really, really like it if the ‘Explain’ included the gender of nouns.
eg. No gender is given for Tat, and this is common for many sentences. (It’s probably feminine, but I’d rather be certain, as I’m poor on cases, and also on separating the different ways in which possessive things work (possessive pronouns and genitive adjectives, etc)
I understand where you’re coming from.
In this case, specifically (gender of nouns), I would however give you the advice to use one of the many included links to dictionaries (written by humans) to look up the gender yourself.
Why I’m saying this: The ‘Explain’ feature really is just an artificial neural network that has no understanding of language. It merely produces some words that may look plausible and have a certain probability to actually hit the nail on the head but could very well be wrong.
Do yourself a favor and avoid learning wrong pronouns due to trusting an unreliable artificial neural network that cannot be trusted, when there is output written by humans that is definitely correct.
I find the Explain to be pretty accurate. I catch it in occasional minor oddities (though some of those are due to the sentence on Tatoeba changing), but on the whole it’s very helpful and definitely beneficial to me. It’s the reason I chose to pay for Clozemaster and I’m still happy with that decision.
What included links to dictionaries? (I’ve just realised that the ‘transliteration’ is no longer appearing for me, but I cannot find a setting to change that)
I suffer from RSI in my shoulder. The more links I click, the worse it gets. (typing is a lot easier than clicking) If it’s in Explain, that reduces the number of clicks.
Part of the reason I want the genders is to improve my understanding of adjective endings in different cases and conditions. If the AI gives a gender, I can at least see if it’s explanation of the ending is internally consistent.
Click on any German word and the pop-up window will show you links to Google Translate, Tatoeba, Wiktionary, DeepL, and others.
“internally consistent”. That is exactly the danger I described. The output often looks “internally consistent” or in other words, plausible, but is nonetheless wrong.
On the whole, yes. But what about the cases when it’s not helpful?
Perhaps a macabre comparison: A sadistic doctor who saves 8 out of 10 patients but intentionally kills the other 2 is also “on the whole” helpful. Nonetheless, I personally would never let myself be treated by him. It’s like playing Russian Roulette.