"There is a big square."
Translation:Ott van egy nagy tér.
The default meaning of "There is a big square" is more or less "A big square exists." Is it the same situation in Hungarian? Or would the Hungarian meaning here have to be rendered with emphasis on the "There" in English: "THERE is a big square" or, much more likely, "There is a big square there."
As I understand it, the HU-EN task is a different unit than the EN-HU task. The source sentence can't be changed, but more translations can be added. In this case the English translation has to stay like this, but Hungarian sentences can be added. In the reverse task it's the other way around.
At least that's what would make sense to me, but I can't be sure.
As I understand the Incubator, first the team writes Hungarian sentences. Then they have to provide a suggested English translation of each Hungarian sentence. (Of course, they can include other English translations, too, likely many.) For that suggested English translation they then have to provide all the Hungarian translations. However, this is all under the heading of the single original Hungarian sentence
Incidentally, that same suggested English translation is the same sentence the system uses as a prompt when translating into Hungarian. This is actually unfortunate since teams sometimes choose less-than-the-most-natural English sentences to help point learners to the natural ones in the target language, but then these English sentences also show up as "suggested" when translating into English, when really it would be nicer to give the thoroughly natural English equivalent of the target language sentence.
The reason it's hard to alter the "top level" Hungarian sentences is that there's the constant requirement to have three sentences including each Hungarian word in the course. So simply removing a Hungarian sentence could cause really big problems if that meant some word weren't used enough. But the English translations can be changed freely, and the Hungarian translations of the recommended English translation can also be altered freely, for they're not the ones subject to the three sentences per word rule.
In short, they can easily add "Van egy nagy tér" as a translation for the present exercise, but they can't not have "Ott van egy nagy tér" show up as the suggested translation because it's the top-level Hungarian sentence in the system. However, they can change the English translation thereof to actually match it, which would solve the problem.
Also, the forward and reverse trees are connected via some mechanism that if all the words present are taught in both, they can show up in both. How that all words is beyond me. I've seen contributors (for the Czech courses) say they really wish it weren't like that, and they actively try to avoid creating sentences where it can occur.