when I have a mistake in my translation, I get to see only one correction, even when I have multiple. I would like to have corrections on all of them.
I understand what you mean but I don't think you need it. Organic learning is part of what Duolingo does.
Euhm, I don´t think you really understand what I mean; first of all, I am not sure what "organic learning" is, secondly, I don´t know what that has to do with what I need. This is a request for a feature in Duolingo's appearance, not it´s internal mechanics.
If there would be a comment for each of my mistakes rather than just one, it would be easier to recognize them and learn from it. For example, I could have missed an accent in Spanish, and also have the gender of a word incorrect; it will tell me I made a mistake, but only point out the gender. It is easy not to notice the missing accent, and hence learn it incorrectly.
I did understand what you mean.
What I mean by "organic learning" is that even if you have a sentence with 2 mistakes and only one is pointed out you will still learn to do both parts eventually. You don't need both mistakes pointed out at the same time - correct the one you are told about, the rest will come later.
The reason I say it is organic is you will be using DuoLingo for some weeks, if not months, and will surely come across the same things again. You don't need to be shown exactly what you did wrong all the time in a robotic and procedural way because you are not learning maths or programming - language is more fluid and over time you'll learn, sometimes by osmosis.
For example - you aren't told, in the early parts of the Spanish lessons, the rules for present tense verb endings but you soon know that verbs for "we" are likely to end in mos. You aren't told that there are male and female versions of the word cat (gata, gato) but you encounter it an it becomes know to you.
You just learn it.
Of course you are free to disagree with me.
Ah, then I misunderstood you; because you said "... what Duolingo does" I assumed you referred to some kind of machine learning I wasn´t aware of.
You may be right, it´s an interesting view on the matter. For example if you were to have 12 mistakes it would be fairly pointless and demotivating to get 12 remarks in red, I agree. I suppose the optimum for learning can only be examined by scientific research.
That may be the way you prefer to learn it, but it is not the way everyone prefers to learn. A simple explanation of a concept such as gender does not render the process "robotic and procedural." For many, it is advantageous simply to point out that "we" verbs end in "mos," rather than relying on "osmosis." The DL comments are heavily laden with people asking for explanations of things grammatical, and questioners are often referred to off-site resources -- which, I think, reveals a weakness in DL's approach.