How about this: The system uses the time it takes one to answer each cloze to determine whether it was Hard, Normal or Easy. My reasoning is that the faster one is, the easier it must be.
I can imagine a scenario in which the system automatically sets a cloze to mastered if one answers it blindingly fast. Conversely, if one takes long to answer a 75%-mastered cloze, it would automatically be reset to 25%.
Of course, this would need a lot of calibration to get right, so it should be accompanied with a means for individual configuration. And it should probably be optional.
Maybe I’m using this site differently to other people, but for me not everything is about the cloze word; in fact often very little is. The cloze word is just a means to an end. The word itself may be a pronoun, or a possessive, or a basic adjective of size or colour. Sure, I could knock over such basic cloze words in no time flat but the CONTEXT in which it appears may be very different in each sentence. There may be idioms, or colloquialisms, or other expressions that I just haven’t come across before (or don’t remember having come across before because my memory is rubbish, which is why I’m on fairly short review cycles) that I want to research before I move on. After all, the site motto itself is “Learn language in context”.
The flip side of that is that there are some words that I always seem to spell incorrectly if I speed through the input, which means that it takes me longer to answer them even if I know the word and its meaning off by heart. (It’s less of a problem in Italian which is delightfully phonetic, but it still exists.)
In short, my concern with this suggestion is that I don’t want the system deciding when I’ve mastered a complete question by what is often (for me at least) an unconnected metric; that is, the time that I take to type the cloze.
My standard position is not to try to knock any idea on the head if it will work for some other people… as long as it remains optional. I still don’t want the system deciding when my reviews come; imperfect as the system may be (though it’s not currently THAT imperfect IMHO), I want to keep control of that completely in my own hot little hands.
I can definitely see your point. I absolutely abhor the Duolingo concept of Timed Practice, and I don’t want to have anything like that. But still, I think the time to answer has some bearing on whether the sentence is mastered or not.
I’ve thought about that, and it has its upsides.
But ultimately I prefer being the one who decides whether I want to review the sentence soon or not. Besides, I’m quite opposed to any incentive to use the app faster; first because there are enough stressful things in daily life, second because I do not believe that studying faster will yield better results. Sure, you might remember the cloze in a flash, but perhaps you might not have taken the time to listen to the sentence and ingrain its subtleties.
I mostly agree with you, but let me just clarify one thing: my idea was only to measure the time to answer. After that there is no hurry: you would have all the time in the world to take in the sentence in all its glory.
I see this feature as a way to keep myself from fooling myself that I know a word, even after I have wondered for quite some time, maybe even having used the color hints to try out different answers until I finally fall upon the right answer.
That’s a fair point, and I certainly do fool myself too, occasionally. But I still think a system based on time measurement would bother my relaxed state of mind when studying. Of course, this is quite personal.
To put it in more exagerated terms so that you see my point of view, this potential system evokes to me the principles of taylorism - more efficient on paper, but more stressful for the user.
