The above link makes a very clear case about the relative usefulness if writing correction.
same link
Ed, this is very interesting. I would like to post this link on my blog and get more discussion going. In my experience I find that writing correction, or correction of my spoken mistakes, helps only if it is used to make me more attentive to certain patterns. This increased attentiveness helps me when I am reading and listening, so that eventually my brain catches on. That is why I find importing the corrected texts, and discussion report, helpful. I read them and save words and phrases.
I do not expect to get it right the next time, I also do not read the explanations of the mistakes, which I find not at all helpful. Either I know why I made the mistake, or at least what the correct form should be, or I don’t. In either case the explanation does not help. I just remember to try to notice this form more when I read or listen.
I feel, that with enough input, and noticing, I will eventually improve.
@Steve “I do not expect to get it right the next time, I also do not read the explanations of the mistakes, which I find not at all helpful.”
I fully agree to the first part of your sentence. I cannot understand the second part, although I don’t often linger over explanations either. If a tutor just corrects writing without providing an explanation in some (a few) cases at least, some students may feel at a loss to understand what’s wrong with their writing. I tend to give more detailed explanations to more advanced students because it might help them avoid certain mistakes. With beginners or intermediate students it’s often a waste of time - yet they might return to their previous writings later and find an occasional explanation useful.
I always import corrected texts etc. and draw my own conclusions, then try again to write without thinking about any explanations that it’s too early for. Nonetheless, I wouldn’t say they are “not at all helpful”. But I can see your point, students differ in many respects and it’s up to them to decide what they use, just as I will decide what I want to explain.
The linked article suggests that word and sentence level form focused corrections are a waste of time, across all student demographics, backed by many research studies.
Reinhard, I am just reporting how I react. I rarely read the explanations. If I used the wrong case, I will see that in the correction. The explanation of why this is the wrong case is usually superfluous. Either I know the reason, but can simply not remember the correct case when using the word, or in some cases the explanation does not yet make sense to me. I am unable to connect the explanation with correct usage. What I find most useful is the frequent exposure to the correct usage.
In this regard, I am listening to a recording, about 60 minutes long, of a number of examples of correct usage patterns in Korean., As long as I can see what the meaning of each pattern is, in this case in Japanese in my little book, hearing example after example, repeated four or five times, is quite effective in giving me better awareness of these patterns when I come across them. I find that it is at this level, and not at the level of explanations, that my brain starts to familiarize itself with the language.
I now find that my reading and listening comprehension of Korean has improved, since I notice these patterns better. My ability to listen to this kind of stuff is limited, and I have to alternate it with meaningful content. However, I think I will listen to it, once every couple of weeks or so, each time picking up something new. It is the examples, not the explanations that cause the bell to ring, at least for me.
Steve, I know it works in this way, it works for me too. However, students who do not listen enough and just occasionally write may not notice the patterns unless they are pointed out and sometimes explained to them.
@ Edward “word and sentence level form focused corrections are a waste of time”
Sadly I must agree to this as well, seeing students make the same mistakes over and over again. I might just as well stop marking their writing and just grade them based on the impression I get from reading their writing. It’s the bureaucracy that forces me to mark mistakes and even distinguish between different categories of mistakes. It does not help students to learn the language any better.
I admit that it’s my “teacher’s mindset” that makes me want to explain even though I know that it rarely has an immediate effect. A teacher thinks, why, it’s so easy, don’t you understand? - No, they don’t 
Without involving myself into all said above, I just want to note this.
Reinhard, the comments I’ve got from you on our two (I think we had just two) conversations in English were the most “stuckable” in my brain. For example, I had said something like “We with Helen…” - and you commented that it was a typical Russian usage, that in English we’d say “Helen and I…” And it has readily stuck (I hope), may be because you correctly mentioned the Russian usage pattern.
What I mean to say is that you hardly have to reconsider your good teacher’s skills 
“We with Helen…” a typical Czech usage as well. Ilya and I belong to the same language family.
“If ESL/EFL writing teachers are really concerned with improving their student’s grammatical competency, they should, in lieu of offering grammar correction feedback, constantly stress in their classes the importance of outside reading.”(Ronald Gray)
He contends that teachers should “constantly” stress in their classes the importance of outside reading. I wonder how many of their students follow their advices constantly.
"Minimizing grammatical error feedback has the advantage of greatly simplifying teachers jobs, giving them needed time to spend on concentrating on other important elements of the writing process, while also removing a significant impediment to their students learning how to effectively write. "(Ronald Gray)
Writing correction “is” useful, but grammatical error feedback should be minimized. This might be Ronald Gray’s message to ESL teachers.
Ronald Gray’s article is mainly based on Stephen Krashen’s hypotheses, and he is concerned with language teaching in brick-and-mortal classrooms. I think that the LingQ system including the “Post and Submit for Correction” function is different from ordinary educational institutions.
"I think that the LingQ system including the “Post and Submit for Correction” function is different from ordinary educational institutions. "
LingQ could test the efficacy of writing correction on its site by tracking error types per student-teacher pair over time. Tutor error would muddy the waters a bit, but it still would be interesting.
There are other factors as well, like the inherent complexity of the intended text. Maybe it would not be so easy to do. I would have read an actual study and see how they set it up.
My personal experience as a classroom teacher, like Reinhard, tells me this paper is bang on in its findings. This includes the most motivated students I have… a group to which I guess LingQ writing students would belong.
I think the LingQ correction tool is useful however. It provides a channel for meaning focused and discourse focused feedback. It also allows students to see correct examples of language that function the same way as language that they have trouble with, as judged by the tutor.
As long as we realize that having mistakes corrected does not correct the mistake in one’s usage, I think it is useful to get some feedback, because some of it, maybe only a small amount, actually sticks, since it helps us to notice things in the language, at least that is my own experience. On the other hand, the actual act of writing or speaking is far more important than the correction.But getting the correction back, or the discussion report, is also a motivation for writing and speaking.
I think that the value of a correction depends on the kind of mistake and how the correction is presented.
I’ve spent well over one hour (maybe up to one hour and half) on a text that gave me ~700 points. Should I just have corrected the mistakes and given no comments? It would save me a lot of time, but I imagine it wouldn’t be that helpful.
Some texts I’ve corrected over the years (at all of the websites where I do such work) have contained all kinds of random mistakes. At times I’ve wondered whether the writer had paid any attention AT ALL (to the texts he/she has read), not beacuse of the low level in general but rather the “randomness”.
Has the submission been written in a couple seconds? Good spelling, bad spelling (even when the same word has already been in use), passable grammar, abominable grammar… Call me stupid, but when there is a kind of “logic”, I think it can be explained and taught. So that’s what I do. I’m giving the correct words, a basic rule (if there is one), and maybe some related examples. All I’ve had contact with have appreciated that.
Why are learners (said to be) having problems with as articles like a/an, the…? Or when to use plural? The articles are everywhere, in any text of any level. Still it seems that some learners don’t grasp it, at least not by pure exposure.
Would a simple explanation be beneficial or detrimental?
In my experience, and based on the research, the explanation is largely ineffective, but can help in some cases. I wonder how many of our learners at LingQ look at the explanations? I think the perspective of the learner is more important in this regard than the perspective of the teacher.
(jeff)"Why are learners (said to be) having problems with as articles like a/an, the…?..
…Would a simple explanation be beneficial or detrimental?"
You have a simple explanation? (That is valid beyond a few example cases)
Steve: I agree that the perspective of the learner is important. I don’t write an essay for each corrected mistake, but my feeling is that if I’d just written the right word and nothing else, the learner might have chosen another tutor next time due to not getting valueable input.
Ed: You’re the native English speaker… I’m sure you can sum up when you say a car/an apple/the car/the apple/cars/apples/the cars/the apples, and why.
(Jeff)" You’re the native English speaker… I’m sure you can sum up when you say a car/an apple/the car/the apple/cars/apples/the cars/the apples, and why."
If you confine yourself to concrete nouns, sure. But even there, there are problems … issues with countability, collective nouns, use of “every,” “any” the list goes on.
But most people, even beginners, want go beyond the concrete in their texts.
So, neither I (nor you) could sum it up in a way that is both useful to the learner and actually valid beyond a few examples. (BTW being a native speaker , by itself, is usually a disadvantage in describing language.)