I wish correcting were more collaborative in LangCorrect.
On one hand, I often see excellent corrections that have been submitted by others, and I wish there were a way that I could “like” them or otherwise indicate that I think they’re good. This is especially important when a sentence has received two or more corrections and one is, in my opinion, better than the other(s).
On the other hand, I also see corrections that are really good, but have missed one or two errors. This is very common even among professional editors, to say nothing of average Joes like us: we get distracted by the major errors in a sentence and focus on fixing them, only to fail to see other, less prominent errors such as spelling mistakes or an incorrect article. It’s tedious to recreate an entire correction just to fix a spelling error. I wish I could either augment the correction directly (in a way that wouldn’t offend the original corrector), or import the correction into my own set of corrections and edit it.
We actually used to have correction voting functionality, but it was hardly ever used. I think the primary reason was that downvoting a correction required the native speaker to also add a correction to the correction and a reason why it was not corrected. I do like your idea of importing a correction. Do you have any ideas on how to augment the original correction without offending the original corrector? I think this would reduce the vertical “lengthiness” of journal entries too.
RE correction voting, I’d just like the ability to upvote/like. Downvoting seems like it might be too contentious.
RE augmenting another correction: how about a means to select several corrections (via checkbox) and then, when I click the “make a correction” button, these selected corrections would automatically be populated into the appropriate correction boxes (without change markup). I would then edit these corrections and correct other sentences if I wanted to, and publish them in the normal way. To the original poster and other viewers, it would look like I had done my corrections from scratch—there would be nothing to indicate that I had augmented another correction. This would have the advantage of discouraging arguments, but would do nothing to shorten the scroll length of the corrections, unfortunately.
Maybe I haven’t understood your propostion correctly, but how would it differ from actually correcting from scratch? Especially since the current system allows users to correct a few sentences only, so it’s already very easy to correct one or two minor mistakes a first corrector would’ve forgotten.
As for the collaborative aspect, it already exists IMO, since you can comment and discuss matters under a correction. Of course, it requires users to be mature (being a able to give constructive criticism and most importantly, accept criticism) but it can be helpful sometimes.
The main issue with corrections is that learners obviously can’t tell which correction they received is the best — and very often, there is no “best”. A lot of stylistical choices are arguable, and with regional variations, a sentence looking alright to some speakers can look weird or plain wrong to others. Let’s say a text was corrected by an American and an Irish person: if people reviewing the correction are mostly Americans, it would make sense for them to upvote the American correction, but it wouldn’t prevent the text from still looking weird or clumsy when presented to an Irish public.
BTW, digressing a bit here, but a way to indicate very localized vocabulary would be useful. Learners should definitely know if a word/expression is very likely not to be understood (or to look very funny, at the very best) outside of a very specific region.
That being said, I saw more than once corrections containing a lot of grammar or spelling mistakes, which were obviously not due to the corrector being forgetful. Maybe some people think they’re fluent in a given language, although it’s actually not the case. Even natives make mistakes. We tend to forget it, but native doesn’t mean linguistically perfect. It’s very hard, if not impossible, to know about every facet of one’s language.
Let me show an example:
OP: Celine wondered how those leaders expected their subordinates to be honest if they were dishonest.
CORRECTOR_1: Celine wondered how those leaders expected their subordinates to be honest if they themselves were dishonest.
I like the correction, but I would like to be able to augment the correction with commas: Celine wondered how those leaders expected their subordinates to be honest if they, themselves, were dishonest.
Right now, I can do this by opening the same post in two windows, clicking Copy under a correction that I want to augment, pasting the correction into the appropriate box in my correction window, making my augmentations, and repeating this process for each correction that I want to augment. I would like a way to reduce the copy-pasting between windows. And I think it would be helpful to the OP if my augmentations could be highlighted differently than Corrector_1’s so that the OP could easily spot the difference.
Note: The above example is just for illustrative purposes. Whether or not commas are actually necessary in this example is beside the point.
Oh, ok, I see your point! Well, this could befinitely be a good idea for one corrector augmenting a sentence, but in the case of 2, 3 or more correctors re-writing a sentence, all the successive modifications being displayed would become kind of a mess to read, IMO. Or boxes could display only the modifications brought since the last correction only, but we’d kinda lose the point of your system.
A good way to circumvent the issue would perhaps be allowing users to see the different corrections of a single sentence under this sentence, maybe with a system of boxes and subboxes to display the successive augments. So, modifications wouldn’t be directly marked, but at least, it would be very easy to compare them and see what was added to what by whom. Maybe I’m saying nonsense here, I’m not tech-savvy at all and don’t even know if this kind of feature would be implementable (I’m fairly confident it is though, cause “boxes and subboxes” system already exist on sites like Reddit for instance, where you can answer the answer to the answer of an answer etc of a comment and it’s very readable who answered what)
I like many of the ideas expressed in this thread, and while I agree that downvoting can be contentious as there’s room for revenge downvotes which don’t reflect on the quality of a correction but rather express a hurt ego, I think it would be a good thing. One way of handling this would be a system similar to https://stackoverflow.com, where the person downvoting also loses a point - so they have to be quite serious about the correction (and/or are forced to make more corrections to compensate).
The thing is that there’s people who not only consistently overlook errors, but on occasion introduce new ones. While I have no problem adding a correction to make up for them, I’d also like to express my dislike of what they did without having to go through the pain of commenting on their flawed versions individually.
One of my favorite LangCorrect features is the ability to edit my corrections after I’ve submitted them, because I often notice mistakes and fix them after the fact. If an up/down voting feature were added, there’d need to be a way to distinguish votes that were added before the edit. For example, if I upvoted a correction and the corrector subsequently went back and edited it, maybe my vote should be erased, since I might not like the revised correction anymore.
This is a really good point. It highlights the fact that few of us are professional editors or proofreaders (I certainly am not) and therefore we’re bound to make mistakes. It’s human nature to see what our brain knows should be in a text, not what’s actually there. It takes a lot of concentration, and multiple rounds of proofreading, to see all of the typos. The average LangCorrector can’t be expected to produce great corrections even in their native language. But how to encourage correctors to augment each other’s corrections without offending anyone and driving them away? I tried commenting on other’s corrections in the past (on a different site), but 4 times out of 5 the original corrector got defensive, so I stopped trying unless I already knew the person was thick skinned and open to constructive criticism. I wish there were a way to guide people to expect to have their corrections corrected.
My suggestion would just be a “thumbs up” option. If I agree 100% with another person’s correction, I should just be able to thumbs up it rather than try to re-type the same correction. This alleviates the issue that was brought up in this thread about the OP not knowing which correction was best.
To accomplish this efficiently, once I click Mage a correction, under each line, I should be able to see all of the corrections made up to this point.
