-9

I'm suggesting a reject reason for when the changes a user suggests are ok but they missed other potential (and largely obvious) fixes in the same post.

This scenario is not catered for, happens frequently, and as a result reviewers are currently using the wrong reject reason, which is in turn giving false feedback/advice.

Users seeking rep and badges from suggested edits just make a quick and uncaring change without considering the rest of the post.
Reviewers then have to click "Improve" and and fix all the issues they did not see.

I know sometimes people just don't see things, myself included of course. But I often see a suggested edit and there are lots of other obvious edits which can hardly be missed if the user had any care to actually improve the post, and motivations were not badges and rep.

We already have an issue now whereby reviewers click "reject and edit" when they should have used "improve".
So it's no more of a problem introducing a potentially good reject reason, to cater for when users suggesting edits have just not taken care to edit enough - so we can educate them :)

I'm not saying we should not use the "Improve" option when reviewing suggested edits, of course if a "few" things are missed then use it. We're all here to make improvements and we, I, don't expect users to catch everything.
But if we do this when there are "many" and "obvious" improvements missed, then users are never going to learn to try a bit harder to see those other potential fixes if we do not tell them.


New Reject Reason

incomplete edit
There were other possible improvements which were not included in this suggested edit. A suggested edit should include all potential improvements.

I'm not suggesting this would be used when a user makes a good edit and misses one or two little things, we cannot expect people to be perfect and go over posts with a fine tooth comb.
We already have "Improve Edit" for that.

This is because I often see a user has just not taken much time at all, and there are numerous and obvious other potential improvements.

The edit should be non-trivial and substantial, but more than anything should include at least all the obvious things wrong.


An example:

https://stackoverflow.com/review/suggested-edits/8426653

There are lower case "i" throughout the post, non-capitalised "ruby", many missing prepositions which degrades readability, general grammar could be improved to improve readability, some words are incorrect ("install" should be "installing") etc.

The edit changed one lowercase "i", so why not at least the rest of those if nothing else? And the preposition added was arguably not even needed.

The close reason we use for such cases as the above example is: "no improvement whatsoever", as per the reviews there (and what most people use).

However, this is factually wrong to use in such scenarios, as the edits were actually an improvement!
Also, using the description from the reject reason, the edit does make the post even a little bit easier to read and more accurate. Changes are not completely superfluous and do not actively harm readability.

Wrong message to users

We shouldn't approve such edits as the example, based on their being a good change, because we then pave the way for users to just post any old suggested edits just to get their badges and rep etc.
Even without badge/rep hunters, we should be showing users a better approach to editing, and what we want and need to improve the sites.

Suggested edits are from users under 2K, so arguably are "often" new(ish) users. And we should be telling users (especially new users) that they missed things in their edit, instead of some inaccurate and canned response to them which states "your edit is wrong".

Their edit is not wrong, they just didn't take enough time to check the rest of the post while editing something else.

If we want people to improve and help the sites, then we need to convey accurate feedback to them, not give them incorrect reasons as to why their actions were wrong.

Wrong feedback is not helping make users become better editors, or priming them to be better reviewers when their rep increases to 2k.

  • 1
    My initial reaction was to say "what's wrong with Reject and Edit?" It would probably be good if you addressed the inadequacy of relying on that at some length. – Nathan Tuggy Jun 13 '15 at 4:54
  • @NathanTuggy and what is the feedback and information provided to the user when we "reject and edit"? – James Jun 13 '15 at 11:56
  • It says in the Community ♦ rejection message in such cases: "This edit did not correct critical issues with the post - view the revision history to see what should have been changed." – Nathan Tuggy Jun 13 '15 at 13:45
6

I disagree with your thesis and I particularly don't like your suggestion of a new reason.

incomplete edit
There were other possible improvements which were not included in this suggested edit. A suggested edit should include all potential improvements.

Many edits aren't "complete", even (or perhaps especially) by established users. Having this as an option would simply make people with fewer than 2K rep stop submitting edits should they get this reason because they accidentally forgot one or two things. We want people to submit constructive edits, we don't want to get picky about what makes an edit "complete".

I know mods who will make a couple of minor changes but miss some pretty major stuff. I've called them out on it before in chat and usually the reason is that they were focusing only on tags, chatty text like "thanks", or the title rather than on the entire content of the post.

As stated in this answer to a question on Meta.Movies.SE:

editing - Is it too much to ask that we be thorough?
First of all, sure, of course it is encouraged to fix errors in other's posts, so yeah, if someone edits a post, it's great if he fixes any errors he notices. However, this is not required at all. Every edit that improves a post is, well, an improvement.

I find that I agree with this. If the corrections made in an edit are valid, however incomplete, they're still improving the post. As far as I know, we don't have any requirements about the level of completeness of edits (other than the minimum character requirement).

In your example, it's perfectly possible that the only thing that editor was trying to fix was to correct "done" to "don't" (which was probably the most important change that needed to be made for readability) but was required to change more than that to meet the minimum requirement of characters, so they met the minimum and moved on.

Honestly, I think that's acceptable and, as a reviewer, I would "accept and improve" the edit.

Remember that SE is a multilingual community and not everyone speaks perfect English and can catch every grammatical, spelling, capitalization, or punctuation error... In fact, I know mods who aren't native English speakers and you don't want them correcting grammar as they'll likely just make it worse.

As being a regular on ELL.SE has shown me, it's very common in some areas to do things like not always capitalize the pronoun "I" or to put a space before a comma instead of after it (like ,this). I'm not saying that this makes it OK to leave the question this way in the long run but it does show you that some people are "blind" to these sorts of errors and should not be penalized for it - you shouldn't ban them from editing.

That being said, here's the text as written in the reason:

This edit does not make the post even a little bit easier to read, easier to find, more accurate or more accessible. Changes are either completely superfluous or actively harm readability.

Does the change in your example fulfill this, even to a small degree? - Yes.

So it's actually incorrect for reviewers to reject this edit. The correct course of action is to "accept and improve"... which none of the reviewers bothered to do, so now you've got a very poor question that doesn't even have a couple of small fixes that someone took at least a little time to make.

Think about it... if SE required "complete" edits, then "Accept and improve" wouldn't even be an option... if all edits are required to be complete, then you should never need to "improve" them.

Would it be better to have a more complete edit? - Yes... but it doesn't hurt the system to have something be edited twice... or even three or four times to get it to a good "complete" state.

  • And the char min doesn't even affect 2k+ – Tim Jun 13 '15 at 10:03
  • "Many edits aren't complete, even (or perhaps especially) by established users." so we should most certainly have this new close reason then, if there are many required scenarios to be handled. – James Jun 13 '15 at 12:10
  • "Having this as an option would simply make people with fewer than 2K rep stop submitting edits should they get this reason because they accidentally forgot one or two things." It's pretty speculative to state users will simply stop editing. Some people fight back. Read some review reject questions on the Meta sites ;) Besides, users already get put out now, because people "Reject and Edit" when they should "Improve" - why is my proposal any different? At least with mine users are given more helpful feedback than the current one, from which they can learn, and less likely to be frustrated. – James Jun 13 '15 at 12:14
  • "In your example, it's perfectly possible that the only thing that editor was trying to fix was to correct "done" to "don't" (which was probably the most important change that needed to be made for readability)" based on what, your opinion? Surely the massive lack of prepositions make readability worse, and for most people? "Honestly, I think that's acceptable" Read the help center, and many Meta replies by staff and mods. We reject based on "minor" edits: meta.stackoverflow.com/a/251659/2632129 - let alone minor edits when they also missed other potential fixes! – James Jun 13 '15 at 12:20
  • When users currently miss out valuable fixes within a single edit, they get told "your edit was no good, "no improvement whatsoever" "not even a bit easier to read". And you say my proposal would frustrate users? Your logic is flawed as users are being frustrated now and my proposal would resolve this :) – James Jun 13 '15 at 12:46
  • "I know mods who aren't native English speakers and you don't want them correcting grammar as they'll likely just make it worse" I see no logic in your statement. If a user cannot improve an edit or post, or do not know what to do with a review, then they simply shouldn't do it! That is promoted by staff as well as many others, and I agree. We do not cater for non-English speakers to try their hand at edit improvements. – James Jun 13 '15 at 12:51
  • "Does the change in your example fulfill this, even to a small degree? - Yes." It does not. The edits are arguably acceptable, you even argued this yourself further up your post. The reject reason is wrong because we're telling them their edits were not good, but they were, they just missed out other things. "So it's actually incorrect for reviewers to reject this edit. The correct course of action is to "accept and improve"" No it is not. Why should we allow people to be lazy, or not bother doing much? Their edit was a waste of site and user resources! – James Jun 13 '15 at 12:53
  • 1
    @James Are you not aware of how edits work? Established (+2K) users don't have edit reviews so this "reason" would be pointless for them unless you are implying that one should go around and roll back edits that are "incomplete" to punish people for failing to fix every error? How does it make logical sense to have a less correct answer simply because someone didn't catch every error? As your example shows, clearly the reviewers were too lazy to actually fix the post, while the person who tried was rebuffed because reviewers were being picky about the completeness of the edit. – Catija Jun 13 '15 at 13:23
  • You are completely missing the points. Or perhaps I'm not being clear. Either way, surely it's obvious what I'm proposing? Where did I even remotely give the impression I want to punish reviewers? Of course I understand how edits work, why would you think otherwise? Where did I insinuate that a "suggested edit reject reason" which is by users with less than 2krep, would be used for someone with editing privileges who doesn't get reviewed? My question even states as users less than 2k rep are usually "new"..etc. Did you read anything I wrote? – James Jun 13 '15 at 13:35
1

I also disagree with your proposal.

Many edits do miss some other mistakes, but it's not like it is the end of the world here. Seemingly, if they miss something that you can catch, then edit it yourself. The current reject reason for insufficient editing is:

This edit does not make the post even a little bit easier to read, easier to find, more accurate or more accessible. Changes are either completely superfluous or actively harm readability.

If the edit does make the post more readable and easier to read, just accept it. Then go fix the changes yourself. Don't waste time of others and don't be an extreme grammar critic. I mean just accept an reasonable and acceptable edit and edit the post again to fix the uncaught mistakes. There's nothing wrong about missing some mistakes. Well as long the editor didn't miss like 80 percent of the mistakes and the post is now improved and easier to read.

Edits should only be rejected if the edits are too minor and don't make a good impact on the post. You could decide it's good or not anyways so go ahead and reject if necessary! Anyways, there is already a minimum amount of chars you must add/change before the edit can go to the review queue (if you have insufficient rep).

Really, whether what is too little of an edit is decided by you and your fellow users who are reviewing in the Suggested Edits queue.

  • "if they miss something that you can catch, then edit it yourself" and keep on doing this for them, giving them their rep and badges?We need to educate them surely? I've read various posts by mods/staff who state if users miss various things in suggested edits then reject them, as they have to learn. – James Jun 14 '15 at 0:01
  • "don't be such a grammar critic" this has nothing to do with being a grammar critic, it's to do with other users shoving any old quick edit in there just to get their rep and badges, and I/we have to fix the issue because they couldn't care less! – James Jun 14 '15 at 0:02
  • "Well as long the editor didn't miss like 80 percent of the mistakes" Did you see my example? They missed more than 80 percent, and many they missed were the same as one they corrected (lower case "i" to uppercase) – James Jun 14 '15 at 0:03
  • "Edits should only be rejected if the edits are too minor and don't make a good impact on the post" This is my point. The example review in my question is too minor based on there were many other things to be improved. Should that review have been approved then? (even if using "Improve")? – James Jun 14 '15 at 0:17
  • No because clearly the edits were too minor and they barely make an impact to the post. – Anthony Pham Jun 14 '15 at 0:20
  • ok, so you agree it should have been rejected. Then, as per my proposal, the close reason is wrong, and sends the wrong message. This is not helping make users better editors or priming them to be better reviewers when they reach 2k – James Jun 14 '15 at 0:22
  • You do know that it is rejected, not should been, right? – Anthony Pham Jun 14 '15 at 0:29
  • "You do know that it is rejected, not should been, right?" Yes, but not with a good close reason, hence my feature request. The best/closest reject reason we have when someone does a quick edit and doesn't bother checking much at all is "no improvement whatsoever", and as per my example, most times these pitiful edits are actually correct and useful edits, they're just not even half what could be improved, because users either want rep/badges, or don't take enough time over checking. Hence my proposal, to give them more accurate feedback with a better close reason. – James Jun 19 '15 at 0:15
  • Then couldn't you just make a feature-request whereas people can type in their own reason for rejection? Just like how you can do that for flags? – Anthony Pham Jun 19 '15 at 0:17
  • We already have that "causes harm" and a box to type in. I have some saved text for various scenarios. But "causes harm" is not right either, plus not many people write out rejection reasons, that takes effort. Some do, but a pre-made one is better and easier for everyone - time wise and because then the site can make it sound professional and official. – James Jun 19 '15 at 0:19

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .