19

I just had the following edit review task: https://stackoverflow.com/review/suggested-edits/3992632

The editor added some stuff that actually should have been posted as comment or as answer.

The only right thing to do for me is to reject this edit with the "Invalid edit" tag.

However as soon as I pressed on "Reject" I got the message "This edit was already approved".
How is this a valid edit? How can someone approve this? Am I understanding something wrong here? Or is it just the reviewers who didn't pay attention?

  • 34
    Nothing's wrong with rejecting that edit. The fact that it was already approved just means the robo-reviewers were faster than you. – Wooble Feb 6 '14 at 12:42
  • 3
    @RononDex, you might be interested in this feature request: meta.stackexchange.com/questions/137755/… – Chris Forrence Feb 6 '14 at 12:54
  • I did a rollback on that edited answer to remove the edit – RononDex Feb 6 '14 at 12:57
  • 9
    I pinged the three reviewers that approved that edit; hopefully I can help them pay better attention in future. – Martijn Pieters Feb 6 '14 at 13:04
  • If you aren't happy with the edit, you can always roll it back. I understand that doesn't necessarily solve the problem because another edit could get submitted; however, use the tools to the best you can, that's what they are there for. Cheers. – Anil Natha Feb 7 '14 at 0:45
  • Note that the voting wasn't unanimous, two reviewers dissented. It was approved by a 3 out of 5 vote. – Robert Harvey Feb 7 '14 at 0:45
13

There is nothing wrong with rejecting the edit, and it should have been rejected had the reviewers not been asleep at the queue.

It is a Bad Edit

The following issues are clearly visible from the suggested edit review:

  1. It has a giant EDIT: in it
  2. It has two contradicting answers
  3. It says 'now' without indicating how to distinguish

At the very least, the edit should be improved before being accepted. Accepting as is creates an unclear answer ("Which answer is right, and how do I tell which is right for me?"). Improving it should include taking a look at the question and other answers to make sure that you are improving the signal of the site. And if you do that you will find...

The Content is Already Available on that Answer

If you look at the actual question, there is this comment posted over a year ago with more upvotes than the answer:

That's deprecated now. Use the BitmapDrawable(Resources, Bitmap) constructor now.

The Content is Already Available in Another Answer

The accepted answer to the question already covers the same topic as well:

Sounds like you want to use BitmapDrawable

From the documentation:

A Drawable that wraps a bitmap and can be tiled, stretched, or aligned. You can create a BitmapDrawable from a file path, an input stream, through XML inflation, or from a Bitmap object.

Since anyone looking at the question will already find the information they are looking for 'above the fold', there is no reason to edit the lowest voted answer that is incorrect. The appropriate behavior would be to reject the edit, upvote the accepted answer, and downvote the incorrect answer.

And all is well with the world.

  • 4
    The problem is that answer is not wrong, it is just outdated. Secondly, who said that we need to improve only top-voted answers? In addition, the reviewer doesn't see other answers and probably shouldn't consider them as any influence over the review decision. – sashkello Feb 7 '14 at 0:44
  • 1
    @sash, It isn't useful. It is duplicate information that adds noise not signal to the Q&A. If something from the second answer adds value to the top answer, then it should be added to the top answer as supplementary information that doesn't drastically change the meaning of the answer. – jmac Feb 7 '14 at 0:47
  • And this is less of a criticism, but just a statement of fact - examining the existing answers is almost impossible. If you do that thorough of a review on a suggested edit, it is impossible for your vote to count since others will actually reivew, so the fact that the info is available in another answer is irrelavant. – psubsee2003 Feb 7 '14 at 0:47
  • @psu, in an ideal world, you would actually make an educated decision based on what information is already available. The fact that robo-reviewers cause that to be impossible is an issue of its own, but that doesn't mean the principle should be tossed out due to an implementation issue. – jmac Feb 7 '14 at 0:48
  • 3
    And the fact that the info is in the comment should be irrelevant as well as we are consistently told that comments are 2nd class citizens that can get deleted at anytime. – psubsee2003 Feb 7 '14 at 0:48
  • 2
    Quote when we should edit: "To include additional information only found in comments". – sashkello Feb 7 '14 at 0:49
  • 3
    @jmac but we are not dealing with an ideal world. given the status of the network as it is right now, other answers should not factor into an edit approval – psubsee2003 Feb 7 '14 at 0:49
  • 2
    You don't see other answers, and I can't imagine if everyone would examine other answers every time an edit seems too major. I just don't see it happening... Also the top-voted or accepted answer can change. – sashkello Feb 7 '14 at 0:51
  • Rejecting the edit is ultimately the proper decision because it does not add any signal or additional value to the Question and Answer. That this is not easily accessible through the suggested edit review given the time limit due to robo-reviewers is irrelevant. That edit as is should at a minimum be improved if you are going to accept, and if you are going to improve it you should review the question anyway to make sure that you make the Q&A better as a result. – jmac Feb 7 '14 at 0:54
  • And read your own answer here: meta.stackexchange.com/questions/200055/… – sashkello Feb 7 '14 at 0:55
  • 2
    @jmac I'm not going to dispute most of what you just said, but your answer says nothing of this. You are pointing to comments and other answers as reasons to reject and in my opinion, neither are valid reasons to reject a post. – psubsee2003 Feb 7 '14 at 0:57
  • @psu, yeah, I acknowledge that and just added it to the body of the answer to explain in more detail. Apologies for the omission. – jmac Feb 7 '14 at 0:59
  • @jmac ok, I am retracting my downvote. I don't agree with most of what you said, but I don't disagree anymore. – psubsee2003 Feb 7 '14 at 1:00
  • @sash, I know what my answer says in the other location. The goal of any edit should be to improve the quality of the Q&A. Point me to as many rules and posts as you'd like, but if at the end of the day they indicate doing something which adds no value whatsoever I will just reference "Ignore All Rules" which is really the proper rule in this case. – jmac Feb 7 '14 at 1:01
  • 1
    @jmac This is a cheap way out. You are basically saying that every possible thing we say here is irrelevant as you can ignore it later. No discussion is possible in such case. – sashkello Feb 7 '14 at 1:04
-2

In the privileges page, it states that edits should be done to "correct minor mistakes or add addendums / updates as the post ages". So, even if lots of people disagree with this, this particular edit is clearly aligned with the editing recommendations. In addition, there are a few discussions on meta which indicate that this is fine:

Edit old answer and obsolete methods

What to do with obsolete posts?

If community strongly feels like this is a bad edit, then that wording should be changed. Otherwise, this looks like a by-the-book edit (yes, it can be improved, but I don't see why it would be rejected).

I'm actually shocked seeing where the discussion goes here, as there was a huge fight in a few posts about it:

What can I do when my edit was rejected unjustified

Suggested Edit rejected because reviewers didn't know programming language

and I was repeatedly told that I'm wrong and don't understand the rules and that answers are there as a collaborative effort and if you can improve answer, do it. I've got it, OK, if this is how it is supposed to work, and... now I see this? Sorry for the rant, but that's frustrating.

  • 2
    The issue brought up in the other discussions is that you were advocating leaving a clear and obvious mistake in an answer despite the fact that it made the Q&A worse. The issue being brought up here is that this edit is bad and does not improve the quality of the Q&A. The goal of SE is to create a good resource with a high signal-to-noise ratio and any and all discussions/rules should be interpreted in that context. – jmac Feb 7 '14 at 1:04
  • 1
    @jmac: Can you explain how the edit made the quality of the Q&A worse, if it added information to update an obsolete answer? – Robert Harvey Feb 7 '14 at 1:11
  • 2
    @sashkello: I hear what you're saying about collaboration, but the rules that suggested editors follow are (by definition; search for things in meta here like "substantial edits") somewhat different than those of a trusted editor, and this edit just feels wrong to me (in a "I know it when I see it" kind of way). Note that the rules you quoted all apply only to editors with the "trusted editor" privilege. – Robert Harvey Feb 7 '14 at 1:12
  • 1
    @Robert, it didn't add any information that wasn't already in the Q&A in a different place as well as adding mixed signals due to having two different answers in the same place without giving any means to distinguish which should be used when or what 'now' is referring to for future readers. – jmac Feb 7 '14 at 1:16
  • 2
    @jmac: OK. Well I would have preferred a new answer be posted, rather than grafting the information on an existing one. The votes that the current answer has are based on the original answer, not the suggested edit, so the edit can't be properly vetted. – Robert Harvey Feb 7 '14 at 1:17
  • @Robert, the comment below it actually says what the edit did, and that had 7 upvotes. So the edit seems to be as useful as the answer itself I would think. Generally speaking though, adding a separate answer that mimics the content of the first answer is of dubious value I'd say. I'd rather see it as a practical example of applying the documentation from the first example (which would be more visible and useful without being a radical change). – jmac Feb 7 '14 at 1:23
  • 2
    @jmac: Why would you add a separate answer that mimics an existing one? I didn't say you should do that. – Robert Harvey Feb 7 '14 at 1:54
  • @jmac The very fact that I disagree is the reason why you can't simply apply the rule of improve vs make worse, because what you are saying seems like worse to me, you see. This is a matter of opinion and that's why we have rules to guide people about SO, to avoid having everyone doing what they think is "better". – sashkello Feb 7 '14 at 2:18
  • @RobertHarvey 1. Some people with this amount of reputation have different opinions on it as well, as you can see from the answers & comments here. 2. I don't see what's the difference, because if a "trusted editor" thinks that an edit is right, it shouldn't matter if it has been done by another "trusted user" or not. – sashkello Feb 7 '14 at 2:22
  • @Robert, you said "Well I would have preferred a new answer be posted, rather than grafting the information on an existing one." The new answer would mimic information already available in another answer. So if it's a supplement, I'd rather have it edited in to that answer rather than be treated as a new one. I think we generally agree here, but I'm probably not communicating properly. Apologies for that. – jmac Feb 7 '14 at 2:56
  • 1
    jmac: A new answer would only impart new information. I don't see the need at all to repeat anything from the original answer in a new one. – Robert Harvey Feb 7 '14 at 3:50

You must log in to answer this question.

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