A pet peeve of mine is people who copy and paste text from Wikipedia for tag wikis, don't source their text, and then get approved in the blink of an eye.

I'm aware that you can Improve a tag wiki if you have 20,000 rep, but I'm not quite there yet. So my options are:

  • wait for the wiki to get approved anyways, go back and either add a source or clean up the text, or ...
  • reject the edit, and hope that two other people notice the problem and flag the same.

You can guess which option happens much more often.

So, I have a few thoughts on how to handle this:

  1. One or two (preferably one) "Copied content" rejection should immediately reject the tag wiki edit.
  2. Each "Copied content" rejection increases the number of votes required to approve the edit by 1 or more votes, in the hopes that others are given enough time to flag appropriately.
  3. Any "Copied content" rejection voids any previous approvals, effectively starting the review process over.

BONUS FEATURE: Once an edit has been flagged as "Copied content", if you try to approve the edit, you get a warning first along the lines of "There are one or more "Copied content" rejection flags, are you sure you want to approve this?"

DOUBLE-BONUS FEATURE: As above, but once any edit has been flagged for any reason, if you try to approve the edit, you get a warning first indicating that the edit has one or more rejection votes.

...

Personally, my vote would go to #1 or #3, as #2 might be difficult to implement (but I have no idea what the code base looks like, so I could be completely wrong here).

The only problem with #1 I could see is that the "Copied content" flag would have to function differently for tag wiki edits than regular edits; otherwise, you could have people flagging everything as "Copied content" just to reject an edit quickly.

Option #3 could function as-is for both tag wiki edits and regular edits, since I don't think we wouldn't want copied content to be approved for either questions, answers or tag wikis.


Update 2013/02/04

After a week of giving the review queue another try again, I'd be in favor of Option #3 plus the DOUBLE-BONUS FEATURE.

It's a small sampling of data, but I pretty consistently came in as vote #3 on a Copied Content edit, would vote to reject, and then watch as the edit got its third approval vote (fourth vote in total) shortly afterwards.

If approval votes got cleared out from a Copied Content rejection and all future reviewers would see upfront that there is a rejection flag against the edit, this would go a long way to making sure the edits are handled properly.

share|improve this question
    
Maybe when you do a 'copied content' rejection you should be required to state where the content is copied from? Then that information could be used in your double-bonus feature warning (e.g. You'd get a nice "Someone thinks this was copied from Wikipedia" message when reviewing it). –  Keen Jan 28 '13 at 19:30
    
And here's an edit that could benefit from the DOUBLE-BONUS FEATURE: stackoverflow.com/review/suggested-edits#suggested-edits/…. Or a slap on the wrist. Or both. –  LBT Jan 29 '13 at 14:14
2  
I think you should also note that separate from the wiki, tag excerpts are particularly bad because you can't even really add a citation there due to its limitations. I ran across this as did you and I have no idea what action I should take. –  Some Helpful Commenter Jan 31 '13 at 20:10
    
What about a simple disclaimer on the edit-tag-wiki page? Something like "Text that is copied and pasted without proper citation will be rejected in the peer review process." I presume many of the users who edit tags (myself included, up until it was pointed out to me) had never given it a thought. –  Doc Oct 29 '13 at 18:18
add comment

4 Answers

up vote 26 down vote accepted
+50

I've suggested this elsewhere, probably in the comments, but it applies well here:

Whenever a user rejects an edit as either plagiarism or vandalism, the suggested edit should display a large notice indicating that. Plagiarism and vandalism can sometimes be hard to spot for those who don't take as much time to thoroughly investigate every aspect of an edit, and someone rejecting it as either of those should be a red flag to further reviewers to pay attention to that particular aspect. Something simple should suffice:

Another reviewer has rejected this edit for (plagiarism/vandalism). Please pay extra attention.

Not to say that this should be the only feature, but I definitely think it should be implemented along with anything else. Currently the only way you'd know is to click the Reject button and see that there are already votes for those reasons, which never happens if the user doesn't think the edit is Rejectable.

Note: This is slightly different from your bonus feature, as it warns them before they even do anything. Warning them after they've already clicked Approve is more likely to result in them just clicking Ok to move along.

share|improve this answer
    
Yeah, I like this variation much better. +1! –  LBT Jan 28 '13 at 19:31
add comment

Here's what I'm thinking:

When I go to reject an edit, add a text field to the copied content reason, a place to indicate where the content was copied from:

enter image description here

Since we now have this information, we can show it to subsequent reviewers - along with a big red alert bubble:

enter image description here

The link would be clickable, opening a new tab with the allegedly-copied-from website. Then, I, as the subsequent reviewer, would be instantly alerted to the potential problem - allowing us to have a better filter for plagiarism.

I realize it's a big change - it would probably have a slight hit on performance. But it would have a huge hit on the amount of plagiarism that gets through - making the internet a better place.

share|improve this answer
add comment

I'll propose another option to help quell this issue:

  • Have there be a 'do a Google search for this text' link on the review page. Have that link do a search with quotes around it (ensuring exact matches only).

This will help quickly identify copied content not just from Wikipedia.

share|improve this answer
1  
I've suggested that in the past, and it got shot down for whatever reason. I can't remember off the top of my head, I'll try to dig around and see what I find... –  LBT Jan 28 '13 at 19:32
    
Curses, foiled again! –  Keen Jan 28 '13 at 19:36
add comment

I have two main comments about this request.

  1. If you're going to reject it needs to be quick.
  2. Your rejection reason might be wrong.

If you might be wrong I think option #1 is out. You might ruin a lot of another users work and it can be used vindictively easily (though not really directionally). On option #2 I kind of agree; I think it's unnecessarily complex.

However, option #2 plus bonus feature is a good idea. It may very well be that people have not noticed that something is plagarism/vandalism etc. Resetting the vote count does little harm and then there's a warning to alert new reviewers to what's happening.

I would add an additional feature though. If you know it's plagiarised, chances are you've got the link to the original content. It would be good to have a little box that enables you to post this link. As a double bonus feature this could be done after rejection so that you can still reject quickly.

If this were implemented this would enable users to verify exactly whether the content is, in fact, plagiarised. It would massively increase the confidence in such flags and as such could be used for edit-banning purposes :-).

share|improve this answer
add comment

You must log in to answer this question.

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