What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 131 Stack Exchange communities.

I find the minimum requirement on edits is fairly aggravating. Below: Reasons why, and a possible method of alleviating the problem.

Firstly: because there may only be a few characters wrong, it is odd to force me to change other, subjective parts of the post to be able to correct the blatant error(s). E.g., someone mis-spelling “etc.” as “ect” in an otherwise fine post.

Secondly: The error message is obnoxious. “Oops,” it starts. “Oops”? I sure didn’t “oops.” I was correcting an “oops.” If someone should be saying “oops,” it should be the system, for preventing my correction from taking.

Thirdly: This policy discourages me and, I must assume, others from making edits. Is that really the intention of this policy?

Lastly: I’m guessing the reason for this policy is to prevent moderators from becoming bored by approving tiny edits. If that is the case, perhaps a better solution would be a streamlined, inline, character-level diff view… perhaps show a whole screen of extremely short edits for approval as a series of one-line highlights. Like a Gmail inbox of tiny edits. Check, check, check (or select all & uncheck some), approve, done. Then the negative consequences above are avoided. Result: Happy mods and a better-written Q&A corpus.

share|improve this question
5  
Why not make more substantive edits? Why do they have to be trivial, pokes for poke's sake? –  random Sep 26 '11 at 23:42
2  
I gave you an upvote for effort and commiseration, but let me say this: using words like "aggravating" and "obnoxious" , etc. is generally counterproductive. –  Adel Sep 27 '11 at 6:15

5 Answers 5

Thirdly: This policy discourages me and, I must assume, others from making edits. Is that really the intention of this policy?

Actually, yes. Trivial edits (like changing "ect" to "etc") are not worth the reviewer's time. Wait till you have full edit privileges to make such simple changes; then you'll only be using your own time. You can't seriously argue that this edit improves the post to the degree that it should be reviewed (or that the dev work should be put in to make reviewing easier).

share|improve this answer
    
What about my suggestion in the last paragraph that mitigates this issue? –  Alan H. Sep 26 '11 at 23:30
4  
And if this is the intention, then a warning should be displayed. Currently if I go to edit something, the message reads, “Avoid trivial, tiny one-letter edits unless absolutely necessary.” But even if I deem the edit necessary, I am prevented from saving. This is a UI bug. –  Alan H. Sep 26 '11 at 23:40
    
I agree you @Alan H. A ton of friction could be averted by simply stating that you can make small edits at the 2000 point level in the denial message instead of implying that edits of less than six characters are valueless. –  Sean Vikoren Nov 18 '11 at 22:22
    
"Wait till you have full edit privileges to make such simple changes" is unreasonable on most SE sites as even high quality users with plenty of rep here don't come near the required rep. on the smaller sites. –  Nicholas Jul 12 '13 at 14:32
    
@Nicholas On the smaller sites, not as many reviewers are needed! But that's missing the point: Extremely minor edits simply are not worth wasting people's time. –  Matthew Read Jul 12 '13 at 14:41
    
I guess I was proposing that one's entire SE network rep could be taken into account; or alternativly, if one has earned the privlidge on SO, one may be granted the same on other techie-related SE sites, perhaps? –  Nicholas Jul 12 '13 at 14:45
    
@Nicholas That has been suggested before and I like it. I think the reason they've not done it is because many edits will require site-specific expertise, and rep on that specific site is the best proxy for said expertise. –  Matthew Read Jul 12 '13 at 14:47

I’m guessing the reason for this policy is to prevent moderators from becoming bored by approving tiny edits.

Well, that's a fine reason, but there's a bit more to it than that.

  • While your tiny edit sits in the queue waiting for approval, no one else can edit the post. So if there are a dozen problems, and you fix only the one that really bugs you, you've now discouraged a more conscientious editor from making a more useful edit.

  • Every edit bumps the post to the front page of the site. This is a good thing - it gives the community at large a chance to review the edit, respond to a now-clear question, etc. But bumping for trivial changes just creates noise. So you're not only wasting the time of the folks who are approving the edits, you're also wasting the time of everyone who drops in to see what changed.

  • Tiny pollute revision history. When I open up the revision list for a post, I want to see what's changed since the last major revision at a glance, not trudge through a half-dozen minor changes one diff at a time. Could the system be enhanced to make this easier in the face of many small revisions? Sure - or you could just spend the time and review the whole post before submitting your edits.

  • And yes, it does waste the time of folks approving the edits. Altering the UI to make it easier for them to mass-approve edits is counter-productive here: it's already plenty easy to approve an edit without really reading it, so if you're gonna make that even easier you might as well just give up on the whole notion of review. Which probably won't be happening.

Think of the edit-approval period as preparation for having full-on edit privileges. The sort that let you make as many trivial edits as you can stand. Hopefully by the time you can do so you'll have developed the habit of not doing so whenever possible.

share|improve this answer
1  
Sounds to me like these problems could be fixed by software & business logic changes. E.g.: Not blocking further edits (in the age of git and diff-merge-patch, can we merge the edits if they are both approved?). Exclude tiny edits from bumping. And as suggested, rethink moderation experience for tiny edits. –  Alan H. Sep 27 '11 at 18:03
2  
Or you could just... y'know... make more substantial edits. –  Shog9 Sep 27 '11 at 19:18
    
Oh thanks, such wisdom. –  Alan H. Sep 27 '11 at 19:46
1  
@AlanH. Automated merging of parallel edits on SO posts would be far riskier than it is for code. Firstly, you're dealing with smaller texts than the content of a typical source control repo, so the chance of collision is higher. Secondly, you're dealing with natural language, which I suspect is more flexible about where you insert something than code is. If one editor inserts a clarification at the beginning of a paragraph, and another inserts his own version of the same clarification at the end of that paragraph, there's no way an automated system can realise there's a problem. –  Mark Amery Jan 5 at 19:21

Just to echo @Matthew Read's answer, the ability to make small edits is a privilege, which you should just look forward to earning. Yes, there are times when a few letters make a big difference, such as the difference between "mutably" and "mutually" - however the current system tries to ensure that only users with a certain level of experience (both on SO and in programming) are trusted to make a distinction between that kind of correction and edits that are trivial, pedantic, and most of all a waste of time for those who would need to approve them.

Regardless of how wise your small edits would be, you must admit that not everyone else of lower rep would be as careful. The current system prevents an overflow of superfluous pecking, while making new users reconsider what makes a valuable edit. And in the interim, users like you have something to look forward to.

share|improve this answer
    
Good details on the distinction, +1. –  Matthew Read Sep 27 '11 at 1:06

I disliked this requirement when it was applied to me. I came around on it.

Same as requiring two reviewers. It seemed silly when the only way to see one was by clicking "edit" on the post, but when I got better tools and could see twenty at once, I'm glad for the verification.

But I'm really in favor of this:

perhaps a better solution would be a streamlined, inline, character-level diff view… perhaps show a whole screen of extremely short edits for approval as a series of one-line highlights. Like a Gmail inbox of tiny edits. Check, check, check (or select all & uncheck some), approve, done.

The current mechanism is very nice for the more complex edits (or simple edits that look complex, like code reformatting) -- but short edits don't often require much context.

One downside is that the edit queue sometimes shows questions or answers that need a lot more love and care than lower-rep editors have put into it so far -- and a condensed view might make that less obvious.

But anything that helps me never see another post with i or ur in place of I or your is worth looking into.

share|improve this answer

I agree that the tiny edit exclusion for edits seems to be wrong for a programming site. Making people wait to get 2000 points to be able to make small edits is not helping the content.

I do understand though that there is a limitation in the edit submission code where a submission locks edits. So until Stack Exchange incorporates a better merge facility, I can understand that there are practical reasons for this limitation.

Here is the example that I posted that earned me -6 points before it was closed (my apologies for not finding this duplicate question). And before y'all start madly downvoting, please try to remember that this is a discussion.

Take the following example: 1-and-1-in-ruby

The accepted answer proposes a piece of code (has since been fixed by a 2000+ points editor):

"foobar".sub(/foo(.*)/, "\1\1") # => "barbar"

The actual output is: "\u0001\u0001"
The line should read:

"foobar".sub(/foo(.*)/, '\1\1') # => "barbar"

("\1\1" changes to '\1\1')

Of course, I could have done more editing in order for it to be a valid edit, but that seemed like the wrong way to go.

This limitation seems to be a bad one for a programming oriented site. It also seems like having a high quality batch merge tool (multiple changes merged and displayed for approval) could solve this problem.

share|improve this answer
    
Don't worry much about losing six rep points on meta; reputation works differently on meta. It'll come or go. :) –  sarnold Nov 18 '11 at 23:51
    
@sarnold: Thank you for the clarification on the difference in meta. It will help it seem less personal, but as a practical matter, my voice, my ability to vote is being drained away as a reward for sharing my user experience of the Stack interface. –  Sean Vikoren Nov 21 '11 at 16:25
    
Ah, I see, yes, this has cost you the ability to vote. The meta stack automatically gives +100 points to people who have > 200 points on another site -- not many people wish to participate in meta as early as you have, so I've never seen this come up before. Good work. :) I placed a few upvotes on some of your questions and answers on the main stack that had been neglected -- you ought to be good to go for a while now, assuming you don't come up with any ideas that were as unpopular as some of mine. :) –  sarnold Nov 22 '11 at 0:29
    
@sarnold: Thanks! =D –  Sean Vikoren Nov 22 '11 at 15:10

You must log in to answer this question.

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