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.

At the moment the reviewers are split into two camps, one is accepting minor edits, the others don't. But first, let's start with some basics:

too minor

This edit is too minor; suggested edits should be substantive improvements addressing multiple issues in the post.

And now we'll look at this suggested edit. In my opinion this is way too minor to be approved, but the consensus of the Moderators and most reviewers seem to be that those are accepted. Though, Jeff seems to not like them.

From an editors point of view, it's a little bit confusing, because some people are accepting those edits, some don't...so it's pure luck based on who sees it first if it is accepted or not. In theory an editor could even do the same edit multiple times until it is accepted.

That's a very big hole of inconsistency there, which should be plugged. Either the rejection reason should be refined and clearer outlined what is "too minor" or, if those edits are valid, it should be removed altogether (because in that case minor edits are already caught by the system).

share|improve this question
3  
I believe it's worth noting that the devs may have different opinions than the mods: Jeff and waffles specifically. (And that the opinions of either group don't necessarily constitute policy.) –  Josh Caswell Dec 20 '11 at 8:16
    
That suggested edit is from the user mentioned in this post. –  Keith Thompson Dec 20 '11 at 10:15
    
I've done hundreds of minor edits and I have never seen this message. –  Daniel Daranas Dec 20 '11 at 21:26
1  
@DanielDaranas: Whoops, I completely forgot to add the tag. Thanks for reminding me. Anyway, since you're 8k on SO I assume confuse edits with "suggested edits" which need to be reviewed. –  Time Traveling Bobby Dec 20 '11 at 21:29
    
I don't really understand what you want from this bounty. The existing text for the bounty reason is about as clear as it can be, in my opinion. Changing it would just be rearranging it for no reason. –  Jeff Atwood Jan 3 '12 at 10:12
    
@JeffAtwood: You mean rejection reason? And no, as far as I can see, it is not, hence this question/request. –  Time Traveling Bobby Jan 3 '12 at 11:00
2  
@bobby the existing text is quite clear; what we can't change is human nature: some people have a Mother Theresa complex and want to help everyone, all the time at any cost, and they'll always believe every byte of any proposed change is sacred. shrug –  Jeff Atwood Jan 3 '12 at 12:29
    
Here one of my experiences stackoverflow.com/review/suggested-edits/3357139 My (skipped) suggestion was Duplicate reject auto-reasons should be disallowed after an approve vote. –  Wolf Nov 14 '13 at 13:54
    
@WolfP.: ...what? –  Time Traveling Bobby Nov 14 '13 at 14:09
    
@TimeTravelingBobby now I'm not sure I commented on the right place ;) ... I see some stubborn automatisms in the review process (on both sides), but I think that especially the reject automatic frightens off new users. –  Wolf Nov 14 '13 at 14:35
    
@WolfP.: I think rejected edits is the same as downvotes...it's part of the game, and somebody who gets offset (or even upset) by it might no belong on SE. That said, I'm sorry, but your edit is a perfect example of a too minor edit that should not have been approved in my opinion, there is so much more to fix in that post... –  Time Traveling Bobby Nov 14 '13 at 15:50

5 Answers 5

My two cents.

This edit is too minor; suggested edits should be substantive improvements addressing multiple issues in the post.

Emphasis mine. Let me explain how I interpret that:

  • If you're only fixing one thing in the post, it needs to be the only thing that needs fixing. For example, a suggested edit formatting some code but totally ignoring the poor formatting and bad spelling of the introduction is in my mind too minor.

    On the flip side - if you're just correcting a spelling mistake and that's the only thing I can reasonably see needs doing - sure, why not?

  • Ideally you should fix "multiple issues" with a post because there probably are multiple issues with the post. If there aren't, see above. But look to fix them all. I personally see nothing wrong with an edit that adds a missing semicolon and alters a spelling mistake on the last line if these are the only things wrong with the post.

So my interpretation of this is:

Too minor edits are ones that fall short of addressing all the issues in a post.

I say by some margin here as I'm likely if someone has fixed 7/8 issues in a post to just click approve, or improve - I use this as a guideline and apply it case by case.

However, there's a context here you're missing - namely what happens if someone happens to grab hold of search, find a word and correct every occurrence of that they can find, as happened in your linked post.

I am personally against this kind of approach not because I don't think it needs doing - but because:

  • I do not believe it is possible to fix all the issues that might exist in the posts this way.
  • If you don't have edit privileges, you're causing work for users. Now, nobody minds this, unless you're not fixing all the issues in a post and it's clear there's no thought going into the process at all.
  • Editing bumps things to the SO home page. Again, nobody minds and this is by design, but if you're fixing one tiny issue, is that a reason to bump 100 questions to the home page? I'd say if you're fixing all the issues with a post - including flagging what should be flagged, closing what should be closed etc - great. If you're not, you're bumping a whole lot of stuff...

My overriding feeling therefore is that doing anything "en masse" implies an increasing tendency to not fix all the issues that might exist in any of the posts.

share|improve this answer
    
And on this post, it looks like that spelling error was more or less the only thing that needed fixing - except for the salutation at the end...I'd say this was good edit –  Adam Rackis Dec 20 '11 at 20:08
    
I'm still not comfortable on drawing a line between "improve" and "reject" to the point where I skip over blurred cases like that mostly. –  Flexo Dec 20 '11 at 20:28
    
@Adam: there's at least one other misspelling, very generic tags, a tag in the title; a few judicious ``s wouldn't have hurt either. –  Josh Caswell Dec 20 '11 at 20:31
    
@Problem - I'm sure you're right -- I've never been very a detail-oriented person :-O –  Adam Rackis Dec 20 '11 at 20:34
1  
@Adam: My brain gets itchy when I see writing mistakes, sometimes even before I'm aware of them. It's a mixed blessing. :/ –  Josh Caswell Dec 20 '11 at 20:36
1  
@Problem - My brain gets angry when I see poor punctuation. Singular possessives that should be plural, commas that should be semicolons, colons splinting a preposition from its object, etc. It's definitely a curse :/ –  Adam Rackis Dec 20 '11 at 20:40
    
One comment on this -- I also reject edits that are just plain too minor. For example, changing that to which in a single sentence. It might be correct in that case and the solitary issue, but the value is so limited it's not worth bumping the post for. –  Matthew Read May 7 '12 at 22:40
    
If the problem is with bumping, then add a checkbox to the edit review process that says "Don't bump this post". –  Dan Dascalescu Feb 25 '13 at 4:27

I'm in favor of making the reason both clearer and stricter. I think your example edit is far too minor, although I should say that I don't think that, on its own, it is a cause for tremendous concern.

A large part of the kerfluffle today was the fact that the minor edits under discussion were also massively serial. The user making them had apparently done a search for a misspelled word in question titles (I believe there were at least three different words) and was single-mindedly fixing only that issue, across hundreds of posts. Many, if not most or all, of the posts had multiple other issues that were left unaddressed. The one you've linked to has two sign-off lines, a tag in the title, a misspelled word ("overlayed"->"overlaid"), could potentially use some code formatting and more specific tags, and (for a bonus edit) has bullet points that are (ahem) pointless.

Should we expect the user to have caught all those items? No, but at least one of them -- sig lines -- is glaring and would have doubled the quality of the edit while requiring only a little further effort.

The usual response (most recently expressed by Robert Harvey and NullUserException) to concerns about too-minor edits is "What is the harm?", or "Every little bit, no matter how little, helps." I haven't found a thorough and unequivocal rebuttal of that response; admittedly at the risk of running a little outside the bounds of your question, I'm going to provide one here.

A number of users expressed indignation about the rep or badges that are gained from such edits; this is, to me, not the main point. Rep is in some way significant to all of us, and I can sympathize with the view that a badge like Strunk & White is devalued by trivial, serial edits, but I am more concerned with the following:

First, other users are prevented from making any improvements on a post while a suggested edit is pending. If another user comes along and wants to make a substantial revision, the only option is to wait until the trifling edit has been acted on by reviewers. If this other user doesn't want to or can't wait, or simply forgets to come back, then the post languishes unedited. This is harmful.

Also, the suggested edit queue is a fixed size. When it is full, no edits can even be suggested until it clears up. The user under scrutiny today went a fair way towards (if he did not actually accomplish) filling the queue up himself. If two or more users took it into their heads to make tiny edits like this at the same time, it would actually prevent more thorough, more valuable edits from being made.

Next, as waffles mentions twice, reviewing suggested edits takes up time. Users with the time, inclination, and privilege to review edits are not infinite. To determine whether any given edit is actually helpful takes a moment and some effort. If reviewers get frustrated by feeling that they have to do more to every suggestion (see the answers on Do we need a 'reject and improve' button?), or that their time is being wasted rubber-stamping edits that could be performed by a machine, they may review less thoroughly or less often. This is harmful.

Further, if a significant portion of edit reviewers believe that a class of edits is bad, then, a) democratically speaking, it probably is bad, and b) in order to keep them happy and continuing to review, perhaps it should be considered bad.

Another "flooding" issue is that any edit, no matter how minor, makes the question active and bumps it to the front page. This is by design, to get eyeballs on an edit (a defense against malicious edits, even by users who don't need to make suggestions) and to allow OPs to "freshen up" their questions or answer in hopes of more attention and votes. What this means, though, is that massive, petty edits push newer, active, interesting questions out of a very popular view. This is frankly unfair; it harms the ability of askers to get as much attention as they can (and probably irritates some users of the page).

Next, a post is automatically converted to Community Wiki mode when it is edited by at least five users. Every edit, no matter how piddling, moves a post closer to this. When the post is converted, the owner's reputation is no longer affected by votes on the post. When the post is further worked on by an editor making real improvements, the count is higher than it reasonably should be. An insubstantial edit harms the owner of the post.

There's also a signpost issue, especially in the case of this search-and-edit program. When a single easy-to-search problem is removed from a post that needs more work, the post is made more difficult to find by editors who actually want to complete the corrections.

Finally, a statement of my opinion: I don't think that hundreds of title typo edits are anywhere near as valuable as a dozen careful, comprehensive cleanups of entire questions. Title edits are often held up as the single most helpful thing any editor can do, the idea being that incoming visitors see titles first. I don't agree generally, and certainly not in the case of simple spelling errors. Making titles clearer or more specific is one thing; adding a missing character to a title and leaving the rest of the post just means that visitors will be lured by a pretty sign, order a meal, decide the food is crap, feel cheated, and never return. When a huge number of negligible edits are being made, brakes should be applied.

Edit suggestions which fail to address a majority of the problems in a post should be rejected. The volunteer labor of edit suggesters is valuable, but not more valuable than that of reviewers. Suggesters should be inculcated with the idea that thorough, substantial, even bold, edits are the proper, most helpful, procedure on Stack Overflow. To that end, the rejection reason should be worded more strongly:

too minor

This edit is insubstantial; suggested edits should be comprehensive improvements addressing a majority of the issues in the post.

share|improve this answer
2  
You've mentioned a bunch of technical issues here, like "other users are prevented from making any improvements on a post while a suggested edit is pending" or "the suggested edit queue is a fixed size". Well then, those issues should be fixed, in the interest of improving the substance and presentation of SE content. Having a policy against just because of technical reasons is as silly as arranging keys in a QWERTY manner so that typists type slower because otherwise the typewriter would break (whether this is an urban legend or not is besides the point). –  Dan Dascalescu Feb 25 '13 at 4:30
    
The most significant issue by far is the time of reviewers. The signpost issue is a close second. You're right that technical difficulties should not dictate policy in general. It would be quite a feat, however, to come up with an understandable UX for editing while a suggestion is pending. –  Josh Caswell Feb 25 '13 at 4:36
    
Queue the suggestion and warn the user that another suggestion is pending. Quora deals with that quite well. As for the most significant issue being the time of reviewers, I showed here how that is generally not case, sometimes by several orders of magnitude. TL;DR viewers spend, in aggregate, far more time combined (hours) than one or two editors take to review a diff between two revisions (seconds). –  Dan Dascalescu Feb 25 '13 at 4:59
    
And then what do you do when the suggestion is approved and the edit conflicts with it? Someone has to come back through and do a merge. Reviewers' time is more important than the time of suggesters. Anyone with over 2k rep can edit as little of a post as she wants to. I disagree with your impression of the signficance of the typo correction you suggested on Fitness. In addition, there are at least two other typos that you could've fixed. –  Josh Caswell Feb 25 '13 at 5:16

Personally I would have rejected that edit as too minor. But before I reject it, I consider whether the mis-spelled word is likely to have an effect on search results - IOW is it imperative for that word to be spelt correctly to show in a search result that a normal user would execute? Is it a key word they would search on?

With people who play the part of spelling and grammar nazis, a few rejected edits will soon stop them and cause them to think more carefully about their proposed edits. To a certain degree the problem is not the minor edit proposals (because that is behavior that should be expected), it is the people who approve it because technically it isn't wrong, when in reality they should be rejecting it because it isn't right enough.

share|improve this answer
1  
"it is the people who approve it because technically it isn't wrong, when in reality they should be rejecting it because it isn't right enough"... ?! It's people like you who drive away users who just want to help. –  Dan Dascalescu Feb 25 '13 at 4:33
    
@DanDascalescu I'm not sure what you're trying to say? I personally don't set the site policy, that is set mainly by the community. I frequently edit spelling mistakes where they matter, and there has been a policy about grammar nazis since before I joined. In any case some people just shouldn't try to help. –  slugster Feb 25 '13 at 5:29
    
@DanDascalescu And part of the problem with the whole it isn't right enough is caused by people who are on rep fishing expeditions by searching for and correcting certain problems while ignoring other glaring problems with the posts. This simply isn't helping. –  slugster Feb 25 '13 at 5:32
    
How is other people rep fishing any of my fault when I'm prevented from fixing a typo that demonstrably wastes hours of collective time? –  Dan Dascalescu Feb 25 '13 at 6:05
    
@DanDascalescu You should stop tripping. Yours is a whole different example to the one I was commenting about. It's even on a different SE site which probably doesn't suffer the same problems as what SO does. If you have an issue then raise it on the meta for that site, or message Matt or something - don't hijack an answer that had nothing to do with your specific situation. –  slugster Feb 25 '13 at 6:56

I like editing posts to fix every conceivable error, I can find.

Posts which have spelling errors, usually have other errors as well. If you fix the spelling errors, it makes it harder to find for other people who would fix more of the errors.

I think it should be re-worded, to indicate that if you're going to edit a post to fix one thing, that you should fix everything you can.

share|improve this answer

From my point of view, it all boils down to one question:

Are spelling mistakes worth editing even if it's one single word in old post?

I believe that suggested edits like adding one or two backticks for year old post would be rejected, the problem is with the spelling mistakes correction - some people believe it's worth editing no matter what, and some think it's too minor.

So I would suggest for the team (moderators plus developers) to have a conference and reach a decision:

  1. Spelling mistakes are worth editing no matter what - in such case, edit the "too minor" reason to something like: This edit is too minor; suggested edits should be substantive improvements addressing multiple issues in the post or fixing spelling mistakes. (Or something more fit of course)
  2. Fixing single spelling mistake is too minor. In such case there's not much to do except ask nicely from the team members and members here not to approve such suggestions.
share|improve this answer
1  
I would also suggest: A spelling error in the title is more significant than in the body of a question. Ie. the "too minor" bar is much lower in correcting the title. –  Richard Dec 29 '11 at 11:36
    
@Richard - agreed, thanks. –  Shadow Wizard Dec 29 '11 at 13:21

You must log in to answer this question.

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