Is everyone in favor of the new "everyone can edit" idea? Is this really such a good idea?

Question

As originally requested here and described in Diff is Hard, Let's Go Shopping!:

[The team is] planning to open the floodgates and allow everybody to submit content edits on posts

Am I the only one who sees this undesirable? In fact, the more I think about it the more I think it has the possibility to be a really bad idea.

I think that, to provide good edits, you must have a good idea of how the site works. I am very afraid that if you give everyone the power to edit, we'll get a lot of low quality edits. The /review link was just added to help weed out low quality questions and answers... if there's a need for such a facility for questions and answers, why do we believe that edits will be much different?

I understand that these edits will have to be approved by the community:

2 Users with N reputation will have to approve each edit. There may be a capped reputation incentive (at the moment we are thinking you can get up to 1000 points 2 at a time for good edits) – waffles♦ yesterday

But I think that this potentially doubles the work for people (like myself) who like to edit. Now, instead of just doing the edits ourselves, we have to review someone else's edits and it now takes two users to approve the edit another user made, so what could previously have been done with one user now takes three.

There's another consequence to this:

@Robert ... this helps fight information rot... so many old questions need TLC. pending edit will block further edits and be actionable inline. – waffles♦ 23 hours ago

So, if I understand that correctly, that means that if some low-rep user makes a bad edit, A higher rep user can't correct it until a second higher rep user concurs the first edit was bad?

Or even if a low rep user makes a small, good edit, a higher rep user can't make a second edit until the first one is approved? I foresee all sorts of problems here.

As it stands, reaching 2,000 reputation is a great milestone on the sites. It means you have proven yourself and, most likely, you know the flow of the sites and what's good and what needs work. Congratulations, you now have the power to edit and make these sites you clearly love a better place. I think that giving editing powers to everyone greatly diminishes the privilege of being able to edit at 2k, and I think that's a bad thing.

Am I the only one who feels this way?

Is the new "everyone can edit" feature set in stone or do we, the community, have any say?

Answer 1

Am I the only one who sees this undesirable? In fact, the more I think about it the more I think it has the possibility to be a really bad idea.

Theoretical problems are fantastic things. I have over 1000 theoretical problems with over 1000 theoretical features.

---

What I can talk about is the actual process we are following, and why I believe this will work.

• We are turning on the floodgates for a few hours at a time, during the next few weeks and gathering proposed edits from anonymous and users without full edit rights.

• During this process we are tweaking stuff so handling the proposed edits is super easy. We are also adding some more restrictions to edits, such as you must enter a substantive comment for any suggested edit, edits must not be no-ops, etc.

• Edits can be approved inline (on the question/answer page) by 2 users with the pending edit approval privilege, earned at 1k rep.

• The vast majority (like 95%) of edits are approved/rejected by the team during the test runs.

• We are building analytics bits to see the effect of the test runs.

• At the end of this process we will analyze the data, tweak algorithms and ensure that the feature has a general positive effect on the site, figuring out rep incentives caps and so on is a phase 2 thing.

• Nothing is set in stone, everything depends on the quality and type of edits that are submitted in our real world trial runs.

Answer 2

I have a bunch of rep points on SO, waaay more than I have any need of. And I hate answering questions.

I have rep points on SO because I wanted to edit. When I first read about SO, six to eight weeks before the site even existed, I thought to myself, "A wiki Q&A site! Brilliant! Finally, a solution to this endless parade of crappy questions with their horrible formatting and broken ingrish."

And then I joined on, and... couldn't edit anything. It sucked. Every day, I'd check in, see something in need of editing, and... facepalm

I got over it, eventually, after posting far too many mediocre answers. But every day, there are users showing up on SO who can't use the site's single most important feature, the one that first set it apart from every other crappy Q&A site, the one that daily breathes new life into mediocre posts.

It means you have proven yourself and, most likely, you know the flow of the sites and what's good and what needs work.

It means you probably know how the site works. How Markdown works. What sorts of edits are appropriate and which ones aren't. A user just signing on probably wouldn't know this stuff. But that's ok, because under the proposed system their edits will need to be vetted first.

It doesn't mean you suddenly have the desire or ability to edit effectively. Nor does it mean that someone just joining the site doesn't have anything of value to add... Right now, we force them to post answers, even when what they're contributing would be better off as an addition to an existing answer. That sucks.

There's nothing anarchic about this. If anything, it strengthens the idea that new users need to spend time learning how the site works, and provides a way for them to do just that while keeping them under strict supervision until they've had some time to get the hang of things.

It's a learner's permit for editing. Perfectly appropriate for the single most powerful feature on SO!

Keep in mind, this feature is still under design, with the team open to suggestions as to how it works and how it looks. Figure out a good way to implement it, and we'll all benefit...

Answer 3

The alternative is that the flawed questions/answers stay that way or that 2k-users have to correct them all by themselves. Just approving edits is less work than editing all those posts yourself, assuming most edits are constructive and useful.

I think it is at least worth trying out; it might be a very useful addition. And if it turns out that the low-rep users don't know how to edit constructively, it can always be turned off.

But I really think that the mere fact that you're trying to improve someone else's post shows an interest in the quality of the site. That is probably a good indicator of being able to edit constructively. Most bad quality posts I see are from users who don't care... they wouldn't spend the effort to edit other posts.

Answer 4

I'm going to throw out a possible solution to some of the problems I see, having read some of the discussion in chat and the very insightful comments and answers here. I think Fabian made a great point:

It gets pretty annoying when you are used to that power and visit a different SE site and can't even fix some annoying typos as you don't have enough rep there.

I have experienced this myself on the Unix, Gaming, and Apple StackExchange sites, and have even lamented about it in chat. So I can relate to that, especially since I doubt I will ever have 2,000 reputation on gaming.

Waffles has now answered authoritatively on what is actually happening and it is a big relief to me to know that this is happening in a much more controlled, measured manner than I originally thought. It's also good to know that tweaks are being made, so I don't know if these suggestions are even relevant, but here's my thinking:

• I am still bothered by the zero barrier to editing. I'd really like to see some minimum rep required, even if it's 50, or maybe even 20. I think this will cut down on the potential for extremely low-quality edits

• Tim Stone made a great suggestion in chat that, if a given user continually posts edits which are rejected, maybe they should be rate-limited from editing. Or maybe editing should be a privilege which can be lost if abused. Again, I am thinking about curtailing abuse here.

• Also, I love badp's suggestion of a "merge" function, because having to wait for someone else to approve an edit just so I could make a second edit really bothers me.

Answer 5

Opening up editing is mostly a good thing.

A lot of times, I have seen a few posts on some SEs with basic errors like typoes or improper punctuation, or subpar formatting. It shouldn't take 2k rep to fix that.

I'm not sure it'll do much good for the intended purpose -- preventing information rot -- as edits aren't the end-all be-all solution for information rot.

---

<sidebar>

Expanding a bit on this point: there's so much more to editing than information rot. I usually edit typos, formatting; I rarely add a paragraph to complete an answer. Also, there's much more to information rot management than editing.

Basically I see more than one level of rot:

1. The question is now completely meaningless. Say, a question about the App Store approval proces after it's closed. What would an edit solve here? Put a blockquote h1 at the top of the post screaming about the news? Close as localized instead.
2. An answer is now plain wrong. Say, a person asked how to make minecarts go really fast and a user answered to exploit a glitch and make a so-called "booster". This glitch now gets fixed. What would an edit solve here? Put a blockquote h1 at the top of the post screaming about the news? Post a comment and downvote instead, and maybe comment on the question too so the asker may revoke the checkmark.
3. Part of an answer now is plain wrong Cha-ching, I hear you say. Yes! Edit away! ...or use the awesome license we have and post a new answer, quoting the rest with attribution. Multipart questions aren't good questions anyway.
4. An answer is now outdated in a detail. Say that, on Unix, you ask how to make a process a daemon, then the answerer posts some sample Python code which happens to use os.popen, which is being deprecated in favour of the more Pythonic but more complex subprocess module. Except, you know, it really isn't the point of the answer and subprocess adds additional complexity that would only risk making the post not as clear, and os.popen hasn't been removed yet anyway. Sure, an edit would be okay too, but personally I'd rather comment. The answer remains valid.

Yes! Editing can help. I just don't think it's the end-all be-all solution to information rot.

</sidebar>

---

In theory, I'd love having a mod-flags-like counter for edits pending review -- maybe only for those edits that already received an approval, just to speed the process up.

Yes, edit proposals acquiring a lock on the post is really a problem. A solved problem, too: just do merging. Maybe discard an anonymous edit that can't be automatically merged.

Answer 6

I think this is a great idea. There have been so many times I've hit the other sites and there's just no way to fix up a post.

My only concern would be that edits in a pending approval state shouldn't need two 2000+ rep editors to approve. After all we already have full edit rights. It would appear to be having an approval feature for the sake of adding a feature rather than streamlining edit workflow.

Answer 7

Wikipedia works without even requiring that users register and they cope just fine.

Personally I think that the Wiki part of the site is under emphasised - we need more edits!

Answer 8

Editing someone else's post used to be a privilege that you repped up your hard smoked EXP by. You craved for that sweet post that rained the magic and arbitrary points to level you up to a platform of casting edits on posts.

It was something to strive for, to add extra payback to the community in addition to the quality answers and questions you were posting.

The "everybody has a valid voice (pending)" edit feature proposed makes it seem more like a carnival where they hand everybody the keys to the ferris wheel just for turning up and kicking around some dust.

No, those pants you have to work your way up to wear. You can't wear daddy's pants. Even if he's asleep. When you climb over him and bury his body in that tracksuit you bought at the op shop, then you can.

Answer 9

I'm not liking the "review edits" feature at all. When I click the tab, I have to consider whether someone else's decision to reformat code, or make a content-neutral neutral stylistic edit to a title is worth approving or not. The decisions to be made here are just too trivial. For example

• Someone edited the a question title so that it would be in the form of a question -- but didn't otherwise change the content. There are a lot of questions whose titles aren't in the form of a question here, so I hit reject.

• Someone else edited a code sample to reformat the code just a little (move a bracket to somewhere more in line common coding conventions, and remove a linebreak from inside a parameter list). I think I hit the refresh button on this one and saw that someone else had already answered it.

I don't really care whether these changes were accepted or rejected. I think that someone who's accumulated 2000 rep will realize that that particular edit is unnecessary, and won't propose it. But if they've been around long enough to accumulate 2000 rep and decide to make that change, let them. Don't bother the rest of us with the details. It's just too trivial for me as a reviewer to actually care about.

(And there's a serious race condition for making a decision where you'll frequently be beat by someone else unless you make a split-second decision.)

Please go back to the old system where users under 2000 reputation just can't edit anything.