15

Lately I've seen a lot of small edits here on Meta to posts that are obsolete -- for example, past Winterbashes, site functionality that has since been changed, and bugs that were years ago. I've even approved a couple of those edits without noticing the age of the post. The posts then get bumped and sometimes pick up new answers and comments, presumably from other people who didn't notice that the issue being talked about isn't relevant now.

Is this a problem? Should we apply any friction to editing and/or reviewing? I wouldn't want to prevent edits (we do want to continue to maintain our content), but maybe the combination of a small edit (e.g. typos) and probable obsolescence could add a notice to the edit and review pages? Just something like "This is an old post (or resolved issue); are these small edits worth bumping this post for everybody to see again?". (Not those exact words; "bumping" is possibly jargon.)

This is a discussion, not a feature request. I don't know if all these bumps are actually a problem, hence the question. Several sites on the network are approaching (or already past) being a decade old; there are a lot of old posts out there, and apparently some people like to look for and fix typos, "thanks"s, and the like.

Improve this question
10
  • Related, not a dup since this is a discussion: meta.stackexchange.com/questions/122567/….
    – Pika Supports Ukraine
    May 10, 2019 at 16:39
  • 6
    I shouldn't be discouraged from making an edit (if I want to) just because it would result in bumping the post. Such discouragement implies that it's a bad thing to bump the post. If that really is the case, then it shouldn't be the edit that's discouraged—but the actual post-bumping behaviour for such edits that's changed. (Which goes to the feature request mentioned in the other comment.)I think that If you want this to be a discussion about the worthiness of bumping minor edits, you should change the way that you've framed the question.
    – Jason Bassford
    May 10, 2019 at 16:49
  • 1
    @JasonBassford not discouraged so much as "look again at the situation; are you sure?". As I said, we do want to maintain our content -- not suggesting we should let old stuff rot -- but if you're going to bump a long-dead post, try to make it count or just let that small typo remain. (And no, I don't want this question to be about when/whether to bump; that's covered elsewhere. This is about trivial edits on obsolete posts.)
    – Monica Cellio
    May 10, 2019 at 17:11
  • 1
    @MonicaCellio I seen no difference between allowing trivial edits on new posts and allowing trivial edits on old posts. If bumping isn't a consideration, then it seems to be a red herring. Why bring it up?
    – Jason Bassford
    May 10, 2019 at 17:21
  • 1
    Bumping is a consideration, but I'm not asking whether we should, or saying we should, change bumping behavior. I'm asking whether we should apply a little more care before making the edit in the first place. Silent (non-bumping) edits bring other problems that shouldn't be the focus of this question. (Unmonitored abusive edits have already been brought up elsewhere.)
    – Monica Cellio
    May 10, 2019 at 17:24
  • @MonicaCellio Again, I see no difference at all between trivial edits to new posts and trivial edits to obsolete posts with respect to edits themselves. If bumping is a concern, then the focus should be on the bumping; if it's not a concern, then the focus should be on the editing itself. I can't see what bearing the age of the post has on anything if you aren't making a normative statement about bumping. Unless, bumping or not, you simply believe that obsolete posts should never have trivial edits made to them. And that would be something different. Why shouldn't they?
    – Jason Bassford
    May 10, 2019 at 18:00
  • 4
    Because the probability that you are actually improving anything goes way down if the post is old and obsolete. Regardless of the mechanism (bump, review, something else that Robert will come up with), humans invest time in reviewing edits. We shouldn't waste their time. Again, what I was asking about is a gentle nudge, not a restriction, to prompt the thought: "is this edit actually helping the site?".
    – Monica Cellio
    May 10, 2019 at 18:02
  • Re: "...humans invest time in reviewing edits. We shouldn't waste their time." In that vein, I'd say we should rather focus on making reviewing edits easier. On most sites there's a group of "trusted users" whose actions I wouldn't bother to double check, but I'd definitely like to have quick overview of the substantial recent edits made by newer users (which the current UI doesn't offer).
    – user437611
    May 10, 2019 at 21:09
  • @MonicaCellio But a trivial edit, although trivial, is still an improvement of some kind. It should be up to each person to determine how they spend their time, not us. And if you're going to object to trivial edits as a waste of time, then the same principle should apply to new posts too.
    – Jason Bassford
    May 10, 2019 at 21:53
  • 4
    We should certainly do something about it: there have been days where the edit-bumping has rendered the front page of Meta effectively unusable.
    – Mark
    May 11, 2019 at 2:24

3 Answers 3

Reset to default
25

Interesting discussion I've noodled around with many, many times. Personally, I think we might be looking at this problem from the wrong end of the spectrum (in my opinion).

How we handle routine edits has some big UX gaffs. Bigger picture: these systems need broader participation, not less. What we really should be looking at is how we can allow routine, low-friction edits… without being so damn imposing and disruptive to everyone else in the process.

Bumping everything all the time is the problem here. A few caveats aside, every time I want to improve a bit of grammar, or clarify a run-on point, it starts to feel like I'm committing a low-level assault on the user and the system. Capitalizing a few sentences puts everyone on high alert (a post bump for everyone); breaking up a wall of text notifies the author that I'm "touching their stuff." It's like I'm declaring an act of congress where I need to first justify if the post (in your example) or the improvements are "substantial enough" to justify my existence. It's a bit nutso, if you think about it.

We certainly need to maintain a level of accountability when folks edit something, but that accountability should be proportionate to the actual risk of who is doing what. The system could certainly do a better risk assessment following any change:

  • What is the experience and history of the user making that change?
  • What did they do?
  • Could we notify a few people to make sure changes don't go completely unnoticed?

Certainly nobody should have zero-accountability with "silent edits" — but in a bigger picture where we're trying to find ways for more folks to participate in these sites, while simultaneously folks feel there's nowhere they can engage, adding yet more friction to the most harmless, non-consequential corners of the site doesn't sound like a productive path to follow.

We should be heading down a path of improving the systems and behaviors we haven't looked at in eons… so these well-meaning activities by well-meaning users are more productive. But not by systematically blocking users trying to do the right thing

— in my opinion.

Improve this answer
8
  • These are good points. We do need accountability (you and I have both seen high-rep users rage, so "make silent edits" should never be a rep-based privilege), but if we can tailor the visibility of certain edits without sacrificing accountability, that'd be great. (I assumed that would be a hard problem.)
    – Monica Cellio
    May 10, 2019 at 17:35
  • 2
    @MonicaCellio It's not a hard problem if I can get some designs across regarding a type of meritocracy where users get to do more of what they're good at while incrementally throttling someone back where they just don't get it. Reputation is only one variable. "Who is the user? What did they do? Are they (generally) pretty good at this stuff?" Even the most "trustworthy" user would never get 100% "silent edits" (as you put it). But personally I don't need to know what Monica Cellio is up to every time you edit a tag or fix a title. Spot checking important, but generally I trust you <grin>
    – Robert Cartaino
    May 10, 2019 at 17:48
  • 1
    On Wikipedia they have the concept of Minor Edit you can choose to be included or excluded about notifications for minor edits. In my experience there it addresses the concerns that Monica raises.
    – James Jenkins
    May 10, 2019 at 17:53
  • @JamesJenkins That has been declined, and a similar request to implement that option for a different purpose (reopen queue bumping) has also been declined.
    – Sonic the Anonymous Hedgehog
    May 10, 2019 at 18:46
  • @RobertCartaino There was some experimentation done with the home page a while back to make it not simply be based on last activity, but that ended up being put to rest due to resources IIRC. Maybe we can start that up again?
    – Sonic the Anonymous Hedgehog
    May 10, 2019 at 19:06
  • 5
    Thank you for at least acknowledging that the root problem lies in the UI. Minor edits are not bad if they leave the posts even slightly better than before (fixing typos and re-tagging fall in this category). Lots of well-intentioned users receive disproportionately high amount of flak for bumping posts due to minor edits and soon lose interest in editing altogether...that's definitely not what we want.
    – user437611
    May 10, 2019 at 20:21
  • 4
    [cont.] In fact, I find the whole "main page" concept flawed. Why do even need to display to the visitors who modified what x minutes ago? That information is mostly needed by the active users (who're interested in moderating the site), not the visitors and the irregular users. Previously I did propose changing the default "main page" to the "Newest" questions page before (in chat), but not many people seem to like that idea either. There must be a better alternative.
    – user437611
    May 10, 2019 at 20:26
  • @Blue I'm speaking as an active user, one who wants to see new answers and substantial updates, not just the newest questions. "Activity" is all or nothing; either you take that along with the trivial edits or you miss out entirely. We need some solution to this that doesn't invite abuse ("hey, I'm gonna vandalize the site, but I'll check "minor edit" so no one will notice"). Maybe custom filters will provide some sort of path here.
    – Monica Cellio
    May 12, 2019 at 3:13
7

I was discussing this with a fellow mod - and to a certain extent, part of the problem is 'unique' to MSE, being a high importance, low volume site.

There's a handful of users with pet peeves - some folks have been doing spelling fixes for years, I know one user's focused on fixing mentions of http to https for some reason, and a few users seem to be very focused om curating community wikis.

As someone who worked on a pet peeve before (getting rid of unwanted tags) - I tend to feel doing these sort of tasks with low friction is a squishy problem. It involves being aware of the front page (and being mindful that sometimes your name is on every single posts for edits).

The friction has to be folks thinking "does this help?" and "dangggg... I just flooded the front page" over hard driction.

The difficult bit is of course, how to tell folks who are involved without crushing their spirit (unless they're the sort who keep doing it anyway). We do want people to edit, but also be mindful of the front page and whether they're really helping.

Improve this answer
3
  • Keep in mind that there was a prior attempt at making the homepage not simply be the last activity, but rather "recent" questions gauged based on interest. The experiment is technically still up and running, see meta.stackexchange.com/home/recommended/debug.
    – Sonic the Anonymous Hedgehog
    May 11, 2019 at 4:53
  • 3
    But its just as much about the user as the mechanics. Part of the difficulty is working out the right motivation to moderate behaviours that are positive in moderation but non positive in excess. I mean, sure its showing the last activity, but its also multiple folks involved in activities that are slightly less core - things other than posting, on posts that are of significant age.
    – Journeyman Geek Mod
    May 11, 2019 at 7:04
  • 1
    We had a similar circumstance a while back on Interpersonal Skills where a user was doing a lot of tag edits on old posts which was flooding the front page. I remember seeing a mod poke them about it in chat and I believe that they reached an agreement to do no more than 5 such edits per day so as to not overload the front page with too many old questions.
    – Rainbacon
    May 11, 2019 at 16:33
-2

I joined this site after a few years on other sites for two reasons (a) to be able to contribute to the development of the network by voting, and (b) to inform myself better about the operation of the network. However I find the many of the posts which appear on the main page as active are edits to old posts which I suspect are obsolete or relate to the epoch when this was the meta-site for StackOverflow alone.

One possible way forward would be to use the tag when it applies. Perhaps we could encourage wider use of this tag so that when people edit an old post they add this tag when appropriate at the same time. This would mean that people could still improve old posts but I could add that as an ignored tag.That would not address the problem of StackOverflow only posts unless they are actually obsolete but I can live with that

Improve this answer
1
  • 3
    Rather than adding a tag (that people would have to learn the semantics of), let's close them when applicable. One of the close reasons here on Meta is "this problem can no longer be reproduced". Even if it's a discussion, if it's a discussion about SO (not the network) from 2010, I don't have a problem closing with that reason.
    – Monica Cellio
    May 12, 2019 at 17:08

You must log in to answer this question.

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