183

TL;DR: We've put together a code of conduct (CoC) that is a bit more comprehensive than our existing be nice policy because we feel that our current policy isn't meeting our needs.

Some background, our reasons for doing this and a link to the draft (Google Docs | GitHub) can be found below. We'd like your feedback by July 11, 2018.

In the beginning, there was 'Be nice'.

From the earliest web archive snapshot of http://stackoverflow.com/faq (circa September, 2008):

Be nice.

Treat others with the same respect you'd want them to treat you. We're all here to learn together. Be tolerant of others who may not know everything you know. Bring your sense of humor.

And that's all we needed when we started. Stack Overflow began with a large group that were fairly avid readers of Coding Horror and Joel On Software. While many of us sort of knew each other from interacting on both blogs and via mediums like Twitter, what can be said is, through our shared interests, we had much more in common with one another than we had differences. For many of us, Jeff & Joel spoke to the majority of what we valued and were passionate about.

We had essentially one rule: be nice. But, it was seldom, if ever that we actually had to enforce it. We were a group that, despite numerous geopolitical differences, essentially behaved the same through the handful of circumstances one might encounter while using the site.

Sure, the mechanics of the site encouraged sincere collaboration, but many missed that we were already a community that (mostly) agreed on the fundamentals of many things needed for our group to self-govern; it wasn't just the software and gamification that was creating a runaway success.

From our bootstraps, we showed tough love by editing and voting ruthlessly, but we were a small enough community that we could sweep away criticism by accurately stating this is how we like it. We and the site were something new, and people wanted to be part of both things. And we grew, wow did we grow; what an amazing machine where one could plant a programmer and grow a good communicator.

And then, there was 'Be nice'.

Maybe in context, we could call it 'Be Nice(r)?'

We expanded our policy a little bit, because it was simply too ambiguous. In fact, a key balance point that we struggled a bit to find in our initial RFC for the re-write was just that, breadth vs. depth:

Some people wanted more detail and examples (for clarity), others wanted less (for broader applicability).

We knew we needed to create something where folks embraced the intent, or spirit of the document, rather than using the document as a checklist of stuff to avoid moderators, while giving some examples of what's bad for use as landmarks. But, the voice of the policy was still better not do that rather than here's how to not do that.

Since then, we've absorbed an enormous amount of feedback from people that have interacted with our sites for the very first time, some stuff has been pretty consistent across rather vast groups of people.

We fell short in our 'Be nice' re-write in the following ways:

  • We needed to write for the best of folks in our community. Off-putting things tend to mostly come from folks who will probably only ever blow their top once. We have moderators to deal with the tiny fraction of people that never care about rules, so our code of conduct needs to mostly resonate with the overwhelming majority of people we really want to keep. We need less over-posturing for troll dispatching and more guides to help decent folks avoid more common pitfalls.

  • Codes of conduct help identify your community to the outside world, and help people decide if your community is a place where they'll feel safe and flourish. Our expanded 'Be nice' policy doesn't hold up to much scrutiny if you're fresh from seeing a bunch of nasty comments go unmoderated because they weren't technically rude.

  • We're at a point where 'Be nice', which first originated in a group where we pretty much already had a strong social contract, just isn't enough. What worked well for hundreds isn't working well for millions, and we need to write something more comprehensive.

But our policy is short on other things, too. What we need is a formal code of conduct that's similar to what free / open source projects and even conferences use in order to set behavioral expectations and norms.

The difference between what we have now and what a formal code of conduct would look like is best expressed by simply showing you our first draft of a formal code of conduct.

For your feedback, we're presenting our initial draft of a real Code of Conduct.

We'd very much appreciate your time in taking a moment to read our draft (Google Docs | GitHub). If you have some cycles to spare, we're specifically seeking the following types of feedback:

  1. Even if you don't agree with all of it, do you feel that this is a reasonable code of conduct? Does it affect your recommending Stack Overflow as a resource in any way? How so?

  2. Is anything in this document, including its purpose, unclear to you in any way? Are there any instances where you'd suggest alternate language or copy? Where? Why?

  3. If you're a long-term contributor, how does this document impact your feelings toward new users and their needs, if at all? How does it impact your feelings toward the company? Or, more broadly, did this document inspire any metacognition at all?

  4. If you're someone who wants to contribute more often, or have felt yourself sort of stuck on the fringes, how does this document change your perception of the site, if at all?

We're open to all feedback.

We understand that not all of you feel like this is necessary, and we understand that truly being inclusive means making sure everyone has seats at this table too, as long as we can possibly extend one. That means we need a civil dialog, so we're asking for any negative feedback to be respectful and honest. Please remember, we're listening to you as well as many, many people that simply aren't comfortable coming here, and a big part of our goal is going to remain to include them.

We'd like feedback by Wednesday, July 11, 2018, however this question will allow for feedback until a final draft is released, at which point this question will be closed as a duplicate of that one. We'll monitor for new answers as long as this remains open.

Please, post an answer and avoid comments if you have anything of substance to say.

We'll respond to comments that are readily and practically answerable with a few dispassionate sentences, but if you've got anything even remotely substantive to say, we're asking that you leave an answer. Using comments instead is actively harmful because it requires us to spend more cycles moderating and archiving than actually engaging.

In closing

We'd like everyone to remember that we're talking about ways to be nicer to one another - so please, let's have a conversation about this in the spirit of that intent.

We're in the process of gathering a lot of feedback through UX research and having a bunch of conversations with interested groups to gain a variety of perspectives. The more open and specific you can be about any concerns that come up as you read this draft, the better our next draft will be. For instance, it's fine to say "that scares me but I'm not quite sure why".

Thanks for reading this far, and let us know what you think!

5
  • 3
    There's an old saying, "Children don't hear anything you say, but they see everything you do." Mission statements fall into the "anything you say" category, and they're not really directed outwardly. I've seen so many organizations agonize over mission statements without considering that nobody outside the organization cares about, heeds, or even bothers to read them.
    – Robusto
    Commented Jul 19, 2018 at 12:51
  • 3
    Curious: if you wanted answers by July 11, why is this just now being featured in the "FEATURED ON META" block of SE sites? Commented Jul 19, 2018 at 14:10
  • 1
    @GalacticCowboy It has been featured since July 3 ... I saw it the next day on the handful of sites I regularly check on - can you tell me where you didn't notice it until now? That's .. definitely a bug .. if so. If you look at the views over the short time this has been posted, word really got out. Kinda baffled as to why you didn't see it.
    – user50049
    Commented Jul 19, 2018 at 14:21
  • 1
    Maybe I wasn't paying attention, but I've been on Stack Overflow and a variety of the hot questions pretty regularly. Commented Jul 19, 2018 at 14:25
  • At this point in time I would like to ask that this whole CoC be reconsidered. IMHO it's led SE down a dark path and has made SE Inc. (and maybe even us) less nice.
    – einpoklum
    Commented Oct 7, 2019 at 22:24

111 Answers 111

29

Reporting and Enforcement

...

  • Address it directly. If you’re comfortable addressing the person who instigated it, let them know how this behavior affects you or others. You can also remind them of the Code of Conduct. If the behavior escalates, disengage immediately -- staff or moderators will take action.


Can you clarify how to do that and when? Via comment would conflict with When I should comment?.

I'm surprised that meta pages were never mentioned either. They could provide a platform for discussing problematic cases - maybe more suitable than the comment section below someone else's post. What role do they play?

6
  • Thanks for pointing this out. We'll have to take another look at the commenting guidelines. Re: Meta - opted not to mention it in this iteration precisely because we'd need to clarify its role & set people's expectations about going to Meta - which is a hefty topic in itself.
    – Donna StaffMod
    Commented Jul 3, 2018 at 21:28
  • 4
    @Donna I think that paragraph makes perfect sense for chat. Less so, if at all, for main sites though. Perhaps retargetting it to chat only would be a good idea?
    – terdon
    Commented Jul 4, 2018 at 9:18
  • I see this paragraph a bit as a contrast to being welcome and no put-downs. When you are having trouble with the person, personal addressing of it can often make it much worse. Telling people to clarify it between themself can lead to flame wars, which could be avoided if some third person would help them to cool down instead of shouting at each other making it only worse. And in the end both will tell you "the OTHERperson started it I tried to tell THEM to stop and THEY made it worse". This does not help any of the persons in the conflict.
    – allo
    Commented Jul 5, 2018 at 12:15
  • @Anne Why did you delete your question, posted today (March 24, 2019) on meta.se? I was working on a "fleshed out" answer with examples, as you asked me to do, when you abruptly deleted your question.
    – amWhy
    Commented Mar 24, 2019 at 15:29
  • @Namaste Sorry for making you waste your time! Commented Mar 24, 2019 at 15:34
  • No problem, @Anne. BTW, I did think it was a good question, at least with respect to wondering how "philosophy" might be a valid tag on MSE, if ever! Cheers!
    – amWhy
    Commented Mar 24, 2019 at 15:43
29

Language is bloated; leaner prose is recommended

I'll make two concrete recommendations.

  1. Change

    Our mission is to build an inclusive community where all people feel welcome and can participate, regardless of expertise, identity, or language

    to read

    Our mission is to build an inclusive community where all people feel welcome and can participate.

    Comment: (1) stronger statement when unqualified, and (2) "regardless of expertise" looks to be completely opposed to the basic SE model.

  2. Remove or significantly rewrite this one:

    No subtle put-downs or unwelcoming language. Regardless of intent, this behavior can have a significant negative impact on others. For example, saying “You could Google this in 5 seconds” is a subtle put-down.

    Comment: while this bullet aims at a worthy goal, it is "soft" rather than "hard" as guidance. Too subjective.

    As was discussed in the "let's hold comments to the same standard" Q&A on Meta.SE, within the set of users that is all SE users there are varying abilities with language, varying writing styles, and a non-trivial number of people who use English as a second, third, etc language. (Not to mention different cultural assumptions ...). The term "unwelcoming language" needs to go because it is nebulous at best.

    Removal of this point, since it is redundant, is my suggestion. The CoC goes into considerable detail later on behavior that is not within bounds. If for reasons of required content or completeness it can't be removed, then it needs to be rewritten to be less subjective.

    At the very least, this bullet point needs to be the fourth of four bullet points, not the first of four. The name calling / discrimination / harassment bullet points are all of greater weight.

  3. Some other good recommendations for leaner, more concise language are in this well presented answer by @Vogel612's Shadow

27

No discrimination of any kind. This includes any language likely to offend or alienate people based on (but not limited to): race, gender, gender identity or expression, English fluency, sexual orientation, disability, mental illness, nationality, neurodiversity, physical appearance, body size, or religion.

English is not my native language and maybe I'm nitpicky, but people cannot be discriminated against on the Stack Exchange network in ways that are public or actionable. Why? Because discriminate against means to take unfair action. Suppose I downvote all christians. Or Close-Vote all women. Or never edit post into shape that are from gay people. Or make any other kind of distinction. Even if that happens, it's impossible to catch and/or proof.

I think what you mean is no insults, bullying and harassment. If someone calls me the N-word, that's not discrimination. The site works fine for me, I get answers from other people, I can vote up or down, no functionality was lost. But it's obviously insulting and probably harassment.

So this point should really by a subcategory of "No harassment. This includes: (...) as well as any language offensive to ..."

18
  • The point about offensive/alienating language being harassment, not discrimination, is a good one that has been brought up by others, but I disagree with this claim you make: "Even if that happens, it's impossible to catch and/or proof." The SE staff actually does have the utility to investigate something like that. If a user is downvoting people, for example, because they are women, that means there is some way on the system to know they are women. The voting patterns can be compared against this information. Commented Jul 3, 2018 at 17:59
  • 1
    @called2voyage So assume you find a pattern. Then what? Are we telling people how to vote now? My last information was "everybody is free to vote however they see fit" (as long as you have only one account doing the voting and don't serial-vote a specific other account).
    – nvoigt
    Commented Jul 3, 2018 at 18:30
  • 2
    A down-voting campaign on women would be form of serial voting. It breaks the voting system because it is voting based not on content but on user. Commented Jul 3, 2018 at 18:32
  • It may just be my interpretation, but I think this has more to do with targeting users based on affiliation, with disparaging words. More about people writing nasty things in comments, answers, and chat, than about voting.
    – apaul
    Commented Jul 3, 2018 at 19:00
  • 10
    I'm just saying "discrimination" is... fishy. If a person harasses others on the street, you don't see a cop being like "Sorry sir, you have now harassed 58 slender white women in the past 3 hours, if I don't see you harassing a brawny black dude by 6 PM I have to arrest you for discrimination". I would simply hope we go with the very first harassment to nail the offender, instead of patterns and statistics of what might be discrimination.
    – nvoigt
    Commented Jul 3, 2018 at 19:19
  • 2
    @apaul nvoigt's point is that what you describe is usually referred to as harassment, not discrimination. Commented Jul 3, 2018 at 19:20
  • @called2voyage harassment is bad, harassment for reasons tends to be an extra level of bad and is probably worth escalated consequences.
    – apaul
    Commented Jul 3, 2018 at 19:24
  • @apaul I agree, but for example in my workplace harassment for reasons is still listed in a separate section than discrimination. Both discrimination and harassment sections have the list of the categories that you should not discriminate against or harass on the basis of. Commented Jul 3, 2018 at 19:25
  • @called2voyage kind of a 6 vs half dozen then. Discrimination would be a biproduct of harassment, right?
    – apaul
    Commented Jul 3, 2018 at 19:28
  • @apaul Not necessarily. Someone can discriminate without harassing someone due to subconscious bias. Commented Jul 3, 2018 at 19:30
  • 4
    Matter of fact people can discriminate against others on purpose (just think not hiring women on purpose) and be super-polite about it. Just because somebody doesn't swear or insult people doesn't make it ok to discriminate against others. But posting racial slurs or personal attacks based on any of the mentioned attributes is not discrimination, it's harassment.
    – nvoigt
    Commented Jul 3, 2018 at 19:32
  • Half of the premise of this answer seems untrue to me, at least in principle. If I were to declare on Meta that I will henceforth not accept any answers from white people, and then follow through, that would not constitute "harassment" of anyone by any ordinary interpretation of the word but would be a clear-cut, detectable case of discrimination. I'm not sure whether we've ever had a case like this in the entire history of the site, but it seems fair enough to declare in the CoC that, hypothetically, such conduct would be against the rules.
    – Mark Amery
    Commented Jul 4, 2018 at 8:10
  • 4
    You cannot guess from the user what is race or gender. In fact, what I have suspected, and a couple of times see proof (reflection on photos) is users posting with a female named and photo on the belief they will attract more help. Commented Jul 4, 2018 at 9:00
  • 2
    @RuiFRibeiro For some sites on the outer rim of SE, giving that information in the post is needed for a good answer. For example questions on "Parenting" mostly start with "As a mother of" or "As a father of". And IPS.SE... well, if you have interpersonal problems, it gets personal, there is no way around that.
    – nvoigt
    Commented Jul 4, 2018 at 9:08
  • 2
    I think this paragraph is trying to say "no hate speech" without getting hung up on the highly charged term "hate speech". And indeed, I'm glad it does avoid that term, but I agree that the content is mostly covered already as harassment. What isn't covered there would be covered by broadening "No name-calling or personal attacks" to simply "no put-downs or attacks", i.e. of individuals or groups. Commented Jul 5, 2018 at 19:52
27

The Our Expectations section sets up a dichotomy between people who are here "to help others" and people who are here "to get help" that I think is unhelpfully misleading about the site's core purpose.

To quote a recent comment of Shog9's:

a lot of the problems we have with Stack Overflow arise from how poorly we describe it to folks.

It's a reference site. That's how almost everyone uses it, and how it is primarily useful. You don't go to Stack Overflow to learn anything, you decide to learn something and sooner or later end up on Stack Overflow to fill in the gaps in your knowledge. I'd guess somewhere north of 90% of everyone who has used Stack Overflow uses it that way.

But we describe it as a place to ask questions... As though that's anything but a means to an end.

This point goes hand-in-hand with others' feedback about the mission statement in the CoC. We're meant to be "working together to build a library of detailed answers to every question about programming". Question askers are supposed to be asking themselves whether the question they want to ask is a useful addition to that library; that is the basis upon which other users will vote on their questions. If it is a useful addition, then the asker can very much claim to have "helped others".

Indeed, an asker whose question gets a good answer may not have personally received any "help" from it at all. For all we know, they already implemented a crappy workaround for the problem they were working on, shipped it, and moved on to other things, all before ever getting round to asking on Stack Overflow - and then posed their question purely out of altruism to help the next programmer to encounter the same issue.

The dichotomy presented by the CoC does a disservice to askers in several ways. It presumes that they are asking for selfish reasons, when they needn't be. It devalues their contribution to our library by presupposing that only answering questions can "help" anyone, when in fact there is great value (and a fine art) to creating tightly-specified, concisely-answerable, broadly-applicable, easily-Googleable questions for answerers to hang their answers upon. But perhaps most importantly, in the context of our current problems with both question quality and civility, it sets up the reader with false expectations about what we expect from question askers here, by making us sound like we are primarily a helpdesk when in reality we are primarily a library.

I'd rather all the language about "helping" was culled, and the primary goal of Stack Overflow* - building a library of questions to help future readers - was emphasized. Doing otherwise sets up askers for a negative reception that they won't understand.


* It occurs to me after posting that I've completely neglected all the Stack Exchange sites other than Stack Overflow in this answer. However, I think this commentary applies to most of them, besides the oddball ones like Puzzling.

27

Since you seem to be strongly focused on making people feel welcome, this information might be worth a new answer despite the existing 103 existing answers:

If this Code of Conduct is adopted in its present form, I would feel less welcome on this site.

To avoid misunderstandings, I want to emphasize that I strongly agree with important parts of it (basically everything about discrimination, harassment and personal attacks) and that I'm well aware of the importance of taking the perspective of others (e.g. those who feel uncomfortable or discriminated against) and of not imposing our perspective on others.

However, the draft sets this consideration, as important as it is, as absolute and takes an extreme position that I believe would be as harmful as it's well-intentioned.

If someone points out that your behaviour is making others uncomfortable, stop doing it. Sometimes, people unconsciously say things that negatively affect others. Even if this wasn't your intent, apologize and move on.

It's very important to be aware that we can unconsciously make others uncomfortable or affect them negatively. It's also very important to understand that we're not the sole judge of whether this has happened or should have happened. It's good that you're so aware of this and are emphasizing it. But you're throwing out the baby with the bathwater. Making just anyone who "points something out" the sole judge isn't that much better than making yourself the sole judge. If this item is taken literally, it would force me to apologize if I justifiably asked someone to refrain from a certain behaviour and someone "pointed out" that this made that person uncomfortable. That way madness lies. Not all forms of feeling uncomfortable can or should be avoided. Of course many forms can and should be avoided, and I appreciate that you're focusing on that, but it seems to me that you've somewhat lost sight of other, also very important considerations in the process.

No subtle put-downs or unwelcoming language. Regardless of intent, this behavior can have a significant negative impact on others. For example, saying "You could Google this in 5 seconds" is a subtle put-down.

That's neither subtle, nor a put-down. It's a criticism of behaviour that's worthy of being criticized. If you want to discourage this particular form of expressing that criticism because you believe there are better forms of expressing it that achieve the same effect while making fewer people feel less welcome, that's cool – but then you should acknowledge the constructive intent of the remark (instead of dismissing it with "regardless of intent") and concretely point out how you'd suggest improving on it. In its current form, this item reads as if there were no reason to write such comments, no problem with people spamming the site with questions that they could have answered in 5 seconds by googling – as if the only goal worth aiming for is to make people feel welcome and comfortable, no matter how anti-social their behaviour is. If you deter people from deterring people who should be deterred, that may do more harm than all the welcoming atmosphere can do good.

I'm a political activist; I've spent a large part of my life trying to understand other people's perspectives and deconstructing my own. I'm really not writing this because I can't relate to the ideas and principles that I feel have inspired this draft. To the contrary, I strongly believe in many of them. But I also believe that it's excesses like this draft that are giving them a bad name.

Edit:

Some more thoughts on the googling part: Further down, in the table, you use a different example for a subtle put-down: “This is obvious, just Google it.” That you're effectively equating these two comments illustrates the lack of differentiation in the draft. “This is obvious” is indeed a put-down (though, again, not a very subtle one to my mind). It takes what's obvious to me as a measure for what should have been obvious to someone else; and it implies that the other person isn't intelligent enough for it to be obvious to them (whereas it might just not be obvious to them because they lack some piece of prior knowledge).

This is completely different from “You could Google this in 5 seconds.” We're not talking about people here who might have a disability that prevents them from accessing the Internet. The person was able to find out about Stack Exchange and to figure out how to post a question here. It's not conceivable that they've never heard of search engines or are incapable of using one, yet capable of posting a question on Stack Exchange. I'm also not talking about a case where it might require some level of expertise to know the right search terms or to interpret the search results. There are often questions that would be answered by simply googling the central term in the question and reading its definition on Wikipedia that comes up as the first hit. There are also other indicators of complete lack of effort: Questions whose title is nearly identical to the title of an answered question, where you can check by entering the title in the “Ask a question” form that that question was suggested when the person asked the question, yet they don't even mention it, let alone explain what it is in the answer that they don't understand. Of course one can overlook things; I'm not saying that it's OK to insult someone for this; but if it's no longer OK to point out that the question could have been resolved with minimal effort and should thus ideally not have been asked, then that in my view is a problem, both in terms of the harm done to the site and in terms of me feeling welcome here with the judgements that I do sometimes have about what sorts of behaviour I find worthy of criticism. (Note that I'm pointing out that you're making me uncomfortable, so you should stop and apologize. ;-)

4
  • 16
    The "sole judge" argument hits the nail on the head. It's one of the things that were wrong with Jay Hanlons earlier "welcoming" blog. And now some SE staff wants to enshrine it in a new CoC. There are two sides to every story; the person who feels uncomfortable is not automatically right. Commented Jul 16, 2018 at 13:27
  • 9
    @S.L.Barth Not to mention that in any conflict, the person who first decides to summon down punishment on the other party by uttering the magic words "you are making me uncomfortable" needn't be the only one who is uncomfortable. At that moment the one thing you know - in an environment where a CoC like this exists and is well known - is that the person uttering those words has made a deliberate, conscious choice to exercise power over the other person and compel a public signal of their submission. That user, in my view, is the one who is likely to be toxic - not their target.
    – Mark Amery
    Commented Jul 17, 2018 at 14:37
  • 7
    (It also has the effect, of course, that people who are uncomfortable will no longer be able to voice that to their peers in good faith and try to have a conversation steered away from an uncomfortable conversation amicably, because the CoC makes just uttering those words the web equivalent of obtaining a restraining order against the other party. It robs the uncomfortable person of the chance to reach an amicable, non-coercive resolution to the situation without aggressing against the other party, even though that's likely to be what they actually want most of the time.)
    – Mark Amery
    Commented Jul 17, 2018 at 15:04
  • 2
    @MarkAmery: Plus one for pointing out the extreme polarization inherent here, that this kind of binary "everything is fine until suddenly someone's feelings are hurt and then bans all around" is actually anti-cooperation. Commented Jul 18, 2018 at 2:36
25

I like this. I'd like to tweak the language about enforcement a bit; right now it says "this is how moderators generally handle misconduct" and then describes warnings, suspensions, and expulsion. That's all good, except that we sometimes see users test limits. "Well, I did that and I didn't get a warning, so it must be ok -- how 'bout I try this? No warning? Great, then I can..." While often this is malicious, we have to assume that in a worldwide, diverse community where not everybody speaks English well, sometimes it's more innocent. Either way, we don't want to give people the impression that no mod response means they're doing fine; on larger sites we might not have noticed. (Which is why people need to flag, but I digress.)

I think all we need to clarify this is to inject a phrase like "aware of", as in:

This is how moderators generally handle misconduct that they become aware of:

It feels like that small change could prevent a fair bit of drama on metas and in chat.

I'm talking about the section on penalties, not about specific examples in the document.

6
  • Stackexchange has been pretty good about shutting down excessive rule lawyering imo
    – Magisch
    Commented Jul 3, 2018 at 16:10
  • 5
    @ChristianRau I think examples that are clearly examples and not definitions are helpful; whether the ones in this document need to be tweaked is a separate matter that I hope you'll address in an answer. (My answer isn't about examples; it's about how we describe application of penalties.) Commented Jul 3, 2018 at 16:11
  • 3
    Of course adding the awareness bit lessens the expectation of something happening because you explicitly mention that moderators might not be aware which isn't necessarily something you want to do in an enforcement section of a CoC. Especially the enforcement section should make the impression that breaking the rules has consequences, not that breaking the rules have consequences if and when arbitrary things apply.
    – Helmar
    Commented Jul 3, 2018 at 17:03
  • @Helmar that could also be "when they become aware", so the fact that you sneaked it past people a week ago doesn't protect you when somebody finally notices and flags it. Commented Jul 3, 2018 at 17:30
  • Sure, it's a question of emphasis. Should the document emphasize that "appropriate action will be taken when there is CoC violation" or emphasize that "things that look like the CoC is ignored might be due to moderators being human and not because it's accepted." I'd argue the CoC should emphasize the first one while the description of what moderators on SE are hold a valiant defense of them only being human.
    – Helmar
    Commented Jul 3, 2018 at 17:40
  • I know I've mentioned this elsewhere, but people of different cultures, people with conditions such as autism, and people who are not very fluent in English can wander into what seems like trying to push boundaries when they simply do not know.
    – user316129
    Commented Jul 6, 2018 at 16:42
25

Our Mission

To expand on other comments on the same sections, these are concerning to me.

Our mission is to build an inclusive community where all people feel welcome and can participate, regardless of expertise, identity, or language. Whether you’ve come to ask questions or to generously share what you know, join us in building a learning community that is rooted in kindness, collaboration, and mutual respect.

I thought out mission was to develop a searchable community of answers to common questions?

This doesn't feel like it describes the community as a whole, and is explicitly at odds with many of the largest communities which are predicated on referenced facts, not discussions.

We can be respectful of each other, but the rest of that statement just feels like something pulled from modern US politics and doesn't seem to really serve the SE communities in part or as a whole.

No discrimination of any kind. This includes any language likely to offend or alienate people based on (but not limited to): race, gender, gender identity or expression, English fluency, sexual orientation, disability, mental illness, nationality, neurodiversity, physical appearance, body size, or religion.

Ignoring the fact that the statement is pretty much cut and pasted from US law, which is by nature inconsiderate of the other nations that participate here, the broader scale impact of such subjective language is something I think needs more thought before including it in the SE Code of Conduct that applies to all the communities within Stack.

We are all just a little circular picture and a bunch of words on this site. Almost none of those identifying characteristics are known or possible to be known without our own act of making them known. Including it just doesn't make much sense here. We are as anonymous as we wish to be.

Each Stack Community is by nature, exclusionary and discriminating.

The English Language communities require posts to be in English, the German community requires posts in German, etc.

The SO community requires questions to be technical in nature with regard to software development. Questions about computer hardware, politics, management techniques, or puppies are closed as off-topic.

Are you going to shut down the board on Catholicism, or Judaism, or Islam because discussions there could offend someone from a different religion?

Are you going to shut down Language Specific boards because they exclude someone who is not fluent in the language?

Are you going to remove the downvote button, as receiving downvotes (for a poor quality question) is likely to make someone feel unwelcome?

I think the SE code of conduct should remain as general-principal as possible, and allow each community to further restrict the conduct as makes sense in that community. This one-size-fits-all type of Code of Conduct under-serves our community as a whole. As written, it disallows the more fact-based communities the ability to be concise without dragging feelings into everything; and I fear the discussion communities will be dampened because no discussion is without risk of offending someone.

10
  • 1
    "We are all just a little circular picture and a bunch of words on this site. Almost none of those identifying characteristics are known or possible to be known without our own act of making them known. Including it just doesn't make much sense here. We are as anonymous as we wish to be." True, but some sites such as IPS may require divulging such information from time to time if you wish to ask about yourself instead of phrasing it as "asking for a friend". Even if you are "asking for a friend", you could still get targeted for, say, asking about LGBT issues. Commented Jul 3, 2018 at 18:35
  • That doesnt hold true for most of the communities, as most of them are on topics where asking questions that would reveal that about yourself are irrelevant to the topic, and usually edited to be agnostic. To me, it sounds like that would be a consideration to add to some form of Community Specific addendum to the Code of Conduct for communities where those types of discussions are acceptable. In principal, I'd say the global CoC should be rather loose, and the various communities can tighten the loop where it becomes necessary.
    – Stephan
    Commented Jul 3, 2018 at 18:43
  • 1
    It should be a consideration because people are allowed to divulge information about themselves in their profiles. As long as they are allowed to do so, we should protect them and not allow for them to be discriminated against based on what they share there. Commented Jul 3, 2018 at 18:44
  • 8
    @Stephan: "We are as anonymous as we wish to be." And those who do not wish to be anonymous should not be exposed to any greater threat of harassment or other bad behavior than those who wish to remain anonymous. Saying that you could have been anonymous must never be a defense against these kinds of behaviors. Anonymity is a choice, and we should not punish those for making a choice different from our own. Commented Jul 3, 2018 at 18:46
  • I absolutely agree with both of you. I'm saying that there is not way to blindly discriminate on this site. We as users have to make ourselves known, AND someone has to harass us. I don't see how the verbiage changes the already present and enforced harassment policies. Maybe I'm not thinking about it the right way, but that's why we're here: to discuss :) thanks for the feedback.
    – Stephan
    Commented Jul 3, 2018 at 18:56
  • 1
    "I don't see how the verbiage changes the already present and enforced harassment policies." I don't think it does--it just makes it a clear part of the CoC. Commented Jul 3, 2018 at 19:01
  • 1
    With regard to "We are as anonymous as we wish to be", here's my answer to Is it possible to know the gender or race of another Stack Overflow user?. tl;dr: This can be interpreted as the suggestion people should hide themselves to avoid harassment - that's essentially blaming the victim. Most of the conversation here is about people who do identify themselves.
    – Kobi
    Commented Jul 4, 2018 at 5:23
  • There seems to be some impression that there is such a thing as perfect security. What is making me uncomfortable is there doesn't seem to be that acknowledgment. How far are you willing to go to ensure that there are never any victims?
    – Stephan
    Commented Jul 4, 2018 at 13:43
  • 6
    I also find it very interesting that everyone has zoned in on that one particular sentence and made no comment to the rest of my points. That hyper focus is just reinforcing my point about current US politics rearing it's ugly head in stack.
    – Stephan
    Commented Jul 4, 2018 at 13:47
  • 1
    I just find it funny that the description of discrimination, something everyone knows what it is without it being spelled out, is more discriminatory against non-English readers than the actual dictionary definition. When was "neurodiversity" ever a problem? How many English readers had to look that up? Also, discrimination is a subset of harassment, so technically, that section is redundant anyways.
    – Tezra
    Commented Jul 9, 2018 at 12:55
25

"No discrimination of any kind" means no voting, no flagging, no closing, no deletion, no account bans. It means anarchy.

I know "discrimination against wrong or harmful answers, plagiarism, spam, and foul language" isn't the kind of discrimination that is intended, but it clearly is included under "any kind".

How will the Code of Conduct be enforced without discriminating against violators?

23
  • No harassment. This includes, but isn’t limited to: bullying, intimidation, vulgar language, direct or indirect threats, sexually suggestive remarks, patterns of inappropriate social contact, and sustained disruptions of discussion.

One of these things is not like the others, not just in this item but in the containing bulleted list. Let's give it more emphasis somehow.

I'm talking about:

sustained disruptions of discussion

If you ask most people to talk about types of harassment, they talk about personally-directed stuff -- harassing a person (like stalkers) or a category of people (like bigots). But the disruption you're talking about here, I think, is the kind of behavior that derails chat, stirs up arguments in comments, edits destructively, posts inappropriately... the category that a community manager once explained to moderators as "breaking the site", even if no individual action is sanctionable.

As a moderator I've seen many users dance on that line, and seen mod teams struggle with how to handle them. The CoC won't make that struggle easier, but could we try to clarify this a little in hopes of reducing occurrences or at least better warning people? Maybe "sustained disruption" merits its own bullet point? Maybe there's some other way to work it in? I don't have an answer here, but especially as there's been some concern raised about rules lawyers, I'd like to clarify that if your target is a site or a chat room instead of a person or a specific group, that doesn't make it ok.

22

I like this policy a lot, especially the "Our Expectations" section. I think that this section does a great job of covering all-too-common mistakes that well meaning users can make, which end up sparking fights and alienating other users. I would be proud to strongly encourage all current and new users on the site I moderate (Mi Yodeya) to read this policy.

One point I'd like to discuss:

No discrimination of any kind. This includes any language likely to offend or alienate people based on (but not limited to): race, gender, gender identity or expression, English fluency, sexual orientation, disability, mental illness, nationality, neurodiversity, physical appearance, body size, or religion.

There are matters that are subject to intense disagreement between people who ascribe to different national, religious, or ideological identities, where simply stating the position held by one group, even from a NPOV, is likely to offend some large subset of members of another group.

Some examples (adjectives chosen to not align with any particular examples I can think of, but are realistic in tone):

  • "Religion R considers Religion S to be in the category of 'wackiness.'"

  • "Religious Text T refers to Practice P as 'outrageous.'"

  • "Nation N considers Nation O a 'rogue nation.'"

  • "Nation N forbids Practice P based on their Ideology I which considers people who do P as akin to 'thieves.'"

Given that many people with certain identities are likely to be offended by each of these statements, do these statements constitute "discrimination," which we must not tolerate at all? I would hope not, since that would make it impossible to ask and answer about the quoted sources, even if they're within the scope of a given site

On the other hand, the policy can't be "you can quote anything you want as long as you voice it from a NPOV," as that could open the door to people making statements of exactly these forms, but actually with the intent to offend or alienate, rather than to discuss.

Clearly, at the end of the day, determining how to balance the merits of information and inclusiveness with respect to particular statements is going to come down to moderators' discretion. I'm wondering, though, whether this "None of any kind" kind of policy will make moderators' jobs harder in some situations. I'm sorry, but I don't have a good idea for alternative language, and I am not at all advocating striking this important point.


I note without surprise that this sort of consideration was discussed with respect to the last iteration of "Be Nice."

7
  • Quotes deserve special consideration otherwise stuff that's considered discrimination is not voiced from a neutral point of view.
    – Helmar
    Commented Jul 3, 2018 at 17:12
  • 1
    Excellent point. I personally think that trying for a neutral point of view is (a) probably unattainable and (b) a very bad goal. Trying for an objective point of view is a great thing to do. Attempting to be neutral results in quoting a small bunch of sour grapes in an effort to give "fair" treatment to both sides, or quoting strawmen disagreements with your own view. Neutral writing feels slimy to read. No one is neutral, but it is possible to be objective.
    – Wildcard
    Commented Jul 4, 2018 at 3:02
  • @Wildcard, actually, I think I meant something more like "objective" when I wrote "NPOV." Either way, you can end up with statements that are part of worthwhile discourse but that will offend some people. Commented Jul 4, 2018 at 3:37
  • 1
    That's true. I think it's better to take responsibility for your own point of view—some people use "NPOV" to avoid responsibility for what they have chosen to share. I think you nailed it with: '...the policy can't be "you can quote anything you want as long as you voice it from a NPOV," as that could open the door to people making statements of exactly these forms, but actually with the intent to offend or alienate, rather than to discuss.' That's why I find so much of Wikipedia feels slimy to read.
    – Wildcard
    Commented Jul 4, 2018 at 4:09
  • 4
    I have not yet found any organised religion that compiles without warnings/errors. Same goes to most governments. Commented Jul 4, 2018 at 14:25
  • @MartinJames this just means you didn't try the right programming language. (E.g. try Whitespace.) Commented Jul 9, 2018 at 17:43
22
  1. Even if you don't agree with all of it, do you feel that this is a reasonable code of conduct? Does it affect your recommending Stack Overflow as a resource in any way? How so?

The focus of the CoC in my opinion puts too much effort on the existing user-base that has learned and follows the SE-rules to be nicer rather than asking new potential users to learn our rules and adhere to our standards.

On the other hand, I do know a lot of people that refrained from actively using StackExchange sites because there was no CoC in place and that they will feel more safe with one in place.

In total, I consider the CoC a good thing if we could get more people on board that bring the platform forward.

The focus shouldn't only be to become a welcoming and inclusive community. We should not forget the quality and usefulness of the overall content.

  1. Is anything in this document, including its purpose, unclear to you in any way? Are there any instances where you'd suggest alternate language or copy? Where? Why?

There are certain places where I find the direction off or unclear, e.g.:

If you’re here to help others, be patient and welcoming. Learning how to participate in our community can be daunting, especially if someone is new. Be supportive while others are learning.

This can backfire and increase the culture of always-helping even in spite of questions that better should have been deleted in the first place.

There is a certain noise level of new-member questions that don't fit our standards, yet which that are encouraged by reputation-farming / helpful-to-a-fault helpers.

Also to what extent should I be welcoming and supportive?

Do I break the guideline of being supportive by downvoting a question of a first-time poster? I don't consider it very welcoming to receive a bunch of downvotes on your first question, yet the high standards for content made the platform as successful as it is today. You should embrace that also within your CoC.

If you’re here to get help, make it as easy as possible for others to help you. Our community is made possible by volunteers. Follow our guidelines and don't worry if others suggest changes or edit your question - they’re trying to make your question helpful to as many people as possible.

Please link to the appropriate how-to-ask-pages for each StackExchange network page. Make it mandatory for new user to read the how-to-ask-page. That's something we should be able to expect from new users.

Be friendly, clear, and constructive. Editing, commenting, and sharing feedback are healthy parts of our community. When giving feedback, avoid jokes and sarcasm -- tone is hard to decipher online. Be open to receiving constructive feedback.

Better: be friendly and concise. Avoid noise by not writing "Can somebody please help me?" in your question or posting "Thank you!"-comments.

If someone points out that your behavior is making others uncomfortable, stop doing it.

Please explicate this. As it reads, it the perfect way to completely disengage potential helpers because of fear that they might make someone uncomfortable. People may stop giving valid criticism and feedback just because one person thought a comment made somebody else uncomfortable.

Nobody likes to hear that their question is too broad, off-topic or should contain a better code example. Nobody likes to hear that their answer won't fit the need of the OP.

Should we stop pointing out flaws because it may make someone uncomfortable? Some people are very direct, others feels uncomfortable very easily.

I won't know, despite my best efforts to be constructive, how my text will be perceived.

When you want to be a member of this community, you should be able to handle and accept feedback. That very feedback might make you uncomfortable, which can be a good thing as you have a chance to grow as a person because of it.

No subtle put-downs or unwelcoming language. Regardless of intent, this behavior can have a significant negative impact on others. For example, saying “You could Google this in 5 seconds” is a subtle put-down.

What's the alternative? Downvote and not comment? Just flag the question because of low quality? Without giving the OP a reason to why I did such action? Or should I just let the question stand, because we-as-a-community want to be inclusive? How does this help the quality of the overall content?

I understand that RTFM-comments and lmgtfy-links are not helpful, yet it also showcases a problem this platform has with newcomer questions.

There is a reason why people react with these kind of comments. That should be addressed as well, even though not necessarily within the CoC.

  1. If you're a long-term contributor, how does this document impact your feelings toward new users and their needs, if at all? How does it impact your feelings toward the company? Or, more broadly, did this document inspire any metacognition at all?

It's a good thing to explicate the rules beyond a fuzzy "Be nice". I think the current draft does not highlight enough that new users should also put in more effort than on other platforms.

  1. If you're someone who wants to contribute more often, or have felt yourself sort of stuck on the fringes, how does this document change your perception of the site, if at all?

I may refrain from commenting or pointing out flaws because it may make somebody uncomfortable. I need my stackoverflow access as it is a professional resource for me. I most likely would use only the downvote and flagging option in the future.

3
  • 2
    "On the other hand, I do know a lot of people that refrained from actively using StackExchange sites because there was no CoC in place and that they will feel more safe with one in place." – do they refrain because no CoC document is in place, or because actually people are not conducting nice? Commented Jul 9, 2018 at 17:13
  • 7
    "I won't know, despite my best efforts to be constructive, how my text will be perceived." This underscores one of the biggest mistakes with the tone of this entire document: no Code of Conduct should require the members to read minds in order to comply with the Code of Conduct.
    – Stephan
    Commented Jul 9, 2018 at 18:38
  • @PaŭloEbermann Depends on who I talked about stackoverflow at tech conferences. Some won't contribute to any platform not having a CoC. From other's perspective, people are not conduction nice on StackOverflow. (I don't share that sentiment. I find SE refreshingly direct, yet I may be part of the problem.) It's worth noting that mostly women and/or junior developer are having a hard time engaging or even being "bold enough" to ask their first question.
    – k0pernikus
    Commented Jul 10, 2018 at 7:32
21

Apologize and move on

I'd drop the apologize part personally. Just because someone felt upset about something doesn't mean they are owed an apology. The person who meant no harm may not feel they did anything wrong either and should not have to apologize. They should avoid further offense if reasonably possible, but it's somewhat dangerous to expect an apology for reasonable behavior that is misconstrued and may be seen as overly sensitive by the speaker. Sure, it would be nice for them to apologize, but I don't think it's a necessary requirement and requiring it is likely to cause more problems than it solves.

2
  • 7
    A requirement to apologize is likely to produce non-apologies such as "I'm sorry you were offended", which don't make anyone feel better. Commented Jul 11, 2018 at 0:22
  • 4
    @JeffreyBosboom I'm afraid the non-apology would rather be: "I'm sorry you felt offended by my remark, which wasn't offensive at all." Comment flamewar almost guaranteed.
    – k0pernikus
    Commented Jul 13, 2018 at 10:20
18

My biggest question is whether the "be nice" emphasis of this policy is really directed at comments or whether it affects moderation actions as well.

Should we interpret this new policy as discouraging downvotes and other quality moderation actions? For example, a new user posts a blatantly off-topic, unclear, and poorly researched question. Should I refrain from downvoting, closevoting, flagging, or taking other moderation actions on the question out of sensitivity for how the OP might feel to see their question moderated, or will I be allowed to proceed as normal and just be discouraged from also posting potentially snarky or condescending comments about the poor-quality of the OP's question (e.g. "Do your own homework", "Please stop posting off-topic questions, you are going to get a question ban", or "Did you read the Help Center?")?

On SO, I can sit on the "new questions" page for an hour and use up an entire day's worth of close votes and downvotes just on zero-effort homework dumps, wall-of-code debugging requests, resource requests (always off-topic) and "omg my puter no worky halp" posts. The Question Ban exists to help ensure high quality posts, and downvotes, closevotes, delete votes, and flags are the primary tools by which we indicate quality. By refraining from non-comment moderation actions, we could soon end up with a dramatic drop in quality, both by leaving bad posts on the site and not shutting off people who continually make poor posts. A few low-quality posts can be edited into shape by the community, but many can't. Do we want to just start allowing those posts to stay (decrease quality because it's rude to censor low-quality questions) or do we want to just be nicer while at the same time taking action to ensure high quality? Similarly, I can hang out in the SOBotics chat room and get notified of dozens of "me too" 'answers' that are being posted at every hour of the day and night. Should we be 'sensitive' to the feelings of those who post NAA answers ("I was just asking for help, they didn't have to get all medieval and downvote me and delete my post!") and fail to remove the posts, or should we just downvote, flag, and delete these inappropriate posts silently, avoiding confrontation with the posters but also not telling posters what they did wrong and what they should do next time?

To some extent, it does makes sense to cut the snark and simply silently downvote, closevote, and delete low-quality posts, but there is another issue. A few times a week, someone comes onto Meta.SO complaining that their posts are being downvoted and closed with no feedback, and begging for information on what they are doing wrong so that they can improve. I'm in a quandary.

@JohnHascall 's idea of canned comments for common downvote/close reasons is intriguing. If there was a pre-vetted 'nice' way to ask someone not to post zero-effort homework dumps, I could see myself using that instead of trying to come up with a phrasing myself that could be misinterpreted by the poster or a mod as unnecessarily snarky or rude. That could be an interesting solution - we could have a process to provide feedback to users to indicate what it was that they did wrong, but by having those comments pre-vetted, we can protect ourselves against having our attempts to 'educate' the user misinterpreted as not being nice.

2
  • 5
    (Re your first sentence) That's a good question. I think the fundamental point that must be more widely realized is that allowing Stack Exchange's real mission to become secondary, and failing to take actions to accomplish that mission, is NOT nice.
    – Wildcard
    Commented Jul 6, 2018 at 23:53
  • +1 to the canned responses for poor quality questions part. I have a document with the links to various how-to-ask links that i use to moderate the first-timers. This is kind of a no-brainer feature that would not only save us volunteers a ton of time, but also provide helpful feedback to the first-timers. That alone might be enough to do away with all of this CoC nonsense while achieving the goals it's intended to achieve.
    – Stephan
    Commented Jul 9, 2018 at 18:44
18

I like what's in your Code of Conduct. That's pretty much common sense and good to be manifested. Apart from that, I don't think it's necessary. I'm very active on Stackoverflow, and I haven't seen a single quality post where the user had to suffer from anything of what you've listed under

Unacceptable Behavior

What you are doing with the Code of Conduct is like treating the symptoms of a disease, not the disease itself.

Instead of spending time on rewriting / extending the be nice policy, I'd suggest to put all your effort into

How can people be taught to write good questions.

Once things like “Ask a question” wizard prototype are implemented, improved and really used, most of the so called unacceptable behavior will vanish, because it's no longer needed to ask someone to google before asking a question (and the other things) - your question filter and wizard should already have handled this.


No discrimination of any kind. This includes any language likely to offend or alienate people based on (but not limited to): race, gender, gender identity or expression, English fluency, sexual orientation, disability, mental illness, nationality, neurodiversity, physical appearance, body size, or religion.

That's common sense, I've never seen anything of that on Stackoverflow

No harassment. This includes, but isn’t limited to: bullying, intimidation, vulgar language, direct or indirect threats, sexually suggestive remarks, patterns of inappropriate social contact, and sustained disruptions of discussion.

That's common sense, I've never seen anything of that on Stackoverflow


Apart from that, like said above - it's always better to be in a friendly environment; that's why I like the manifestation.

I strongly recommend though, to look on what is StackExchange doing to force unacceptable behavior. That is, letting people bully existing users with tons of unanswerable/no effort/... questions. It's on Stackexchange to do something against that.

Stackoverflow is the best source on the internet to get answers to programming questions, no matter how hard they are to solve. We shouldn't do everything to bring the quality level to reddit. That exists already.

18

As briefly as possible (with full knowledge simply stating this will make many believe I am a villain:)),

If someone points out that your behavior is making others uncomfortable, stop doing it. Sometimes, people unconsciously say things that negatively affect others. Even if this wasn’t your intent, apologize and move on.

This is highly problematic because people can choose to be offended by anything, and people have different tolerances for what makes them uncomfortable.

Stack Exchange needs to be a civil space, but it's also a public space, and community guidelines have to reflect the median, not the extremes. (The extremes may be said to include both the most uncivil individuals and the most sensitive individuals.)

The likely result of this directive is an inverted hegemony by which people who choose to read in malicious intent to innocuous statements end up becoming the harassers, attacking those they believe have transgressed.

17

Respect requests to stop behavior that is making others feel uncomfortable. Remember that no question, answer or comment is worth getting worked up over. If someone is making you upset, ask them to stop. Do your best to resolve it civilly with the other person or take a break from the situation. If they still won't stop, we may consider that harassment (see Unacceptable Behavior).

This will prove to be the mischievous hobgoblin of the policy. The potential for abuse of this one is of legendary proportions as anyone with any sort of difficulty can claim that practically anything is making them uncomfortable.

6
  • 1
    In my experience it's what comes after the "hey, can we not talk about $TOPIC because $REASON" that really drives things. When I've seen that taken seriously and met with "oh, sorry, I didn't realize. Do you have a minute and are you willing to explain the discomfort a little more so I can understand better..." there's a real moment for people to grow in understanding and connection. Or to smoke out a mischievous hobgoblin who, frankly, won't be able to genuinely engage with that sort of response and it's obvious to everyone in the room that they're trying to bully others with the policy.
    – nitsua60
    Commented Jul 6, 2018 at 17:00
  • 2
    OTOH when it's assumed that the intent is bad and it's met with "oh, another of those, shove off jerkwad..." things go badly. Maybe not at that moment, but that's a room whose "culture" is being set at "don't assume good faith." In other words, bad-faith actors are the mischievous hobgoblins, not bits of policy.
    – nitsua60
    Commented Jul 6, 2018 at 17:00
  • @nitsua60 the corollary to that is, of course, if the bad faith actors are the problem, why assume a policy that can easily be exploited won't attract them.
    – user316129
    Commented Jul 6, 2018 at 17:03
  • That's where users who habitually engage in such behavior and abuse the policy will be quickly detected by moderators. Someone cannot claim everything on SE makes them uncomfortable. Even if it does, then this probably isn't a site they should be comfortable being on and it should be rethought as to whether they actually should be here. e.g. A user is offended by just programming. Not a particular bit of offensive code or variable names. Just the act of programming is offensive. They obviously shouldn't participate on SO.
    – user64742
    Commented Jul 8, 2018 at 14:11
  • @TheGreatDuck There is more to SE than just SO, and I'm already seeing this abuse on the softer sites. Furthermore, this behavior is NOT quickly detected, and it does drive people away
    – user316129
    Commented Jul 9, 2018 at 10:01
  • 2
    @nitsua60 at The Workplace, we've had a spate of people going around changing all the pronouns, BTW, so this DOES happen and it was not resolved quickly, for a real life example to prove that this is not just navel-gazing on my part.
    – user316129
    Commented Jul 9, 2018 at 13:18
17

Unfriendly behavior is not a cause in itself, but a symptom of a deeper problem. Stackoverflow sends thousands of newbie questions to everyone, flooding the reading stream of willing experts with what is spam to them.

Stackoverflow provides insufficient ways for askers (or others) to mark their questions as newbie question, and insufficient filters for experts to allow them to choose what they want to see. What works for thousands of participants does not work for millions.

As an example I prefer to just look at questions who are 5 hours old, have no answer and have a positive score. I might miss a few interesting ones, but avoid most annoying ones. But to do so I have to use a clumsy browser bookmark and a userscript to hide all my search options. Why can't Stackoverflow provide such options conveniently?

This is a sample search query that I bookmarked:

intags:mine answers:0..1 score:2 is:question closed:no duplicate:no hasaccepted:no created:3m..1d lastactive:9d..0d

Facebook has a button labelled "Show me posts like this less often". Not sure how that works, but why can't stackoverflow use this mentality to help readers avoid seeing posts that they do not wish to see?

A different CoC does not change this, just like the last 10 years what was missing was not a better CoC. The CoC is just a meek attempt to appease the PC activists, it wont change a thing.The next logical step is to remove downvoting, because of all the hurt feelings of askers who get downvoted a lot. Downvoting is also "a subtle put-down".

Give experts an easy way to see only expert questions, and experts will stop leashing out against what is spam to them.

This would also help keeping experts interested in the platform, rather than turning away in disgust of all the spam.

3
  • How do you envisage the impact of your proposed feature on non-diamond-mod moderation tasks? My instinctive reaction is that by hiding a lot (perhaps the majority) of questions from some of the users with sufficient rep to moderate them, poor quality questions will accumulate and dilute the usefulness of search results. Although there's a tradeoff against burning out expert responders, so please don't see this observation as a black-and-white rejection of the idea. Commented Jul 9, 2018 at 7:32
  • 1
    I cannot predict the impact on how much moderation will happen on what posts, and what kind of questions would accumulate. Users who enjoy to moderate would still be able to see all posts, I think of better filtering as an opt-in feature. For search, the search algorithm handles quality filtering. The internet is full of trash, but search engines manage to get useful results to the first page. Moderation is most valuable for frequent search results, and those still would get a lot of attention.
    – tkruse
    Commented Jul 9, 2018 at 12:09
  • I like this idea. Maybe it could be turned into a feature request.
    – user380304
    Commented Jul 11, 2018 at 14:02
16

"If someone points out that your behavior is making others uncomfortable, stop doing it. Sometimes, people unconsciously say things that negatively affect others. Even if this wasn’t your intent, apologize and move on." The way I read this is that if for any reason anybody feels uncomfortable in response to something I have written, the source of their discomfort is my unconsciousness and they are automatically owed an apology.

I think this needs serious rethinking and reworking.

1
  • 1
    Beyond the dicey meaning, I think it's also just too long a way to say what they want to say, which can discourage readers from looking at it.
    – user392547
    Commented Jul 18, 2018 at 3:04
15

Does non-discrimination mean we ought to lower the bar to be more accepting of patently dumb questions?

I co-moderate the Christianity StackExchange and there are any number of awful questions one could ask about Noah's Ark for instance. We pretty much allow 'em if they're objectively answerable, but to be objectively answerable they have to show a little bit of research and come from an identifiable perspective.

If we were any nicer we'd allow all manner of craziness.

So, basically, we've created our community rules (same as all the other communities) that supersede base policies.

Could we have 10th amendment to this CoC that says users should abide by community rules as well?

2
14

Tiny but extremely bothersome: the structure of the "unacceptable behavior" section.

The heading is "Unacceptable behavior". I expect a list of bad things which a person can do. Instead, I see

No subtle put-downs or unwelcoming language. Regardless of intent, this behavior can have a significant negative impact on others. For example, saying “You could Google this in 5 seconds” is a subtle put-down.

That seems to say that not subtly putting down people is unacceptable behavior, if you read it without any common sense. It pinches a bit. It doesn't effectively damage readability at all, but it does seem a bit imperfect and weird.

Can we remove the "no" at the beginning of each line? Think of it...

Unacceptable Behavior

  • Subtle put-downs or unwelcoming language. Regardless of intent, this behavior can have a significant negative impact on others. For example, saying “You could Google this in 5 seconds” is a subtle put-down.

  • Name-calling or personal attacks. Focus on the content, not the person. This includes terms that feel personal even when they're applied to content (like "lazy", "ignorant", or "whiny").

2
  • oh, wait, I guess someone could rules lawyer it into non subtle putdowns being ok.... Commented Jul 4, 2018 at 9:46
  • @JourneymanGeek Well, I did warn you that it doesn't damage readability at all ;) It's just... weird... to leave it like that.
    – user392547
    Commented Jul 4, 2018 at 9:51
13

There's a subtle aspect to the rewrite that I think is a HUGE improvement. I don't know to what extent it was deliberate, but I hope that it will be preserved when incorporating any feedback. And that is: attention to the positive behavior we want to encourage, rather than emphasis on the negative.

Glancing over the current (to be replaced) write up of the "Be Nice" policy, the words that jump out are:

  • jerk, name-calling, rudeness, belittling, bigotry.

Glancing over the new writeup, the stand out words are:

  • help, patient, welcoming, friendly.

All by itself this is a dramatic improvement. I feel it will have a disproportionate impact on the actual effect on people reading the "Be Nice" policy.

(By the same token, I've always been sort of ironically amused at "safety" posters that show people banging their heads into things and getting hurt.)

This goes beyond the bolded words and into the more subtle word choices. The "tl;dr" version of the flagging bullet point has a beautiful phrasing for the follow-up sentence: "Every person contributes to creating a kind, respectful community."

The writing overall show a good grasp of the subtle effect (sometimes not so subtle) of word choices on emotional responses and actions. Good on you.

13

The other posts cover most of the concerns pretty well. So I will add what I think this document missed out on.

Namely, Actionable Items. Sure you mention flags and contacting and telling others off (at your own risk); But as a user, when I want an expansion of "Be Nice", I don't care about the definition of harassment. I know what harassment is.

What I REALLY want to know, is how can I take Did you Google this first?, and turn it into something more acceptable without losing the important message (that you are expected to Google this before asking)?

As a new user, I had to stumble into Meta, and how to properly use it to resolve conflicts (like an edit war, or why a post with no comments or answers has 5 down-votes despite appearing to meet all the site requirements). The CoC should link to supporting resources like how to use Meta. (Don't put this all in the same document or it will become a swamp. Just link to related support content with a quick reference)

The CoC also takes the position that only you are at fault. While "An eye for an eye makes the world blind", There really should also be a link to another resource for How should I deal with problem users/content? This policy discourages giving factual constructive criticism, and leaves people with down-votes/deleted content. Giving people electric shocks for "bad" conduct is very harmful to both parties. If I am the problem, I should be directed to a resource that clearly explains why, and what I can do to correct this. (This is currently done with sharing Meta topics, as that is the best way to deal with SEs fluent nature, but is done rarely. The CoC should give guidance to do this short of practice.)

As a semi-casual user, I find the new CoC useless. It conveys no information to me that "Be Nice" didn't. (Other than implying having a sense of humor is a suspend-able/ban-able offense) What it really needs is references to addition resources related to common "Be Nice" issues (like no effort questions).


One last note Unacceptable Behavior and No ... is a double negative. I get what you meant to say, but better to remove the extra Nos.

2
  • 1
    Many of the per-site metas have a question containing some pro-forma comments to be used when welcoming new users, from the review queues, etc. These could be the basis for a similar resource containing comments pre-vetted as sufficiently welcoming. Commented Jul 6, 2018 at 7:24
  • 2
    This answer seems to take a lot of inspiriation from the modern practice of Criminology- that it's possible to systematically and sympathetically help people who engage in inappropriate behaviors (theft, rape, second degree mopery, etc.) without condoning those behaviors. Commented Jul 6, 2018 at 19:24
12

Flag the harmful content. Depending on the content, flagging will alert moderators or trigger an automatic deletion.

Since this is supposed to be for everyone I have to point out new users can't flag untill they have atleast 15 reputation.

4
  • 6
    SE team folks recently bragged here about the plan to test allowing new users to flag comments on their own posts
    – gnat
    Commented Jul 4, 2018 at 14:44
  • 7
    @gnat As soon as that plan becomes reality, I will remove this answer. Commented Jul 4, 2018 at 14:48
  • based on past observations on proposals of real changes that could improve SE network I wouldn't hold my breath waiting for this to happen
    – gnat
    Commented Jul 4, 2018 at 14:53
  • 5
    @gnat the doctor told me that holding my breath for six to eight weeks would possibly be damaging to my health.
    – hat
    Commented Jul 4, 2018 at 16:04
12

I agree that there needs to be an element of nice, although I'd prefer to talk more specifically about being "civil" than "nice" given our tight Q&A format and huge community.

The lack of conforming to requirements by users plays a massive part in the fact "nice" is currently missing. In a huge online community with people ignoring rules and with great potential for nonconformity, "nice" will only maintain civility to a certain degree and mostly only with those who are naturally nice anyway.

People who are naturally nice will all get on with it and in a civil manner, without the need for a CoC. Some just need a nudge and get a bit caught up in the frustrations etc, but those who really won't be nice or who don't adhere to requirements simply need strong rules, not telling "please be nicer" as why would they suddenly just do that?

This is (sadly) the fundamental issue with any large community, including ours I believe - the issue that:
The users who are not affiliated with "nice", who ignore rules for their own means and are uncivil, they drain the resources without giving back which means other "nice" users have to put extra in to compensate. Then those who are being nice are being taken advantage of, which is "not nice", and suddenly those nice users are the ones dealing with a lack of "nice", and as a result "nice" will naturally dissipate from it being unfair - precisely one of the main issues we face now.

We do need more "nice" but that will only happen if you strongly tackle the ones breaking the civility and not conforming to requirements. Being nice will not stop most non-nice people from causing problems, they do it because they are not nice and, as with any community, rules are needed to force them down certain paths! I wish it were not true, but ignoring this in the name of being "nice" will hinder fixing the problem and ironically hinder bringing about more nice.

It's a bit of a gloomy picture I paint on top of all this "nice" talk, but without hard rules in place, made clear, and importantly enforced, we will not have civility, let alone "nice".

 

 

What "nice" should mean and where it should be applied

(taken from another answer of mine but is very relevant here)

  • Comments should "be nice" and explain the rules and requirements of this site in a more welcoming way rather than short and blunt. This has the chance to make people feel welcomed and possibly more likely to improve their question/answer.
  • Question askers should "be nice" and recognise their question is very important site content which if is bad has potential to be harmful to the site in various ways - bad search results, the need for moderation, bad content, etc. Questions produce much work and resource usage, by bringing about the need for edits, comments, votes, flags, and answers. This is a lot of power, and with it comes great responsibility.
  • New users should "be nice" as they're new to a site, and should take some time to learn its ways (we should make this easy of course). E.g. when someone informs them their question has an issue, they should look to fix it. They should be nice and not want to come to a new site and expect everything to be their way, on a site that's been here for many years with specific ways. Once a new user posts their question there is no "us and them" they're entirely part of the "us", and should care about the site and the quality of that question.
  • Answerers should "be nice" and recognise that answering poor questions may well give that one person information they wanted, but the site repeatedly makes it clear this is not what it wants. It adds poor content, which in turn attracts more poor content, and brings about the need for moderating. We have millions of questions and answers, and it only needs a small percentage of those to be bad for it to be fairly noticeable and have a negative impact.
  • Stack Exchange company should "be nice" and recognise that those arguing over the poor and lazy questions, and those who answer them, have to heart the very interest and goals Stack Exchange declares it has.

Some return points on what I've seen debated so far:

  1. Site rules cannot only apply to those who have read and understood them! This would be unfair, so even new users without knowledge should be expected to learn, or need to be issued any relevant penalties (perhaps some new less severe ones to get them to the info about expectations). This can be done while being civil and nice, but is a necessity to maintain order and this civility all round.
  2. Being a new user is not an excuse to not fix a poor question once learned of that requirement. They are getting all the benefits of the site's resources, functionality, and people's time, and it is therefore only fair that they uphold the requirements that make the sites great in the first place which is providing them with the aforementioned goodies!
  3. "Be Nice" is great, but should not mean "leeway to let the rules slip by". That includes bad question after question after... and nothing done about it.
  4. This should be about "all users" being nice, and regardless entirely of any particular action taken. So rather than focussing on commenters being unkind, also see that poor questions are unkind to both people trying to help and quality of the sites.

Most importantly:

  1. If question quality rules were perhaps a bit stricter, but definitely enforced much sooner so bad questions don't stand a chance, then there'd be much less of a problem with people not being nice, because there will be much less to not be nice about!

If we could make a ton of bad questions not happen, and more go away quickly, it would greatly reduce bad comments and people not being nice, because laziness and no care for the site's quality is where a lot of "not being nice" comes from.

We have the Help Center (etc) which lists dos and don'ts, but I think a lot of them read like "guidelines", almost optional in some case, rather than very important rules that keep the site in order. An idea would be to (leave the Help Center as it full of info) and introduce a new page, linked in the top bar, with short and to the point info of the core requirements that keep the sites clean, organised, and civil. Such as a two column list with requirement | penality.

This should include all the things which on face value may seem trivial but harm the sites greatly, and as a result a knock-on effect to other things.
Such as not fixing a bad question causes bad content, low quality answers, snarky comments (not justified but it does happen). And, just for example, something like "3 strikes and cannot ask for a week" (or whatever, not really suggesting without a bigger debate) 3 bad comments and cannot comment for a day. Etc.

I think "nice" is great, but let's not try to clean up a dropped broken jar of jam with a feather duster!

12

Just to add to existing great answers:

I feel that this whole change of COC is focusing on improving new users' experience AGAINST existing users.

I mean I agree we need to be more welcoming and more respectful, however this should happen by improving everybody's experience. This whole story sounds like SE will change to be more welcoming for new users and a lot more strict for existing ones, and that's a very bad direction; might loose a lot of established users contributing to the site in favour of new users who might not want to contribute at all.

My substantial problem with this change that is doesn't offer help for existing users

  • writing comments asking for clarifications in questions and explaining issues with them takes time, writing them polite could take even more for those who aren't very good in English
    moreover explaining close reason is kind of a useless activity from the perspective of the comment's author: just explains things again what should've been read by OP before asking the question
  • down-vote: as mentioned already in other answers, already a down-vote is seem offensive by some people who just come to ask for help. And I've to agree with that, getting a couple of down-votes when you arrive to a site isn't welcoming. (honestly I'd disable down-voting on new users' questions, and just close them if appropriate. Answers are different as those can't be closed).
  • close: also mentioned already that close reasons are not in a friendly tone, that should be improved too
  • my suggestions: we should look for ways making pre-written comments for the cases which occur most frequently (off-topic, unclear, missing details...), which maybe automatically added with close vote / flag, or inserted manually and could be edited easily. Currently OPs see the close reason only when the question is closed, which may even take a couple of days.
  • also, give positive examples for us, I can accept “You could Google this in 5 seconds” is a subtle put-down, but I still don't know what I should write instead of it.
1
  • Yes, answers can't be closed, but the reason for downvoting answers is usually completely different. Questions are usually downvoted as off-topic or something similarly StackExchange-specific.Answers are usually downvoted due to factual inaccuracies or low-quality link-only, which is frequently taken correctly by the author. Such problems aren't frequently as SE-specific: downvote reasons are the same on a lot of internet platforms. Question downvotes are made against people looking for help, and are thus more likely to be taken with concern and disappointment.
    – user392547
    Commented Jul 17, 2018 at 9:52
12

There are several issues with this CoC...

First of all, it isn't an instrumentation of the SE sites core mission, heck, it doesn't even link back to any meta post. There's nothing about it that makes users to collaborate and respect each others. Yeah, it covers name calling pretty well, but so also does respect, and without any of those issues that were already pointed out.

Secondly, this thing also segregates users into groups and give weapons to attack each others. You ABSOLUTELY do not want this. It's called tribalism and while it allowed us to focus our collective towards a single goal, in this case it only serves to separate all users into classes/castes and tell them how to treat the other castes. Remember, tribalism thrives on making other tribes look bad, independently of how they actually are (see this CGP Grey short about it or look for "human tribalism" in Google Scholar). This behavior leads to absurd results because, a) it only allows us to collaborate in specific stuff and b) such stuff is usually meaningless (see SO position on editing). This goes against your intent of "together" and is more like "us". This is dangerous if codified, and surely the end of a collaboratively site like SE.

Thirdly, most of the CoC can just be summarily deleted. Yes, it has examples, and yes, people loves examples, but only give those if they don't understand simpler and blunter words and not in the CoC. The entire thing could be summarized as "be nice to each other and don't be a dick" and it would apply to all of it, from the interactions between any level of users privileges to the humans with the system. Don't codify mbadness (stealing straight from Shog's fingers), the CoC should be a framework and as such it should be very simple. It should be easily digested. Just leverage meta if you want to make a point or dive into a topic.

Now, this is for the ones that wrote it: remember, your opinions are not what you are. It's human to feel aggravated, but it's also possible to humans to be reasonable. Stop what you are thinking right now, it would be meaningless trying to be introspective right now. Cool off, then read this 3 times over, read what you wrote 3 times, and this post 3 times again. I took up until now to write 4 measly paragraphs and I already had read the CoC when there the Google Doc didn't have an absurd amount of animals in there. Others also did a fair amount of work trying to mince the whole thing along with all the context surrounding it. What's the worse it could happen, anyways? Writing it all again?

11

One major question I have is:

How much of this is considered new policy, versus simply formalizing existing policy?

It seems like there may be some new policy here, but I'm unsure.


I have a few thoughts on the Reporting and Enforcement section.

enter image description here

Is this a new policy? Does this mean that when a comment is removed for being rude/abusive (or simply unnecessary?) a warning is generated? As is I don't think this happens, but maybe I'm just not aware of it.

enter image description here

Well that escalated quickly. This list jumps right from some annoying noise to sexual harassment and voting fraud. I'm concerned this may give the impressions that certain classes of comments between these edge cases may be considered only minor infractions live "did you Google it?".

Also, is a comment like “This is obvious, just Google it.” something we should flag as rude/abusive under this policy?

2
  • 2
    That comment has been deleted by mods hundreds of times and personally I've rated those flags helpful in the past as well.
    – Helmar
    Commented Jul 3, 2018 at 21:42
  • 2
    The escalation is really weird though
    – Helmar
    Commented Jul 3, 2018 at 21:43
11

Reporting and enforcement

We take your reports seriously. Those who don’t follow the Code of Conduct in good faith may face repercussions deemed appropriate by our moderation team. This is how moderators generally handle misconduct:

Warning/Suspension/Expulsion

All actions will be taken on a case-by-case basis at the discretion of our moderators. Here are a few examples:

Behaviour/Action table

  1. The two tables don't match. Content deletion just appears almost out of nowhere in the second table. Warning doesn't appear at all. The mismatch could be fixed by standardising the two tables to say Content deletion and warning or Warning and content deletion.

  2. What about chat suspension? AIUI that falls between warning and account suspension on the severity scale.

2
  • "will remove offending content" => content deletion Commented Jul 19, 2018 at 13:57
  • Yes, I spotted that while I was writing the answer and changed "out of nowhere" to "almost out of nowhere". IMO the point still stands. Commented Jul 19, 2018 at 13:59
11

The draft has a section on enforcement that confuses me. Account expulsion is mentioned there, but that is not actually something we do now (spam, trolls or socks are removed, but that's not really relevant for the code of conduct).

Suspensions are mentioned to be "one day or more", but actually in most cases they start at 7 days. The escalating suspension lengths are not mentioned at all, and they are probably the most important aspect to know about the enforcement in general.

I'm not sure what is to gain by the examples given there, I don't think they can represent the way this works well enough to be useful. Most of the enforcement is about patterns of behaviour, not individual actions, so this gives a bit of a wrong impression.

One thing I'm wondering is whether this new code of conduct is actually meant to change the way we moderators enforce the rules, or is it meant to represent the existing rules and customs around enforcement?

11

When I first saw the notice of this Meta post, I immediately felt a sense of weariness and impatience. Now I've spent over an hour going over the document, the post, and about half the answers and associated comments, because I decided to soldier through it because in the end I care about this that much.

My biggest reason is I want to speak for all the people like me who decided not to look at this at all because their immediate reaction is that too much is being asked of volunteers. Much larger again is the group of people who never get invested in Stack Exchange because of that feeling, and the limitations on their behaviour imposed by the culture here.

Some answers have already eloquently expressed this aspect but I think it deserves expansion. I'm concerned SE doesn't understand how much this community has developed a personality which attracts those of like mind and repels others. This whole thread inherently has an aspect of preaching to the choir, it can't avoid it. The bulk of those who disagree won't bother putting time into this, and there are those who only come to SE for a bit and then leave because they smell an atmosphere they don't like. That includes lots of good people.

I think in particular of a person who was drummed out of SE because they said a series of really offensive things. Okay, I get it. But that person also made a lot of valuable contributions. More than that, I have a strong feeling SE was important to them because it helped them with their social problems. It helped them with their isolation. And now it's gone. That must have hurt. I know their intolerance was a burden, and it was wrong. I just wish we were in a place that could allow things like that to roll off our backs just like we allow being called to an extremely high standard when we are only human to roll off our backs. People like that need that sort of support, and we are all that person every now and then. Get rid of offensive comments, sure, warn them, ban them for a bit when they need to cool off and reflect. But be welcoming to them too. They are human too. Why are we distinguishing between some weaknesses as being deserving of understanding, and saying others aren't? At least let's be honest with ourselves about what we are doing. We are taking the easy way out.

What we need are tools. A page of rules is not the real solution to this. The problem here is that text cannot satisfy the human heart, and yet all committed users have an emotional attachment to SE to one degree or another. Maximizing the ability of Stack Exchange to accommodate the widest range of human interactions possible is the solution here.

I wish moderating duties were distributed as widely as is possible while maintaining a minimum standard of moderation quality. A casual assessment might think that is the case, but I disagree. It's where things get sensitive that the bulk of responsibility falls on only a few, and that's where the ball gets dropped, usually because it is too great a burden to ask of a volunteer. I wish it was easier to have private interactions with other users.

99% of what this place is, is thanks purely to volunteers. I wish all the stuff from Stack Exchange staff made clear how much they owe to volunteers. My God, we're awesome.

4
  • 9
    If the person you spoke of, really cared, they wouldn’t have said the offensive and intolerant things. If people are allowed to insult people intentionally, for example acuse them of doing drugs “again”, that’s going to drive away a lot more people than that one person who is being intentionally offensive.
    – Ramhound
    Commented Jul 4, 2018 at 20:08
  • 18
    ...a person who was drummed out of SE because they said a series of really offensive things. ... But that person also made a lot of valuable contributions. Allowing a person that says "really offensive" things to remain in a community just because they contributed some content that was useful is not good for the long term health of the community. Volunteer communities need a constant influx of new members and if I see that kind of toxic behavior tolerated, I don't join a community.
    – ColleenV
    Commented Jul 4, 2018 at 20:21
  • 8
    I'm not inclined to argue, i've already spent more time on this than my better judgement indicated. I've tried to adapt to seeing this as not a community and just use it for what i need. Just don't tell me this is about tolerance. This is about getting traffic and preventing lawsuits.
    – kim holder
    Commented Jul 5, 2018 at 1:15
  • 1
    Alright, i put in an edit to expand on the part that's controversial.
    – kim holder
    Commented Jul 5, 2018 at 14:18