I’ve posted bugs on this website in the past, and the result is always uncooperative and unproductive. Usually a workaround and/or a link to the FAQ is posted as a comment or answer, and then the community redefines the posting as a “support question” and removes the “bug” tag.

Most of the bugs I post are usability bugs. These are especially vulnerable to this problem because the community consists mostly of hardcore coders, who are all too quick to misdiagnose a usability bug as “user is dumb”, which leads them to reclassify bugreports incorrectly. Every time a particular usability bug is raised, the “answer” is a link to the FAQ — which completely misses the point. The usability bug stays, user after user posts about the same problem again and again, and the community is hardened in their judgement that “users are dumb”.

A usability bug is a usability bug irrespective of the existence of a workaround that might work for people who already know it and who are dedicated enough to learn and remember it. That is a very small proportion of users. Most users are not dumb, but they are either new or don’t use the site every day or have other worries in their lives. The point in UI design is that the software should be usable by someone who isn’t already an expert in the specific software itself and who doesn’t have time to read pages upon pages of manuals and FAQs and remember all of it.

In summary, Meta Stack Overflow does not provide any way for individual users to make developers aware of a problem. The user community works to shrug problems off even when the posted workaround is unusable and/or undiscoverable, and the posting is marked “closed” for supposedly being answered, while the actual bug still exists and remains unfixed. From the point of view of the person reporting the usability bug, it’s like I’m talking to a wall.

Stack Overflow needs a communication channel to the developers that doesn’t go through an unforgiving community that acts like a wall. Users need the ability to argue that their bugreport is a bug, and not see their postings closed and dismissed with no realistic opportunity for recourse. Users need an opportunity to make suggestions that will be seen by a reasonable developer who appreciates the concept of UI design, where there is no opportunity (or at least not as much of an opportunity) for a community of usability non-experts to close or reclassify postings and completely prevent developers from seeing the suggestion.

3 Answers 3


There are two separate, distinct and very important pieces that have been intermingled here.

  1. Are Usability Bugs Actually Bugs?
  2. Is The Meta Community Receptive?

I'll handle them separately because they are wildly divergent. Although the second is actually an example of the first.

Are Usability Bugs Actually Bugs?

This has a very important two word answer: No, But...

No they are not bugs. This is because what constitutes a bug is usually defined by the lower level implementation programmers. That is to say, "I was told to make it do XYZ and it does. Not A Bug" A bug is observed when the system does not behave as defined.

But is that really the target of applications? As Timwi states

[Well designed] software should be usable by someone who isn’t already an expert in the specific software itself and who [hasn't read the] manual or FAQs.

He is absolutely correct in this, however this is a design issue. It turns out that someone had assumed, misjudged, or simply not noticed something in the design phase.

What Is A Bug?
These items cannot be considered bugs in the traditional sense, because they cannot be simply "set right". It is not as simple as redesigning it to "spec" because the "bug" is in the spec.

This means a redesign, and any redesign needs to investigate the impact on hundreds of thousands of pre-existing questions and thousands of users. At this point it also could influence dozens of separate sites.

This is why "Usability Bugs", while being actual and real bugs, would fit more cleanly into the classification of [Feature Request] in our little walled-in Meta world. However, Jeff's quote was quite prescient.

"One of my favorite things about UserVoice - which we use [Ed. used] for Stack Overflow - is the way it intentionally blurs the line between bugs and feature requests. [...] [D]evelopers tend to use that division as a wedge against users. Nudge things you don't want to do into that feature request bucket, and proceed to ignore them forever."

Is The Meta Community Receptive?

Now that we have clarified the definitions as

  • Bug: Behaviour that differs from the spec.
  • Feature Request/Usability Bug: Problem/enhancement to the spec itself.

Power Users
We must then ask, is the Meta Community the correct place to claim that there is something wrong with StackOverflow at large. The Meta users are the ones who are most familiar, power users, very involved. Humans naturally distrust change, and these are the users who have the most to lose. We are also the ones who are the furthest from novice users.

Merely Human
The users of meta, being human, are also subject to things like frustration, annoyance, disdain, cliques, claques, coercion and cooperation. It is an imperfect community that is certainly not always mindful of the Be Nice! adage. (Have you read our FAQ?) This is a result of a nearly constant stream of identical complaints from people who have decided that their personal user-experience is a valid measure of everyone's user-experience.

Harsh But Fair
As a community of internet people, the very best that can be expected of Meta is that they be accurate (if scathing), quick (but arrogant), and helpful (though condescending).

Usability Bug?
This raises the question from earlier. Can the behaviour and sometimes difficult new-user experience on Meta be considered a bug. Jeff stated from the outset that a lot of the support and moderation would be done by the users, so currently it cannot be considered a bug, because we are accurate to the spec. Even if we had turned out to be a huge clan of hate-spewing sycophants, it would be according to spec.

This means it is a "feature request" (Replace the cruel Meta People) and the people who must evaluate the request are those same problematic people. This has the telltale signs of a self-enforcing disaster.

What can I do?

Situation 1: I'm a user and someone was offensive to me.
- Flag them as offensive using the flags.

Situation 2: I'm a user and someone was cruel or insulting.
- Ignore them. Do Not Engage. If it continues, flag for moderator.

Situation 3: I'm a user and everyone is repeatedly mean.
- Like Dogs, Bees and Trees, Meta regulars can sense fear. Do Not Engage. They will retaliate.
- It might be time to consider whether or not there is a reason that people are repeatedly treating your suggestions with disdain.

Situation 4: I believe the problem is systemic, and cannot be fixed on a case by case basis.
- This is the definition of a usability bug. These should be taken up with [email protected]. They are the only one who could possibly (if they agree with you) enact the kind of changes you desire.

Situation 5: I am a regular.


  • I fundamentally disagree with your definition -- I would define "bug" as "blocks some part of the core functionality with no viable workaround" .. there's an urgency to the issue and it affects lots of regular users. I'm fine with feature requests of all types, but bugs should get higher priority because they block regular users. Aug 31, 2010 at 2:06
  • @Jeff Your definition of "bug" excludes all community input. Because the site is still 'functioning', you don't have to approve anything we have to say because it isn't "core". You've also said "without viable workaround" as if that makes it not a bug. If my car breaks down but I hitch a ride, that doesn't make it better.
    – devinb
    Aug 31, 2010 at 2:30
  • @devinb I have no objection to feature-requests. All I'm saying is, bugs are higher priority, which is why adding [bug] to a question is not something I take lightly -- it better be a real [bug] or else you're just marking all your outgoing emails with IMPORTANT! aka "crying wolf.." Aug 31, 2010 at 2:40
  • @Jeff Atwood: Why are bugs that affect regular users more important than bugs that affect new users? New users are in the majority.
    – Timwi
    Aug 31, 2010 at 7:11
  • @devinb: Your definition assumes that there is a spec. Where is the spec for StackExchange? If it is secret, then I have no way of classifying bugs correctly, yet I’m criticised for supposedly crying wolf? Realistically though, even if there is a spec, small but important details like the tab order of elements on the ask-a-question page are not likely to be specced — what does your definition have to say then? Personally I start by assuming that it should be normal and that any deviation from the normal expectation is a bug.
    – Timwi
    Aug 31, 2010 at 7:16
  • @Timwi I'm not criticizing you. I'm trying to explain the way it works. Also it takes a mind-numbingly intense arrogance to believe that anything that deviates from your perception of normal is a bug.
    – devinb
    Aug 31, 2010 at 11:32
  • 1
    @Jeff While your opinion makes complete sense, I feel like you're missing or ignoring two important facts. A) Bugs can have their own priorities. You've defined "bugs" as only "critical bugs" and everything else as a feature request. When you say "that's not a bug, that's a feature request" it is equivalent to saying "The problem is not in my application, it is in you the user. And if I have time, I might try to accommodate you." Obviously, this isn't the attitude you intend to convey, but when you start claiming that nothing is a bug, you are giving that impression.
    – devinb
    Aug 31, 2010 at 11:34
  • @devinb: Does it? How do you decide that a particular behaviour is buggy (as opposed to not buggy; forget about feature requests for now) if not by comparing it against your perception of some kind of norm or expectation?
    – Timwi
    Sep 1, 2010 at 21:41

Every "bug" is important and valid to the person who opened it. There's a bit of bias there, though, don't you think? In your specific case, I don't believe that is in any way a bug, but a [feature-request].

The real question is, can you get other people to vote for your "bug" and support it?

If so it might end up on the bugs tab, or the feature requests tab.. those are the lists we tend to work off of.

  • 8
    Totally seconded. However, I can sympathize with the feeling of "talking to a wall": A quick explanation from the powers that be when adding a status-declined or status-bydesign would go a long way, especially for new users, and even if a community member has already outlined the reasons.
    – Pekka
    Aug 2, 2010 at 9:52
  • 3
    I'm going to post some quotes from Jeff. "One of my favorite things about UserVoice - which we use for Stack Overflow - is the way it intentionally blurs the line between bugs and feature requests. Users never understand the difference anyway, and what's worse, developers tend to use that division as a wedge against users. Nudge things you don't want to do into that feature request bucket, and proceed to ignore them forever." I suppose that was then, and this is now. codinghorror.com/blog/2008/11/…
    – MarkJ
    Aug 2, 2010 at 10:36
  • 2
    No, Jeff. The real question is does the bug affect the usability of the site for a substantial number of users? If you’re going to rely on the votes of J. Random People on the site, you are not going to find the answer.
    – Timwi
    Aug 2, 2010 at 10:57
  • 5
    @Timwi and the answer is listening to you and giving top priority to your bug reports?
    – Pekka
    Aug 2, 2010 at 12:32
  • 1
    @markj this is a pretty egregious case of not-a-bug, nowhere close to the grey area. I have no particular objection to the feature request, I object to the strident, combatative tone and whining in the original "bug" report. Aug 2, 2010 at 21:26
  • 3
    The wording may have been a little strong, but honestly, you should hear my users sometimes! Sometimes passion just means the user is keen to achieve something with the software. I think Timwi is hinting at a genuine truth though. We can't judge usability issues for new users solely by votes on meta. Almost everyone here is an experienced user, and experienced users are often bad at imagining the user experience of a new user.
    – MarkJ
    Aug 3, 2010 at 16:42
  • 4
    First off, a dismissive opening comment that implies believing in your own opinions = bias is not very friendly. Second of all, his entire point was that the Meta community is not friendly to users who are trying to express usability bugs. Your response is "there are no usability bugs if you can't convince the meta people". You are essentially confirming his beliefs by saying you aren't willing to recognize the problem.
    – devinb
    Aug 30, 2010 at 17:17

Regarding your specific bug reports and the things you are complaining about, they may be partly due to misunderstandings.

  • Reports closed as duplicates are not declared "not a bug." It is just that an identical or very similar report already exists in the system. There will usually be a link to that duplicate.

  • Reports re-tagged status-declined or status-bydesign have been reviewed by Jeff himself, who said "won't fix". You won't get a better channel to the makers of SO than that - if he won't fix it, no bug tracker in the world is going to change that.

To be fair, undisputable bugs (as in malfunctions) always get fixed pretty quickly in my experience.

Also, "Usability bugs" are not always as clearly bugs to others as they are to those who report them. I, for example, find the envelope a fine solution to navigate to the "recent activity" page, even though I agree it is not immediately discoverable.

  • 3
    Of course discoverability is irrelevant to everyone who has already discovered a given feature. I don't think that in itself is an argument against making something more discoverable. You have to consider everyone who hasn't yet learned the feature. To me the envelope was quite a bad UI initially - it took more than one try to really associate it with the Recent Activity page. I only thought of it as a "show me those new messages" page initially.
    – RomanSt
    Aug 2, 2010 at 10:45
  • 1
    It is ironic that you should use the word “undisputable”. I assure you there are usability problems that any user experience engineer would consider absolutely undisputable, while developers wouldn’t. Who’s right? I’d say, if it’s a user-experience issue, it’s more likely to be the user-experience expert.
    – Timwi
    Aug 2, 2010 at 10:53
  • It took me 5 minutes to find the responses page too. Who's right? The users are right. If they can't find a feature they need, it's not discoverable enough. Any good user experience engineer will know to do usability tests to see how the users get on. It's a quirk of human nature that many excellent developers are poor at usability, and many highly advanced users can't imagine what a product is like for beginners.
    – MarkJ
    Aug 2, 2010 at 11:42
  • @Timwi by "undisputable bugs" I was referring to malfunctions (like "Login doesn't work using browser X when setting Y is set to Z")
    – Pekka
    Aug 2, 2010 at 12:26
  • 2
    And @Timwi by "undisputable" I'm not talking about Developers vs. User Experience experts: I'm talking about a feature that has worked so far for several very well-functioning communities. While I'm not disputing it can (and maybe needs to) be improved, I find it arrogant to call this a bug.
    – Pekka
    Aug 2, 2010 at 12:36
  • @Timwi User Experience Engineers do not highlight bugs. They highlight problems. I attempted to clarify the difference in my response.
    – devinb
    Aug 30, 2010 at 21:08
  • @devinb: Call it what you want, but meta doesn’t have a [problem] tag.
    – Timwi
    Aug 31, 2010 at 7:23
  • @Timwi, You are stubbornly adhering to definitions that no one else here agrees with. It is only a "bug" if you are willing to claim that you are correct and everyone else is wrong.
    – devinb
    Aug 31, 2010 at 11:43
  • 1
    Cc: @devinb @Timwi; To: all; Why doesn't meta have a [problem] tag?
    – SamB
    Nov 28, 2010 at 22:26

You must log in to answer this question.

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