Would it be possible to adapt the software powering Stack Overflow for other purposes beyond the tradition Q&A? For example, to run a survey site for public opinion polls where the asker of a question would usually also provide (all the) answers? Users would still vote them up or down.

This would suggest some additional features, e.g.:

  • Delay the publishing of the question until (all) answers have been written.
  • Predefined styles for A/B (yes/no, good/bad …), Likert scale (“strongly disagree” …), numeric estimation, category assignment, color value, price range and other standard questions.
  • Questionnaires: Sets of related questions.

It’s a bit similar to how moderators are elected.

Also, would a site like that fit under the stackexchange.com umbrella?


No, a site based entirely on a survey of user opinion — and similar ideas soliciting alternative content — aren't currently a good fit for a Stack Exchange Q&A site. Apart from the technical considerations, the primary reason we don't currently host "public opinion polls" and various alternative uses for the Stack Exchange engine is the software isn't fundamentally designed for that purpose. The core software is meticulously focused on a very specific principle — to create a highly-collaborative work of sharing knowledge by having users ask very specific questions that can be answered definitively in the space of a post. When you try to swerve outside those intended roles, the system will almost certainly get in the way. Tenaciously.

Square peg, round hole

We're starting to see a marked increase in proposals where folks want to vote on a variety of other stuff. When Stack Exchange is seen as a simple voting engine, the conversation quickly turns towards how they are going to redefine what is meant by a "question…" and what type of content will fit in those slots called "answers". Comments often become something else still — and you will probably have to explain what a vote actually means in the context of this site (most of these alternative Q&A's have no concept of a correct answer). Of course, most of these stories won't match what the UI is prompting for generally.

A poor end-user experience

Even if you're willing to accept these limitations, the majority of incoming users will not be privy to these meta discussions. Experienced users are already tired of (re)explaining again and again how a typical SE Q&A works; that problem goes up exponentially when the UI itself steers users in the wrong direction. Stack Exchange includes a lot of infrastructure designed to create great Q&A. But what does "closed as primarily-opinion-based" mean on a site soliciting reviews or polls? Or how does someone's "accepted answer" break the ordering when asking for a best-of people's choice list? Those are just two examples that barely skim the surface.

You can moderate away a lot of the expectations people have become accustomed to elsewhere, but when people are constantly mis-using what the system is (apparently) trying to steer them towards, the whole thing becomes one endless harangue as unsuspecting users are just trying to use this poorly-outfitted site.

Maybe someday… but not now

For the foreseeable future, we're not going to see a lot of dev time set aside to accommodate these expanded use cases. Without a lot of customization and redesign, we will be doing these subjects a terrible disservice. The system itself will lead users astray from how the site is supposed to work. The 'help center' will be filled with factual errors, while much of the pop-up and side-bar guidance won't reflect how users are actually supposed to use this site. We've been able to accommodate a few sites with special needs, but as folks continue to push these special use cases even further into the "review" and "idea-generation" space, many of the system prompts go from somewhat-off to outright incorrect.


You must log in to answer this question.

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