Should soft questions be community wiki?

Question

Recently, we had a discussion on meta about soft questions, and now there is a precedent on the main site. Should such types of posts be community-wiki?

Pro: a) there is rarely a single best answer; b) voting on CW does not affect reputation score after the question has been converted into CW. If we want the reputation to reflect mostly comptenence in physics, this might be an argument for CW for soft questions.

Con: c) As Joe Fitzsimons points out, cstheory tends not to turn career-advices type of questions into CW.

Please, share elaborate on these arguments and/or add your own.

This post has been migrated from (A51.SE)

Comments

For anyone who doesn't know what community wiki is, see http://meta.stackoverflow.com/questions/11740/what-are-community-wiki-posts

This post has been migrated from (A51.SE)

Answer 1

Well, I wrote the blog post on it that Logan cites. It probably isn't too surprising that I have a coin to toss in this fountain.

Mathematics also uses soft-question as an "always Community Wiki" option. Consider, on Mathematics, they do have soft questions, but they are not overwhelming of their normal content. Indeed, their top questions by votes are largely populated by non-soft questions, both extremely appealing and technically intriguing. There's a lot of good stuff there that proves that Mathematics is a site about hard mathematics, not simply a place to discuss softer points. And Mathematics currently rounds out at the top of all Stack Exchange 2.0 sites in question volume, standing on this platform of hard questions.

---

My personal stance is that it is a bit off-putting to try and have an entire caste of questions be considered "On-topic, and acceptable, but unworthy of reputation". Which is what you get if you try to say that there's a whole tag or class of questions that must be community wiki. It paints the site in a haughty light, one where the community tells you whether or not your contributions are "worthy", which I don't think is a very healthy light to be depicted in. For a site that already has restrictions on entry (research-level only), it makes the site even less welcoming. Community wiki shouldn't be seen as a reputation filter.

If it isn't about what is "worthy" of reputation, it confuses me a bit, to hear "Making soft questions not earn reputation makes reputation a better metric of who is active in addressing theoretical physics questions" in the same breath I hear "Soft questions should be allowed on a site that is about theoretical physics questions". Are soft questions considered a part of theoretical physics or not? If they are, why are they not contributory to reputation? If they are not, why are they allowed on the site?

This post has been migrated from (A51.SE)

Comments

I really wouldn't compare this site with maths-SX. It is *really easy* to ask an on-topic-and-popular-but-not-soft maths question. It is also really easy to ask a not-quite-on-topic-but-popular-and-soft research physics question. What is hard is to ask an on-topic-and-popular-but-not-soft research physics question. Far better to compare with MO where soft questions are tolerated so long as they don't get out of hand, and the no-reputation-facility of CW is used to ensure that no-one benefits from the otherwise excessive votes that they get.

This post has been migrated from (A51.SE)

---

Regarding the "haughty light" paragraph, what we find at MO is that not everyone reads the sign on the door! Moreover, that sentence doesn't make sense. Surely the voting system *is* the community telling you whether or not your contributions are "worthy". What this (using CW like this) is saying is that the moderators are empowered to act against the "general consensus" when they view it is in the greater interests of the site - something that isn't always in accordance with the majority view.

This post has been migrated from (A51.SE)

---

@Andrew If the votes do the trick, why are we making decisions on *when* the votes actually have that impact? However, since my point didn't make sense, I'll try clarify.

This post has been migrated from (A51.SE)

---

You're saying that we have votes to determine what the community finds useful, which is correct. We also reward useful contributions, that's what the system is pretty much founded on. However, in marking community wiki *explicitly* just to *deny* reputation (and not for the otherwise intended purposes of community wiki), one is saying that the usefulness of the contribution is entirely irrelevant to what the author gets in return. Essentially, the site is reaping all the benefit of hosting what people want to find, while the author isn't worthy of being recognized for the contribution.

This post has been migrated from (A51.SE)

---

My point is that the votes garnered by a post are only a crude measure of the worth of that post and that sometimes a corrective factor is needed. (I didn't say, "**just** to deny reputation, by the way.) A popular post does **not** mean, "Lots of people want to know the answer to this.". It would need a serious change in how reputation and votes were computed for it to mean that. So, yes, your last statement is correct, except that I would say that the person is reaping all the benefit of posting their dubious contribution in a public place, while quality of the site as a whole goes down

This post has been migrated from (A51.SE)

---

@Andrew If the quality of the site is going down on reception of these questions, why are we keeping the post in the first place, then?

This post has been migrated from (A51.SE)

---

@Andrew: if a post is driving down the quality of the site, the solution is to *remove it*. This goes for hard technical questions and soft "field/discipline" questions. Why on earth would you think it is ok to let a bad post drag you down so long as no one gets rep for it?

This post has been migrated from (A51.SE)

---

(I'm waiting for the "would you like to move this to chat" link.) To both of you: because it is never about *just one question*. Questions do not exist in isolation. Sure, if we could identify *the* snowflake that started the avalanche, we'd blast it with the hairdryer. But as we can't, we have to simply watch for where too many accumulate. (I'm happy to discuss this further, but am feeling that the format doesn't fit. If you would like to continue, let's open a chat room for it.)

This post has been migrated from (A51.SE)

Answer 2

I was writing a similar question when I saw this.

Stack Exchange's policy has been to avoid CW. See here. Their argument is that if the post gets upvoted a lot, no one really cares whether it's CW or not. It sets a precedent that the best questions on the site are soft questions, since the hard ones will tend only to be upvoted by those who understand the question. MO has had similar discussions on its meta, but I can't find the post.

That being said, I disagree with their reasoning in our case. The options we have are either allow soft questions as CW, allow them without CW, or ban them altogether. The last is by far the worst, and if we're going to allow soft questions I'm all for people not gaining reputation from them because it makes reputation into a better metric of who's active at asking and answering theoretical physics questions. The fact that these questions will be upvoted a lot seems a problem in either case, but one that is entirely removed from the question of CW.

This post has been migrated from (A51.SE)

Comments

+1; I side with your your opinion. Soft questions as CW looks like the best available policy, at least this stage.

This post has been migrated from (A51.SE)

---

+1, interesting reasoning.

This post has been migrated from (A51.SE)

---

I agree. This is also in line with MathOverflow practice.

This post has been migrated from (A51.SE)

Answer 3

I believe soft questions (if they are not in abundance) are important. Whether a question is CW or not, in my opinion this is not the right question. A question should be CW if it does not have a single answer, like in this case. Admittedly, soft questions tend to be of this kind, but not necessarily. But a question should not be CW only because it is soft.

This post has been migrated from (A51.SE)

Comments

As a side note: not having a single answer is not a good reason to make a question CW.

This post has been migrated from (A51.SE)

---

No, "has not a single answer" is *not* an indication for Community Wiki. I recommend reviewing [this blog post](http://blog.stackoverflow.com/2011/08/the-future-of-community-wiki/) for a better perspective on community wiki.

This post has been migrated from (A51.SE)