Should we open a sandbox?

Question

In How can I let my boss know her methods for management aren't useful? the OP asked for a chat session to improve her question.

After a week or two thinking why we don't have our sandbox thread, is it time to create one?

We have 18 questions closed with various reasons this month, so I think sandbox will help new members to improve questions before posting them.

Here is a discussion on issues that Worldbuilding Sandbox as a post, and possible solutions.

Note that we don't have to follow how Worldbuilding do their sandbox! We have our own issues and challenges!

Answer 1

It's an interesting idea. A sandbox allows you to avoid some terrible questions - and, more importantly, turn bad questions into good ones. I can tell you from experience on Worldbuilding that it really can make a difference for some users, and I'm pleased with how ours has gone.

There are, however, some issues that we'll need to think about:

• The sandbox will need to be cleaned up continuously; some people will abandon their drafts after a while. Currently, on Worldbuilding, we may delete posts that have not been modified for 30 days.
• It's not going to help everyone. Some users won't learn how to write a good question. Most will, but some won't, and it's important to understand that.
• The newest users won't have enough reputation to use meta, unless we lower the rep level.
• There are some functional problems with a sandbox, including visibility and ease of use. We still haven't solved those, although we're working on it.

A good sandbox, ideally . . .

1. Is visible.
2. Is well -monitored and -moderated by both the community and diamond moderators.
3. Is used regularly. If a sandbox turns into a ghost town, then maybe it's not needed.
4. Actually works and produces high-quality questions. I personally think that a question born in the sandbox should eventually be better than average.

To make this thing work, we need to satisfy all of those points, and get around the problems I listed above. It's not as easy as just saying "Hey, let's have a sandbox!" It takes effort.

If people are willing to work at these, then I'd agree with this. But we need to keep in mind what this will require.

Answer 2

I think this is a great idea.

Having a sandbox would allow not only new, but all users with questions that they're not sure how to ask get them fleshed out, and ready for the actual site. As a point, people who answer questions tend to stick to very specific points on questions, not actually answering the question but addressing different and too specific of issues. If we can find the balance between an answer/question that's too broad, and one that's to specific, the site might be more useful for those looking to use it as a general resource.

Answer 3

I don't want to be overly negative, but there are also examples where a sandbox isn't helping much. There is a sandbox on Programming Puzzles and Code Golf, where users can post their questions for others to review, before they get posted for real. In theory.

In practice the sandbox gets little traffic in the form of reviews, leaving most of the criticisms a question will get to wait until the moment it gets posted on main. Many questions then still get put on hold for being unclear, while they have been in the sandbox for a long time.

There are a bunch of recent posts on meta regarding the sandbox on PPCG:

• How can we get more feedback on our Sandbox questions?

• We need to talk about the Sandbox

On the other hand, nothing is stopping users from not using the sandbox. If a (new) user is unaware of the site policy about scope and question quality, there is a real chance that they are unaware of the meta, let alone the sandbox. Pop-ups and pointers to a sandbox seem nice, but then again, we have similar popups about exactly scope and question quality and that's not working. Heck, PPCG even has 10k+ rep users that fail to use the sandbox yet post terrible questions left and right.

I'm skeptical about how much this will actually help new users, whether it will get enough traffic to be of any use and especially whether it wouldn't be more effective to be more clear about what we expect of questions asked, instead of deferring users to a sandbox.

Final remark inspired by PPCG sandbox:

Make sure that if there is a sandbox, it actually gets used by enough users to be representative of the main website, so that it does actually good to potential questions, instead of something to slap bad questions with.

I am not necessarily against trying, but it will require a commitment from established users. Not only now, but also in a few months/years time. It shouldn't end up as a place to redirect bad questions, just to get them off the main site, and then forget about them.

Answer 4

My first argument is that it may help some people when there is a sandbox, of whatever form. _{The form may be discussed further.}

If new users use it, that's great! If not, at least it will help other people.

Regarding HDE's excellent points:

Is visible.

We need community effort to do this.

• Remind new users that there is a sandbox to refine their question before posting (or maybe after it is closed).

  even if it is not obligatory, (it isn't on PPCG), but bad questions still get comments about how the question should've been posted in the sandbox first.^Comment

• Having the link to sandbox in the help center.
• Having Sandbox link in featured post of the meta will help with the visibility and the ease of access.

Is well -monitored and -moderated by both the community and diamond moderators.

Pointing new users to the sandbox only works if that sandbox isn't dead. A sandbox with only questions and nobody that reviews the questions is worse than no sandbox.^Comment

I agree with the first sentence. A sandbox needs community effort to keep it alive. If you agree that we should open a sandbox, join us as reviewers. It does not require personal commitment. Just do the usual if you find a question that is not clear, or is not a good fit for our site: comment for clarification and give suggestion for improvement.

However, I disagree with last sentence.^{See "Failure?" below}

Is used regularly. If a sandbox turns into a ghost town, then maybe it's not needed.

Yeah, maybe ^{See "Failure?" below}.
It's a tool to help us get higher quality question. If it's there, someone will benefit from it being available. Of course, it needs both reviewer and user.

Actually works and produces high-quality questions. I personally think that a question born in the sandbox should eventually be better than average.

With the help of feedback and guidance from community (the reviewers), of course they should be! But we don't even know if it's effective here because we don't start it yet.

Failure?

Noticed what gets repeated in the four points? Community. Sandbox needs community to stay relevant. In the case that the community decides that Sandbox is no longer needed, just discontinue or close the sandbox practice. At least it has benefit some users in its (maybe short) life.

I personally commit myself to seeing Interpersonal Skills.SE grow to be a better site, although I know I still have a lot to learn. If sandbox can be a way to help this site, I will gladly do it. If not, fine.

I don't see any drawback of opening sandbox, then closing it if it lacks of participation, except the "wasted energy" of those who committed themselves to the sandbox. I personally see that as an investment to figure out that IPS.SE does not require sandbox.

Note to potential reviewers: I don't see myself out of this site for at least 2 years to come. If I do "retire" from the site, at least I've spent that 2 years helping users, either in the main site, or in the sandbox (if it is agreed). That is also true to you.

It doesn't matter how long you can be a reviewer. Doing one review already helps. Your commitment is not to the sandbox, but it is to the growth of our IPS.SE.

Answer 5

As of Zizouz212's suggestion, we can have a "Sandbox" chatroom (need a good name suggestion), that functions roughly as "Sandbox" thread:

to help new users to shape their question to fit StackExchange format

A chatroom meets 4 criteria presented by HDE:

A good sandbox, ideally . . .

1. Is visible.
2. Is well -monitored and -moderated by both the community and diamond moderators.
3. Is used regularly. If a sandbox turns into a ghost town, then maybe it's not needed.
4. Actually works and produces high-quality questions. I personally think that a question born in the sandbox should eventually be better than average.

However

1. Chatroom needs 20 rep
2. Meta can be made accessible to new users; lowering rep requirement is possible on a per-site basis. 1

Feel free to add things I haven't thought out.

Answer 6

We already have a sandbox. Go ahead, ask a question on the main site. You'll get feedback on the question, people will edit the question to improve it... all of the things a sandbox on meta is supposed to do. Only the difference is that the question won't be hidden on some meta post where it won't get feedback; everyone will be able to see it and give feedback.

If you really want feedback before you post the question (perhaps because you want to avoid downvotes) from a limited group of people who aren't representative of everyone who uses the site, just ask for feedback on the site's chat room.

The only people who will use a sandbox are the people who don't need really need it--experienced community members who are active on meta. If the goal is to welcome new community members, then you want something like a mentoring feature.