[meta] Is Answers doomed?

UA is not dead, but it is not healthy either. Healthy sites grow, change, and evolve to handle new situations. I see very little investment in this site. The only significant change in the infrastructure I recall in the last 18 months is the addition of line numbers to the source. It took 11 months for something as critical as 'Send Message' in the moderation queue to be fixed, and I've noticed that reported spammers are not having their accounts suspended. There is a real disconnect between the people with ownership of the site and the community who regularly participate in the site.

Demographic changes Changing demographics is the issue most talked about most on the site. Unity is popular, and there is a rising tide of beginning programmers attempting to use Unity. Unity Answers is not evolving to handle the change. Schools are adopting Unity as the platform for programming classes. An increasing number of artists with little or no programming experience are trying to bring their vision alive through a patchwork of found scripts. Because of its visual nature and focus on games, Unity is a tempting starting place for any beginning programmer.

But many beginning programmers don't know how to ask a technical question. Often beginning programmers don't know how to research a technical issue (or asking first is easier than research). Beginning programmers don't have the experience to take an answer or multiple answers to a similar question and remold it to their needs. The result is rising number of duplicate and poorly asked questions. In terms of moderation, UA operates the same today as it did 24 months ago...a few rejections, a few closures, but in general everything is approved . While we may point to the FAQ and the vision for the site, in reality there are few standards, and whatever standards exist have routinely ignored or violated (including by high Karma individuals). Each individual moderator does what he thinks is right (or what he feels like doing) with little guidance.

Other Issues While the changing demographics is the elephant in the room, there are a number of other issues that could be addressed. Here are a few of the top of my head:

• Mediocre searching (both Google and UA) make finding past answers difficult. I can't count the number of times I've gone looking for an answer to direct someone to, only to fail to find it.

• Users using 'Answers' as comments (there have been a couple of propose fixes).

• Questions answered by comments that remain 'unanswered'.

• The number of questions that are never resolved/accepted.

• Closure mechanism does not reflect the values of the site

Addressing the problems First, there must be some resources made available to charge/evolve the site. While there are things that can be done without changing the site, these changes become cumbersome without site support. We need a sense that if we debate ideas and come up with something that has community support, that someone is listening and will give us feedback and resources concerning these ideas...whether they can be done, and if so on what schedule.

As for content changes, there is something interesting about this site...ninty plus percent of all questions that anyone would consider questionable pass through moderation. So all it would take to change the nature of the questions to be answered would be standards embraced by around a dozen active moderators. In reality what is needed is a set of specific community standards that can be applied semi-objectively. This is one of the problems I've had with other pushes to change content...there were no clear standards for what was being closed.

Personally, I'm not in favor of placing limits on the sophistication of questions. Beginners should be able to get answers here. But I do think it is appropriate to push back on the quality of the question (for everyone including beginners). Poorly asked questions cause a number of problems on this site. For me a well asked questions has:

• A title that reflects the content of the question - Google places a lot of weight on the title. Without a good title, it is difficult to find (and build upon) past answers.

• Good context - what defines 'good' varies from question to question, but a solid description of the problem, source when applicable, a description of how things are setup in the scene when needed, how the character is being moved, etc. We should be able to answer the question without a number of requests for information and/or a number of guesses at the answer.

• Some indication that the OP did some research or other investment in figure the problem out for himself before asking for help. This might be a script attempt or a reference to other material.

In addition to 'good question' standards, there are other categories of questions that we should at least discuss if they should be supported. Here is a quick brainstorm of some categories. I think several of them should get a Meta question so we can decide if this is something we want to do:

• Conversions of whole scripts between Javascript and C# - I don't have any trouble helping someone who has made an honest attempt and needs some help, or if someone posts a short section of code that are causing trouble. But there is plenty of material available for anyone, including beginners, to take a shot at the conversion. Isn't it fair to ask someone to at least make an attempt at conversion before asking for help?

• Write scripts for you - I'm including both explicit and implied. Many questions say, "I need a script," but there are also the questions that give a long laundry list of features. Many of these are closed, but many are also answered.

• Not enough context to answer the question - Isn't it reasonable to reject questions that don't have enough information for a target answer? The OP can fix their question and ask again.

• Lack of research - is it unreasonable to suggest that OPs first make an attempt to find the answer before posting a question?

• Instructional materials request - Comes in several forms. Typical one start out with "What is the best book..." Or, "Is there are tutorial for...".

• Legal questions - The questions concerning Unity directly have been asked and answered many times. For other questions, we are not a forum of lawyers. Do we really want to support these questions?

• Multiple technical questions - Often comes in a too general question that has multiple things that need answering.

• Discussion/design questions - I'm about the only one who closes these questions.

• Presentation issues: poor title, source pasted in as an image, large blocks of bold or upper case text, unformatted code. Are these reason enough to reject a question and ask the OP to fix the problems and resubmit the question?

• Off topic - we answer a lot of non-unity-specific questions: C# language issues, geometry/vector questions, game design questions, and most of all debugging questions that are just dealing with programming and have nothing specifically to do with Unity. Do we want support all of these?

Unity Player specific questions - installation or other issues. Some get closed, some get published answered.

• And I'll bet there are a few more categories.

  ---

Once we've defined the standards, these standards need to make their way back into the reject and closure mechanisms. I'd like a single line closure/reject as we have now for closure, but the closure text would have a 'more info' link to short paragraph describing what the closure means, how it can be fixed or addressed, and perhaps in some cases, links to material to help. User should have a common experience. It is not a specific moderator applying some arbitrary standard. It is Unity Answers' standard being applied in a semi-uniform way. In addition closure and rejection lists that reflect Unity standards would give new moderators needed guidance on what should and should not be published.

No. I used to ask the very same questions that veterans hate, but that was when i didn't know an ounce of coding. I would depend on people better than me to help, and after a while, I got a hang of it, and now I'm a decent coder because of the people who were kind enough to do so.

Unlike OTHER people, I haven't forgotten that I once didn't know anything about coding, so I feel good when i provide thorough answers and entire scripts to newbies because I still remember. The more you help the newbies, the more people who actually learn how to code there will be to help out the newbies once again.

I interpret most of the hate towards people asking silly or simple questions, is just whining. What I'm saying is that it's a cycle that people shouldn't get snippety about, But I guess that's just my opinion.

Cross posting from the other long 'What can we do to improve...'

http://answers.unity3d.com/questions/18806/what-can-we-do-to-improve-the-quality-of-unityansw.html?page=6&pageSize=5&sort=votes

A quick win to improve the experience: put a filter option on Questions / Most [Up] Voted to allow selecting only those in the last day, 3 days, 7 days, 14 days, 30 days, including option to show answered/unanswered. That would separate most of the wheat from the chaff with a simple change. Interesting questions upvoted would show, hundreds of how do I press E to turn on flashlight won't.

I haven't posted a lot, but every time I have it has gotten me out of a situation I was stuck in for a very long time. For those new to Unity answering the "stupid questions" is really important.

Probably not.

The whole site seems to run using comments as answers (and I am just as guilty as the next). Maybe we should all try harder to post real answers and not comments? I use comments when I am not 100% sure I am answering the question. :-(

Maybe we should try harder to close questions that are not new/original? I've started picking any question that lists a compiler error number, and add a tag with that error number. I then add a comment with a link to the tag page, and close the question as a duplicate. Is that too restrictive?

Hey Graham, regarding the "comments not answers" phenomenon -- "QA board reverse snobbery" -- , it's an internet wide issue

Some idiot was talking about it here for example: http://meta.stackoverflow.com/a/253687/294884

Note that the new-new thing is just editing answers; "comments-not-answers" is like last Tuesday's cool, you know?