The line between on-topic and off-topic questions has always been on the soft side here. As time goes on, we seem to get further and further away from a concrete line or simple definition of what is on-topic. As a result, I feel like the quality of questions might be slipping. I think that by being more firm about what is on-topic, we can clean things up a bit and improve the quality of questions.
I'd like to propose a simple one-question test for any question, new and old on the site:
Is your question about using a specific web application?
Note that our FAQ says:
What kind of questions can I ask here?
Web Apps - Stack Exchange is for expert and advanced users of web applications. If your question generally covers …
Using Gmail, Facebook, Twitter, Google, or any other website which behaves like an application
Browsers and their features relating to the use of a web application (Greasemonkey scripts for a web application, etc)
So, our FAQ already mentions that an on-topic definition of using web applications. It then goes on to mention an on-topic definition of using [specific popular web application]
Thinking about this issue made me think of something I used to say at my old job. I was a web developer and as such, supported the internal, intranet applications I developed. However, I would get general internet questions, browser questions, anything. If it related to the web, people thought it was a question for the Web Dev department. We had to tell people that we don't support the entire internet. I think this relates here. Make your question specific to a specific web application. If your question is general in such a way that it is asking about the entire internet, it is probably not a good question for our site.
Note that this also includes recommendation questions. "Is there a web application that lets me view when the stars align in the right formation such that I'll win the lottery?" is not a question about using a specific web application. It is asking about something that may exist in the entire internet. We don't support the entire internet.
Obviously I'd like to get some feedback on this. I think it could be a very clear guideline for creating new questions, as well as closing questions.
Examples of existing questions that I'd propose closing because they ask about the entire internet:
Does anyone know of an online SIP Voicemail service?[closed]
- A Web application that acts like a digital version of a book
- A web app where you can pay to have surveys answered.
- Todo with user input
- Are there any good cocktail recipe suggestions web apps?
Rating service for any web content[closed]
- Online editable comparison grid builder
- Free issue reporting software?
There are also questions about alternatives. These tend to be either incredibly subjective and vague, or incredibly specific to a user's needs (if they bother to specify).
Now, a case can be made that recommendation questions can be specific and thusly the category should be allowed. Jeff's question, Web video sharing service with "fair use" protection? is used as an example for this. Look at some of the answers it got though (out of 11):
- Put it on your own server.
- Vimeo is a good choice.
- I would try MetaCafe or Dailymotion.
- Why not use blip.tv?
So even though the question has details in it, the underlying question is still:
Where can I host a video of what I believe to be a fair use 1 minute excerpt of a movie without it constantly being at risk of being pulled?
...which is still attracting a high percentage of bad answers.
Yes, it is a good chunk of the questions we get. Are these the kind of questions we want though? The examples I pulled are from the first page (50) of /questions sorted by activity. When someone first comes to the site, are these the questions that give a good impression of the site and draw users in?
Is there a way we can keep these questions but still improve the quality? I am all ears.