6

I was looking at the Area51 page for Raspberry Pi SE and saw that our question-answer ratio isn't all that great. It is listed as "Needs Improvement" on the page, and on the sidebar here it says there are 77% answered.

area51 screenie

Can we go back through the un-answered queue and see if we can't get this thing up to par?

Also, a question that I have been wondering about, how/when does a site graduate Beta?

  • Geez. Im trying! – Loko Jan 21 '14 at 15:28
  • It's fine :D We are getting somewhere, just thought I draw some community attention to it. – RPiAwesomeness Jan 21 '14 at 16:46
2

Can we go back through the un-answered queue and see if we can't get this thing up to par?

When doing this, keep in mind the option to close questions -- one of our issues here is because the traffic is low, it is hard to get to the pre-requisite 5 close votes.

Below are the 10 most recent unanswered questions > 6 hours old as of this writing. It might have been better to look at questions > 24 hours old, since the longer a question goes unanswered, the more likely it is to remain so.1 However, I think a major reason for unanswered questions is low quality questions, and if those don't get an easy answer fairly soon, they might as well be caught and closed ASAP.

Out of the 10, I voted to close 7 and answered 1, leaving only 2 unanswered. That may seem a bit zealous, but I think some of my points here are worth considering -- if we closed more low quality questions, we would improve the general quality of things on the site, and get rid of potential tumbleweed.

This doesn't have to apply to just recent questions, obviously.

  1. https://raspberrypi.stackexchange.com/questions/12713/setting-up-raspberry-pi-so-as-to-connect-to-a-server

    I voted to close this because it is too broad. Asking "how to stream data from a computer", or "how to store information in a database" without more specifics shows a complete lack of effort in prior research; while someone could answer the question by regurgitating some basic material easily found in wikipedia, I'm sure that neither of these topics needs yet another internet posting.

    "Poorly researched" is the #1 tooltip for downvoting questions, but in terms of getting the point across, closing it is probably more effective -- remember, the OP will automatically get the attention of a moderator if s/he then chooses to edit and improve the question. Downvoters may never bother to return and change their minds even if the question is improved and corrected.

    So if you see stuff like this, vote to close.

  2. 1 motor spradically stops when going forward

    Seems valid.

  3. PWM not found in Occidentalis v0.2

    Seems valid.

  4. https://raspberrypi.stackexchange.com/questions/12709/raspcontrol-help

    I voted to close this question as unclear. While superficially it seems clear, if one considers the steps involved in the problem -- significantly, installation of a web server, which is not even identified -- then it starts to seem more like something very lazy.

    Considering I did not make a comment ("We need more information"), that may seem a little unfair. However, my word is not the last word, either, so by the time we get to the 5 close votes, some of that "unfairness" will likely be resolved.

    I mention that because I think that if you believe something should be closed, then you should vote to close even if you don't have time to make a comment, or anything in particular to say beyond having chosen a reason. This is significantly different than downvoting with no explanation, esp. since a close can be completely reversed by the community.

  5. https://raspberrypi.stackexchange.com/questions/12708/forward-audio-from-usb-input-to-jack-output

    Ditto the last question -- I voted to close as "unclear" with no further explanation because the OP was to lazy to even explain what s/he means by "USB input".

    Of course, someone could prompt the OP for more info, which is fine; I do that sometimes. However, I do think it encourages an environment whereby people think it is okay to be say as little as possible because "if more information is needed someone can ask for it". Closing a question makes it clear that's not the ideal attitude.

  6. Problem installing Ruby on Rails server on Raspberry Pi

    I voted to close this as "unclear" too. The question seems to be centered on the idea that some Ruby component may have a minimum RAM requirement, but rather than researching that or asking the question somewhere more appropriate (e.g., a Ruby forum), the OP simply decides to ask "Has anyone got this working on a Model B?" If we say that is okay, watch the quality of the questions go out the window.

  7. Trouble connecting to wireless network two rooms away

    I voted to close this as "too broad" because it is. Notice also the OP has not bothered to reply to comments.

  8. omxplayer rtsp stream over composite output

    This one was unanswered, but I added one. It is not a great answer, but one of the points I make is:

    If [...] you want a more detailed answer to the question, How do I select the active video output? then ask that as a separate question.

    I often think about doing this, and I'm going to start more often. The other option is the question becomes completely localized, because what should have been something general (and more widely useful) gets answered in a Q&A in the comments.

  9. https://raspberrypi.stackexchange.com/questions/12698/cec-on-sharp-tv-remote-xml-fails-after-tv-power-off

    I voted to close this as "unclear". The effort is so low here the OP never actually asks a question. S/he just sort of implies one, kinda. I guess. Maybe.

  10. https://raspberrypi.stackexchange.com/questions/12690/unable-to-see-webpage-for-webcam-stream-update-can-now-see-stream-in-vlc-but-s

    I voted to close this one as "unclear" although "too broad" would also fit. Adding "updates" to an unanswered question is a poor practice in an S.E. context. Someone needs to rethink but not "out loud", as it were.


1. And 6 hours is pretty brief; I'm sure there are knowledgeable regulars here that check in every few days and go over all the questions since their last visit, as opposed to people like me who just look at the most recent questions regardless of how long it has been. Someone should graph how likely a question is to be answered based on how long it has gone without one.

  • I agree with all of these flags and flagged them the same. Hopefully this site can be improved! – RPiAwesomeness Jan 4 '14 at 15:55
  • 1
    Often the question is "answered" in Comments (some of my questions have been). There is then no way to "accept" a helpful answer. – Milliways Jan 5 '14 at 6:10
  • 1
    @Milliways You can create an answer with what solved it and comment, asking the OP to post a self-answer or mark yours as answer. – RPiAwesomeness Jan 10 '14 at 15:14
1

The guys over at Code Review also had a similar problem back in the day. Then they came up with a great idea and set a goal for themselves.

Call of Duty - We're on a mission

Mission Status:

This section tracks the metrics at regular intervals, at or around 12:00AM UTC. Each week is divided in 3 rounds (3-2-2 days), where the number of Questions with no answers /zombies and the number of Questions with no upvoted answers /targets at that point in time, is used to calculate the net score for that round; the net score is then divided by the number of days in the round.

It's important to understand that those are net figures; they include incoming zombies so they are inherently skewed (downwards). We'll get a much better picture once the data explorer allows us to dig up our own data. I'm including the total number of incoming zombies at the end of the week, under the debrief section.

Incoming zombies are acquired by searching for questions here "is:question created:yyyy-MM-dd"

They also came up with their own memes.

Some "memes"...

A few expressions have come up recently, thought I'd share them here (feel free to add, this is CW!)

  • Running out of ammo: when you have exhausted all 40 votes for the day.
  • Head shot: when you're out of ammo and post a target in chat, and then someone puts in a vote for you.
  • Waking up the dead: when posting an answer on a zombie question causes the OP to come back to CR and accept your answer.
  • Killcam: when you've just shot a zombie with a great answer, and now the zombie is still counting as a question with no upvoted answer so you post your own target in chat. Use with care, abusing the killcam is said to trigger thoughts of "rep-whoring".
  • Zombie looting: when you've downvoted (upvoted?) a question that ends up being deleted (on the same day), and then you get your vote back. The vote count doesn't deplete though, which means zombie looting allows you to appear to have voted like 43 times on a single day, and steal the show on the voters tab.
  • Napalm Strike (or Carpet Bombing Airstrike): when you upvote the question, and all answers are good enough to each warrant an upvote, too.
  • Double tap: when you answer a question that has another answer (with no upvotes).
  • (removed): instead of removing a chat message, why not just say (removed)? Started when Jamal used his vacuum cleaner on a couple of off-topic messages in the chat. Beware though, it is said a baby unicorn dies everytime it is used.
  • ding!: The sound the chat makes when you ping someone.
  • theoretical: when you're out of stars, you can still give a theoretical star. If you're lucky you'll even get theoretical thanks, so you can say "you're theoretically welcome!". It's all about theory. Originated from a question that was said to have theoretical code, which should go into a theoretical compiler before being posted on Code Review.

Anyways, if we could get something like that going on this site, I think it be very helpful in getting us to graduation.

  • Precisely! There's also the added goodness of getting rep on the side! Make it fun and they will come I guess. Lets try this! – RPiAwesomeness Jan 1 '14 at 23:22
  • As for the graduation, do you have to get it and keep it at a perfect level for a certain amount of time? – RPiAwesomeness Jan 1 '14 at 23:23
  • 1
    @RPiAwesomeness Code Reviews statistics have been near perfect for a while now, yet they have not graduated. Not sure why though. – syb0rg Jan 2 '14 at 0:09
  • Could it be you need to be totally perfect? – RPiAwesomeness Jan 2 '14 at 0:30
  • @RPiAwesomeness I doubt it. I have seen sites that have graduated that don't have perfect statistics. Click on them to see all of the different statistics, and you will see that some of them have catagories that still "Need Work". – syb0rg Jan 2 '14 at 0:33
  • What is considered an "incoming" zombie? Reading over this and I can't quite make it out... if you could explain it to me/tell me what the different terms mean, I have an idea in the works... – RPiAwesomeness Jan 11 '14 at 2:36
  • @RPiAwesomeness Incoming zombies would be new questions. What other terms do you need explained? – syb0rg Jan 11 '14 at 2:54
  • Ah. That makes sense now... I don't know any other terms I need explaining, just how the score would be best calculated...hmm. – RPiAwesomeness Jan 11 '14 at 2:57
  • @RPiAwesomeness If you look at the original post on the CR meta, you will see how they deduce the score. Basically, they look at some numbers in the unanswered tab and compare them to the previous "round" (a round being a few days long). We can discuss this more in the chat room if you want. – syb0rg Jan 11 '14 at 3:15
  • Sure thing. Be right there with my idea... – RPiAwesomeness Jan 11 '14 at 3:28

You must log in to answer this question.

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