8

We've got a good set of seed questions to start the Beta with. I think we should refrain from asking any more seed questions. Please consider deleting any seed questions you have asked if they are "softball" type questions that have not gotten answers. These do not help the site.

We have 44 unanswered questions meaning only 40% of our questions have answers. Please look through the unaswered tab and:

  • Answer any questions you feel comfortable answering. Take a moment to search Google Scholar or other resources to find an answer if you don't readily know one but are interested in the subject matter.
  • Vote to close questions you don't feel are answerable, are not good fits for the site or you feel are low quality. Please leave a comment stating why you believe the question should be closed.
  • Downvote low quality questions so that other users may identify, this can help us close them as well. Gathering enough close votes in a private beta is difficult.
  • Upvote good answers you see (answers with no upvotes don't count as answers in the metric)
  • Leave a comment asking for more details if you are unsure how a post should be answered or whether it should be closed.
8

This is mostly a cautionary tale, but "seeding" this site with questions simply to add to its bulk sounds helpful, but in all likely hood, it is actively harming this site. I wrote about this extensively in

Asking the First Questions

…but a few key points are most poignant.

I was a bit put off by the context implied by “seeding the site.” The word seeding suggests to me that we’re coming up with questions just for the sake of asking questions. My concern is, if people feel that the author doesn’t really care about the answer, the whole exercise would likely be perceived as a waste of time.

Sound familiar?

Users like to answer questions when they are helping people, and everyone loves to show off a bit on occasion. But users do not want to be given homework assignments with a bunch of busy work. This seeding has to stop. You're supposed to use this site to ask about problems you actually have; otherwise, this whole exercise will likely be perceived as a waste of time.

The Statistics

You have 49% of your questions answered (CogSci Statistics). Ouch, that's a far cry from the 93% average of the other sites (I don't recall a site ever falling below 70%). That in and of itself would be cause to shut down the site, if those stats continue. Folks who come here should have a reasonable expectation that their question will receive a good answer, otherwise it provides a poor user experience. Seeded questions draw attention away from real problems from people seeking help.

What should we do?

Stop seeding the site. Survey your content and take a good hard look at what is being asked. If it's an intriguing question that looks helpful to the person asking, answer it.

But if the question looks like the typical text book fare of any first-year CogSci class, or seeded from another forum, vote to close it as "not a real question." That sounds counter-productive and unhelpful, but read the first questions blog post again:

…you need a site where people are asking very interesting and intriguing questions, not the basic questions found on every other Q&A site…

The earliest questions on a site will set the tone and topic of the site for a long time.

  • 3
    The problem we are having is a bit more complicated. Surely there are a few questions we could close as "Not a real question", but plenty of the seeded questions are actually very well thought out ones. They were basically thought out up front, and copy pasted here. This caused them to be badly formulated, but we were able to contact the poster and he is now improving most of them. At the moment looking at the stats it might look bad, but the main problem isn't 'quality' posts. Rather, we don't have enough people to answer them. – Steven Jeuris Jan 21 '12 at 23:18
  • 1
    We could close them as "Not a real question" in order for the stats to work out better. But eventually I'm certain a lot of them can easily be reopened after they have been reformulated. Perhaps this is a good way to go? Additionally, you can't just compare CogSci with any other SE site. This is a highly research-oriented site, and it takes a long time to answer questions appropriately. I'm guessing the ratio difficulty of asking a good question vs answering it properly is entirely different from other sites. – Steven Jeuris Jan 21 '12 at 23:22
  • 2
    I do agree that expecting a research oriented site, in a field with many questions and few answers, to maintain a 90%+ answered rate is probably not ideal. It takes me at least 10-15 minutes to find proper sources and organize my thoughts to answer a post here when I already know the answer. Answering on SO, SU ect often takes a couple minutes for a correct answer. – Ben Brocka Jan 21 '12 at 23:25
  • 2
    That said I do believe we should aggressively vote to close some of the already spammy "seed" questions we have, a good start would be downvoted/zero vote questions in the Unanswered list you feel are poor fits for the site or reek of seed questions. I've deleted a couple of my own questions as well, if you have seed questions consider deleting them. – Ben Brocka Jan 21 '12 at 23:26
  • 2
    I don't think we should read too much into the answer rate statistic just yet. The site was in private beta for only 3/4 days and we had one user who asked close to 40 questions (many of which had the potential to be good questions but were generally poorly formulated). It will also be good once we move to public beta and we can start promoting the site on blogs and relevant forums to pull in active researchers. – Jeromy Anglim Jan 22 '12 at 2:52
  • I think Jeromy, Ben and Steven all make great points, specifically with regard to the relatively massive number of questions one user posted. However I was actually on one of the Stack Exchange betas which was closed (as you may remember Robert) due to poorly defined scope and what in the end amounted to basically nothing but "seed" questions linking to other sites. So I am definitely interested in making sure that doesn't happen here as well! – Josh Jan 22 '12 at 4:59
  • Robert, do you think you could join us in chat at some point in the future to review our progress and provide constructive feedback if you feel we're not on track still? I know at least Steven, Ben and myself have been very active in chat and have been discussing meta topics and working hard to help the site succeed! – Josh Jan 22 '12 at 5:01
  • @BenBrocka My answer isn't a concern about the low answer rate this early; it was a caution about loading up this site with "seeded" questions that people may not be interested in answering at all. See my first quote. – Robert Cartaino Jan 23 '12 at 14:59
  • @RobertCartaino point taken – Ben Brocka Jan 23 '12 at 15:03
4

I think these are great suggestions.

A quick look at our top users in terms of rep as of Jan 21st looks like this:

Answers   Questions
10        5  
7         10 
0         38
0         5
4         0
2         5
1         3

Obviously the flood (albeit well-meaning) of 38 questions by one user has potentially created an issue especially at this private beta phase in terms of getting a democratic sense of what the site is all about.

Personally, I'm open to others still asking questions in the private beta. In terms of defining the site, I think it's important to get input from others. If I were to suggest a general rule for the private beta period, it would be that a user should try to at least maintain equality between the number of questions they have asked and the number of answers they have provided. Of course, in the future when the site goes live, there always be a discrepancy between people who tend to ask questions and people who answer questions.

  • 1
    You make a very good point about 38 questions by one user being a bit off balance. Also, I like the general rule of thumb that you should try to answer at least as many questions as you ask, however for some users here (myself included) this may be close to impossible. I'll still try, however! – Josh Jan 21 '12 at 13:27
  • @Josh: Well, following that rule that would mean we should refrain from asking questions for the time being. ;p There are however enough unanswered questions to choose one from in which you are interested, and do some research on it if you have the time. – Steven Jeuris Jan 21 '12 at 19:28
2

I also think this is a good idea. We curently have 724 unanswered questions (15% of all questions) and according to Area 51

90% answered is a healthy beta, 80% answered needs some work. In the beta it's especially important that when new visitors ask questions they usually get a good answer.

so we need to get the answer percentage up to at least 90% if we are to get this site out of beta, and therefore we have a mammoth task of answering a minimum of 220 unanswered questions. This will go up as more questions are asked.

I am steadily going through questions I can answer. It is easy to call them up. Either click on the link at the top of this answer, or just type in the search box at the top

answers:0 closed:no

Another point to note, is not just that we have unanswered questions, but that there are a lot of questions with just 1 answer and our answer ratio is 1.4

Again, Area 51 states:

[A ratio of] 2.5 answers per question is good, only 1 answer per question needs some work. On a healthy site, questions receive multiple answers and the best answer is voted to the top.

  • This is always a good effort, but don't put too much emphasis on the beta metrics. If anything, the main thing to get out of the metrics is that they have stayed the same (and have even improved over time); this is a really good sign since they are not bad. Also don't forget you can vote to close questions which with our new introduced restrictions might be off-topic now. – Steven Jeuris Mar 3 '18 at 12:03
  • That is true, which will also help with our answers metrics – Chris Rogers Mar 3 '18 at 12:05
  • 1
    You are referring to Garbage Valley, a term coined by ChristianHummeluhr and now a recognized problem at our site. Contrary to @StevenJeuris I do find this problem to be significant and indeed it has been addressed elsewhere, e.g. here - I have seen your effort and I have undertaken two even bigger efforts in the past, listing hundreds of questions into the review queue.... – AliceD Mar 5 '18 at 7:54
  • 1
    I totally support this effort to reduce Garbage Valley. As a mod I cannot list questions into the review pane as any vote from my side is binding. As an active user, I would encourage you to go through these old, unanswered questions. However, some of the questions you targeted I have starred and I aim to answer them, as they appear to be good posts. Answering questions should always be the preferred action, as that improves the overall quality of our site, and that is the most important thing.... – AliceD Mar 5 '18 at 7:57
  • 1
    But again, eliminating Garbage Valley is a long term goal of me and I really appreciate your work. Plus, throwing questions in the review queue puts them to other users' attention and that's always a great thing. Kudos for your work Chris. – AliceD Mar 5 '18 at 7:59
  • 1
    And also of importance - answers with less than two upvotes do not count for the %answered metric - so upvoting of good answers is important. – AliceD Mar 5 '18 at 13:32
  • 1
    Oh interesting, @AliceD, on the 2 upvotes needed on answers. – Chris Rogers Mar 5 '18 at 13:35

You must log in to answer this question.

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