I'd like to post this on the per-site metas of nineteen beta sites in the Stack Exchange network:

The Beta Phase: a survey

Hey there, _____s and fellow Stack Exchangeians!

I'm Doorknob, a regular from the beta site Programming Puzzles & Code Golf, and elsewhere on the Stack Exchange network. If you are reading this, then you are probably interested in the growth and development of your own community. I would like your input on the always present issue of graduation.

Your site has been selected because it has been in its beta phase for over one thousand days. Wow. That's a long time. Seeing as you've had over a thousand days of experience with this, you're probably familiar with the beta phase and the process of graduation. You've also probably wished to get out of beta at one point or another. Well, I've decided that by combining the knowledge of the eighteen other sites that have sat in beta for over 1000 days, we'll be able to make some collective progress here.

If you would like to help, I request of you only that you answer the following questions to the best of your ability. I'd really appreciate it if you do this, as it will help greatly in determining what we all have to do in order to graduate.

  • What is the status of the following issues on your site? Please rank them from 1 to 10, 1 being "not an issue" and 10 being "major, substantial problem that's holding us back in beta." If you would like, it would also be great if you summarized the issue or added some extra explanatory text.
    • Questions
      • Question intake (per day)
      • Overall quality
    • Answers
      • Answer intake (per question)
      • Unanswered questions
      • Overall quality
    • Reputation, users, and activity
      • Reputation levels (2k users, 10k users, etc.)
      • New user intake and retention
      • Average views and activity on new questions, and visits per day
    • Moderation
      • Questions not being closed/deleted quickly or at all
      • Spam, low quality posts, off-topic questions, and non-answers
  • How much would graduation help your site? Please answer with a number from -5 to 5, with -5 meaning "extremely harmful," 0 being "neutral / no effect," and 5 meaning "extremely helpful."
  • When do you think your site will graduate?
    1. In one or two months
    2. Four months or less
    3. Eight months or less
    4. One to two years
    5. Longer / never
  • When do you want your site to graduate? (Please answer with numbers above)

After analyzing the results of this survey, I will crunch the numbers and create a graph or two. I will then present this data to you, which you can do whatever you want with, and the community team over at Stack Exchange, to see what they think of it.

My goal with this whole endeavor is that hopefully, we will be able to pinpoint some good and bad things happening on all of these sites, and see what could be done about them. It's also possible that some correlations between all sites might be found in the data, which could be discussed in regards to all sites, and perhaps that a solution be found. We're all in the same boat here, so we might as well help each other out.

Thanks for reading this, and please answer these questions in the answers below!

Now, obviously, I can't just go spamming a thing on 19 sites, so I thought I would ask here first.

  • Can I get "official" sanction to post this on the per-site metas of 19 SE sites?
  • Should I change any major issues with this post?
  • Are there any other things I should do before posting this?
  • 1
    You realize you can get most of this information from SEDE or the API...
    – Shog9
    Apr 19, 2014 at 23:58
  • 1
    @Shog9 And for the rest: "It's my site's graduation, and I want it now!"
    – Tim Stone
    Apr 20, 2014 at 0:05
  • 1
    I'm guessing you're trying to figure out when a site will graduate, and I certainly wish the graduation progress was clearer and better communicated. It's not completely about the numbers though. Some beta sites apparently haven't graduated due to a design backlog.
    – Troyen
    Apr 20, 2014 at 0:11
  • @Shog No you can't; this is purely opinion. Maybe some site thinks that 3 questions per day is okay while another demands 20. It's not a objective fact.
    – Doorknob
    Apr 20, 2014 at 0:21
  • I'm pretty sure the number of questions, answers, users, amount of reputation, time to close, etc. are all fairly objective measurements. Is this like one of those lie detector things where you start off asking for folks' names and ages to calibrate the rest?
    – Shog9
    Apr 20, 2014 at 0:24
  • @Shog There are hard numbers, but they don't mean anything out of context. If they did, then graduation would be a purely statistical process, would in not? Some beta sites have far better stats than graduated ones.
    – Doorknob
    Apr 20, 2014 at 0:26
  • 2
    My point is that you're asking folks to take time out of their day and away from their interests and spend it filling out a form with numbers you could just as well dig up yourself. Do you not realize this? Or do you have a reason for it that I'm not seeing?
    – Shog9
    Apr 20, 2014 at 0:28
  • @Shog I'm asking them to say how much of a problem each one is, which isn't something I can do without becoming familiar with each site's individual culture.
    – Doorknob
    Apr 20, 2014 at 0:30
  • 3
    I'm gonna just suggest that you pick a site - go with Code Golf - and fill out this survey. Then try to explain why you chose the answers you did. It's really, really easy to make a survey that asks unanswerable questions, forcing folks to made decisions based on incorrect or irrelevant data - if you're not seeing the problem here yet, I suspect you haven't tried actually doing this yourself...
    – Shog9
    Apr 20, 2014 at 0:34
  • @Shog Alright, I'm starting to understand your point here. Thanks; I won't post these then.
    – Doorknob
    Apr 20, 2014 at 0:38

1 Answer 1


I don't think it would be appropriate for you to start such a discussion on a site where you don't participate.

On the other hand, MSE seems like the perfect place for it.

It would be appropriate for you to post a link to this discussion in these sites' chat room, as long as you're prepared to answer reasonable follow-ups in chat.

Of course, if someone wants to post your survey on a site where they do participate, that would be fine.

I don't think your survey is particularly useful. It isn't a site's population that decides when a site graduates from beta, but Stack Exchange. There are no firm criteria for graduation. I've participated in several betas through graduation and have no idea when a site becomes ready. The only important factor that's been made public several times is that a site must have enough users at various privilege levels to handle moderations (and this isn't about the data shown on Area 51, which hasn't changed since 2010 even though the decision to graduate has benefited from experience since then).

This is all the more complicated as external factors come into play — for a while, many sites were considered ready for graduation by the Stack Exchange staff but didn't graduate for many months because no one was available to work on their design. Now that Stack Exchange has a second designer on staff, this seems to be getting better.

Overall, I'd say graduation is not that important. A distinctive look perhaps helps market the site — or perhaps not, I've seen users who'd only participated on a beta site remark that they liked its uncluttered look. Ok, there's community ads, I guess that's nice. And moderator elections, but I have nothing against appointed moderators (of course I am biased).

Regarding the content of your survey, I don't see how the “when” questions can be answered, since we don't know the criteria. In the moderation section, you missed an important issue: meta participation.

I'm going to answer for two sites, though with prose rather than numeric ratings, because I have no idea what scale you expect.


I'm a regular but not top user of Cryptography. This site covers a very specialized, highly technical topic. There are many experts there, as well as students, users and the occasional amateur.

Questions: 8/day, generally good. There's quite a bit of off-topic traffic, from people using cryptography as a black box rather than asking about cryptography itself; they are redirected to the appropriate site (mainly Super User for users of software that involves cryptography, Stack Overflow for programmers with a bug in their code, Security for questions relating to the choice of cryptographic algorithms in a security architecture, Mathematics for questions which are purely about the underlying math).

Answers: many topics are too technical for me to judge, but on the whole I think that questions get good answers. Several participants have academic publications in the topic. The upvoted-answer rate is 92%, which I think is reasonable considering that some questions are very difficult. In the self-evaluations, more advanced questions generally do well (meaning that people think they have good answers); questions from students, programmers or amateurs tend to fare poorly.

Users: 2 trusted (20k), 7 semi-trusted (10k), 17 closers (3k), 26 editors (2k); page 2 (72 users) stops at 630. I don't know how to evaluate user retention.

Moderation: These days the review queues don't clog up. There's little meta participation, but big issues don't come up often; the site's scope is fairly clear (to the regulars at least), and conflict is rare.

Would graduation help? Probably in terms of marketing; what's the point of graduation anyway?

When? I don't know. I assume the site needs more users with enough reputation to close, edit and review.

French Language

I'm a moderator on French. The site doesn't make your cut but it will in a month. The site covers a well-defined topic. There is a small but competent public of native speakers, but a dearth of genuine experts (especially teachers of French, whether as a native or second language).

Questions: ~1.5/day. There are two main types of questions: questions from learners, and from natives. The learners tend to come and go; the natives are mostly regulars who also speak English and drifted from other Stack Exchange sites. We have no problem with bad questions (some, sure, but definitely manageable) but the low number of questions is definitely a concern.

Answers: There are only 9 questions without a positive-scoring answer. This is not surprising for a site where any native can have something to say about most questions; all language sites have a similar figure. Self-evaluations tend to be positive but not overwhelmingly so.

Users: 1 trusted (20k), 6 semi-trusted (10k), 20 closers (3k), 30 editors (2k); page 2 (72 users) stops at 687. There is a core of regulars (mostly natives, also a few learners with the occasional question) who have stayed for a long time.

Moderation: The review queues empty quickly. There's never a shortage of editors. Meta is slow, but nothing important comes up these days.

Would graduation help? For a language site, we're missing a vital feature: a user interface in our native language. With only an English interface, we can't market the site to many experts (in particular French teachers of French), who for the most part cannot speak English or are not interested in making the effort. A French user interface is a prerequisite to graduation; while it could happen at the same time, I doubt that we'll have enough traffic to justify graduation while cruising on an English-only UI.

When? First, we need a French interface. Then we can promote the site more, and I expect traffic will pick up. (It's already picking up, albeit very slowly, but it's attracting more non-native askers. At the moment we have more than enough answerers for our traffic level but at some point we'll need more answerers. And our challenge will be to have a good enough site to attract not just random Pierres but professionals.)

You must log in to answer this question.

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