What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 133 Stack Exchange communities.

There are a number of sites which just don't have any activity to speak of. I think there needs to be a more consistent policy about what we should do to encourage them to grow, and kill them if they are dead. What can we do to help encourage those sites to flourish?

share|improve this question
2  
Given that you seem to be referring to Beta sites, don't worry about "killing" them. If they show no chance of getting out of beta, SE does not hesitate to discontinue them. This has happened more than once before. –  Bart Feb 28 '13 at 19:14
    
@Bart: There's a number of sites which basically appear dead now. I could name 3 easily, and another 4 that could use some real work... –  PearsonArtPhoto Feb 28 '13 at 19:18
7  
Then go participate. Or motivate others to participate who you think might benefit from the site. Other than that, there's not really anything that needs to be done. –  Bart Feb 28 '13 at 19:21
3  
@Mysticial I don't really think listing them is particularly constructive –  Ben Brocka Feb 28 '13 at 20:05
    
@BenBrocka Yeah, I didn't see it coming. It's easier to say now since some people are undoubtedly upset in the answers/comments. –  Mysticial Feb 28 '13 at 20:15

5 Answers 5

up vote 11 down vote accepted

I'm a moderator on a beta site that has languished in beta for almost two years.

We've tried quite a few different things to boost our participation levels (at this point, I think our stats have been consistently good... with the exception of questions/day, which has hovered at around 10% of the "healthy" target for months).

At this point, I feel I can confidently say that there is nothing the larger SE community can do to help "boost" a beta site that will be effective in the long term, aside from giving the site's community time enough to explore what works and what doesn't work.

Which is exactly what is happening where you see sites that appear to be "dead".

The success of a beta site has to come from within. It has to have not only a solid core of users, but a clear understanding of its own identity, through scope definitions, policy, tagging practices, and other meta discussions.

No amount of outside help can fix it if the community hasn't been able to accomplish that on its own.

The only realistic thing StackExchange can be doing for these beta sites that I can think of is to get the word out to the public that they exist, and they do exactly that.

Jon Ericson's answer points out a spike of activity in November for Library and Information Sciences. I don't participate there, but I am willing to bet I know what happened: a good question likely got promoted by SE on twitter/G+, etc.. The reason I expect that to be true is because we've seen similar surges of activity on parenting.se when a good question gets tweeted (such as my recent question on letting toddlers win, if you'll forgive me for the self-promotion).

Which comes back to the main point: the community needs to provide a continuing flow of quality content in order to have the SE team succeed with promotion.

So the best thing you can do to help these sites flourish is ask good questions, and then stick around even after you've gotten your good answers.

share|improve this answer
1  
I concur with your guess. It's important to note that the way a site gets positive attention is by having great answers to interesting questions. You example is excellent since the question is one many of us with children have grappled with and the answer combines logic and expert opinion. (The only problem with the question is that it doesn't show up in Google searches unless you use really specific terms. Lots of sites already cover that issue.) –  Jon Ericson Mar 1 '13 at 0:56

Help them find more participants.

There is a critical mass required for a site to grow, and any help they can get to grow the quantity of users will correlate to the probability of them making it long term.

share|improve this answer
    
This works for a specific site that you want to help save, but short of actively monitoring the statistics, a user has to care enough about a site to do this, essentially by themselves, with no indicator of if a decision is being made to close said site. –  PearsonArtPhoto Feb 28 '13 at 19:37
1  
@PearsonArtPhoto If the individual users don't care enough to do this, then what chance does the site have to begin with? –  Bart Feb 28 '13 at 19:42

This is an absolutely terrible idea. The last thing that any community needs is some self-righteous "Stack Exchange citizen" to complain that they suck without ever bothering to participate. Before I get to making my point, let me illustrate why I get pissed off by people like you — you have 0 votes cast and only 1 answer that was migrated from Signal Processing on Cognitive Science and more importantly, not seen since March 6 2012! and yet you claim to be knowledgable enough to say that the "site needs probation"?

If a site has a problem (and lot of sites do) —

  • raise it on their meta.
  • ask them how they think they're doing.
  • if you have constructive suggestions, lay it out so that they can evaluate.
  • try and participate in any way you can to improve it, be it editing/asking questions/answering, etc. (this is the least you could do before you begin commenting on how much they suck)

In other words, involve the folks participating in the actual community instead of a bunch of Stack Overflow meta heads.

I don't object to the fact that some sites just don't have enough momentum and need to be culled at some point. Sometimes an outside perspective might help, but I see nothing of merit in your post other than smugly saying "I think these sites suck and so they should be killed".

share|improve this answer
3  
...and breathe out.... ;) –  Bart Feb 28 '13 at 19:36
1  
I'm resisting the urge to add a hyphen between "self" and "righteous". ;-) Another point to make is that dead and dying sites might still be providing a useful service to the internet at large. This is especially true if the subject is niche and slow-moving. –  Jon Ericson Feb 28 '13 at 19:38
    
Basically, I don't want to be the one to go about trying to help save sites. But someone needs to do it, wither they be an SE employee, or a member of said site. I don't mind, however, helping to save a site if it's one that I'm going to, or actually have, used. And I would really appreciate knowing that my site is on the chopping block before it gets chopped. –  PearsonArtPhoto Feb 28 '13 at 19:42
1  
@PearsonArtPhoto "Basically, I don't want to be the one to go about trying to help save sites" — Then don't. But also don't be the one to go about trying to kill sites. –  Lorem Ipsum Feb 28 '13 at 19:46
    
@LoremIpsum: Sites are going to be killed whether or not someone say something about it. I just want to see them be given notice about it beforehand, and clear goals as to what they can do to keep it from happening. –  PearsonArtPhoto Feb 28 '13 at 19:47
2  
What part of the beta phase and its metrics doesn't already achieve that? It's not exactly rocket-science to figure out if a site is doing well. And let's be honest here, SE seems to give the sites a fair shake at getting out of beta. Closing sites does not seem to be done without any consideration. –  Bart Feb 28 '13 at 19:50
    
@Bart: Of course there is consideration. But historically, when beta sites that have been around for a while are closed, there has been an announcement made that the site will close in a few weeks. Even if the site improved, it would still be closed. The only thing that could be done was to migrate the questions to another site prior to closing the doors... –  PearsonArtPhoto Feb 28 '13 at 19:52
1  
@Beofett: My answer and downvote occurred when the sites were listed in the question. Now that the question has been softened, I think it deserves a second look. (Though, it probably would take a book to properly answer it. ;-) –  Jon Ericson Feb 28 '13 at 20:24
1  
From one meta head to another, I love you –  jmfsg Feb 28 '13 at 20:39
1  
FWIW, @Pearson: we have been trying to improve our communication here a bit - the "self evaluations" are a big part of this effort, but in general we will post concerns on sites we're worried about. If you can't find anything worrying from someone on the team, there's probably nothing to worry about... Or it's so obviously bad that nothing more needs to be said. Sadly, one of your (now-redacted) examples has kinda fallen into that last category. –  Shog9 Mar 3 '13 at 4:12
    
Even for the obvious, it's still helpful... –  PearsonArtPhoto Mar 3 '13 at 21:11
    
We'll always post something before we shut a site down, but really... If you're well under 1 question per day, if most of your area51 stats are red, if you know who's upvoting your posts because he's the only guy besides you still voting... Then you don't need us to tell you there's a problem. –  Shog9 Mar 4 '13 at 1:46

Allow me to introduce you to Libraries & Information Science:

Libraries' heart beat.

I've never visited the site myself and have no particular need to ask or answer questions there. (I do love libraries, but that's not really the point.) The above chart (visits per day for the last half year) comes from Quantcast, which has similar information about every site on the network. I'd like to draw your attention to a few details:

  • The site has a clear "heartbeat". Weekends and Christmas are clearly visible as people log in on Monday morning and start checking out (so to speak) on the weekends. This is a site with regular users.

  • Something big happened on November 27, 2012. I don't know what, but it really seems to have sparked new interest. Perhaps someone posted a link to a question on some site where librarians like to go?

  • There's a noticeable increase in activity over the last couple of months. We can see real growth in the site, though it's not explosive.

The Area 51 stats don't tell the full story. Is the site being used and curated or is it just a wasteland to frustrate future Googlers? Does the site have potential for growth or does it serve a dying subject? Are questions getting quality answers or is it mostly spam? These questions might be important to make a decision, but they can't be answered from the simplistic Area 51 stats. Looking at them can be a bit like calling for a pitcher to be fired because his batting average is so poor. (International version: Looking at them can be a bit like calling for a goaltender to be fired because he never scores any goals.)

Find a niche

It can be helpful to think of each site as a mini-startup. The first thing you need to do is make sure that you've got some market you can serve. Area 51 and the Community Managers do a pretty good job of ensuring that a site at least starts out with a vision of what sort of users it will attract. If you have regular users who care enough about the site to visit it week after week, you probably have found yourself a niche. For the chart above, I would guess that L&IS has found its place in the world.

Pick a growth strategy

The next step is to figure out how the site might grow. In business, it's possible your niche is too small (the old refrigerator salesman in Alaska gag). But on Stack Exchange, it's more likely that the segment of the market you are pursuing doesn't know about the site yet. This is doubly true if the site's subject is not particularly related to programming. Spikes in visits, such as what happened in the end of November on L&IS, signal that there is an untapped market waiting to be introduced to your site. If you turn up in search results or get mentioned by movers and shakers or get in the news, you might find visitors coming out of the woodwork.

Retention

Probably the most important step is to convince new users to stick around. Sites grow when the number of incoming users exceeds the number of outgoing (or rather not-returning) users. Maybe you saw a bunch of people turn up for the hat contest and then disappear. That would be a sign that you are doing a poor job of retention. You won't capture everyone who shows up on your doorstep, but a healthy site will find new users who quickly turn into site enthusiasts. Perhaps the best way to achieve this is by filling your niche well. (See above.)

share|improve this answer

I'm going to try this again, a bit more carefully worded.

  1. Any site that is in danger of being closed (By whatever metric is decided that it is in danger) should be given a warning of some kind, at least 30 days prior to closing. That warning should be posted in meta, as well as emailed out to a list of users above some reputation level.
  2. The letter should express concern about the site, and about what the users can do to help the site flourish.
  3. There should be some indicator as to what kinds of metrics will help ensure the site will remain open. These metrics should apply for some kind of check points, long down the path way. These check points might be as far away as 6 months down the line.

The specific details can be worked out later, but in general, I believe this will help a site's users to realize that a site they value is in trouble, and they need to help more if they wish to keep it as a resource.

share|improve this answer
3  
I don't much like the idea of point #1. It can be agonizing to participate on a site that isn't getting much traction and the temptation is to work really hard asking questions to keep the lights on. Adding a 30 day warning will mean that some poor users have a hard deadline to produce a flurry of activity. Then what? Unless the underlying problem(s) are fixed, the site will either be closed anyway or continue on life-support. Just pulling the plug is far more merciful in these situations. –  Jon Ericson Mar 5 '13 at 18:53
1  
@JonEricson There's nothing to say that the criteria can't be extended for a period of time further than a month away. In fact, I think there should be some kind of criteria for a long term plan. You are absolutely correct, a few dedicated users might keep a site going for a month, but I don't think they could for 6 months. And of course, human discretion would need to be done in any case (A big part of the reason I didn't include specific metrics) –  PearsonArtPhoto Mar 5 '13 at 18:57
    
I'm still confused by why you think a probationary period would help. Your second and third points seem helpful to me, but there doesn't need to be any (published) deadline. I can only see how it might cause pain and confusion; I can't see how it would actually fix any problems. –  Jon Ericson Mar 5 '13 at 19:07

You must log in to answer this question.

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