24

We have a rather disappointing situation on Programmers.SE. Two discussions that one would think would be enthusiastically supported, are now kind of stale.

The first one is on getting our own community blog, and the second one is for our birthday celebration event. The extremely weird stalemate is that one of our mods posted a question suggesting one event, and I posted an answer raising concerns about that event and proposing an alternative. Both are upvoted and no other answer appeared that would possibly break the tie.

I've spammed our chat room quite a few times, trying to get more people to participate on both discussions, with little effect.

I don't really know how to measure Meta participation and what would be considered healthy. The only number I think is relevant is that only two non-moderators have gotten the bronze discussion badge. And we can exclude the one guy, he's just noisy and opinionated and likes to make a fuss about unimportant things.

Given that the community has 40k members, I'd say our Meta participation is pathetic. I'm not really active anywhere else, so I can't really compare.

Anyways, my questions aren't really P.SE specific:

  1. How can I evaluate Meta participation? For all I know, my perception may be completely wrong.
  2. In any case, what can we do to get more people involved? What's your war stories?

I will raise my concerns on P.SE Meta at some point, but I would love to hear how the other communities handled similar issues.

Although as Jeff pointed out P.SE is doing pretty good when it comes to new Meta posts, I wasn't totally convinced, so I gathered the totals for most mature sites1. It appears that compared to other sites we (P.SE) aren't doing very well (sorted on Meta questions per users):

Site          Questions Meta   Users   M/Q       M/U       
--------------------------------------------------------
SciFi              2290   354   4000   0.15459   0.08850
Tex               14014   630   9200   0.04496   0.06848
Gaming            14346  1023  15000   0.07131   0.06820
Photography        4290   394   6600   0.09184   0.05970
Math              34657  1092  19000   0.03151   0.05747
Drupal             7788   240   4600   0.03082   0.05217
Physics            5874   311   6000   0.05295   0.05183
Cooking            4921   356   6900   0.07234   0.05159
English           12764   771  15000   0.06040   0.05140
GIS                5926   192   4400   0.03240   0.04364
CrossValidated     7304   290   6900   0.03970   0.04203
DIY                3456   155   4400   0.04485   0.03523
WebApps            6473   310  13000   0.04789   0.02385
UX                 3203   223   9400   0.06962   0.02372
Apple             12101   347  15000   0.02868   0.02313
Ask Ubuntu        33678   803  35000   0.02384   0.02294

Programmers       16005   857  40000   0.05355   0.02143

WebMaster          7801   204  10000   0.02615   0.02040
Game Dev           6078   201  11000   0.03307   0.01827
SuperUser        108857  1308  84000   0.01202   0.01557
ServerFault      101739   732  71000   0.00719   0.01031

I have no idea of the quality of Meta participation on the other sites, and I'm fully aware of how useless numbers can be on their own. But since I went ahead and wasted 20 minutes, thought I'd share.

1 I gave up when it became really boring.

|improve this question
  • 5
    I personally wouldn't care much about a blog or a birthday contest. Perhaps those two weren't the powderkegs you expected. (I check in on the SO blog when the little indicator goes red but I'm honestly a little disappointed most of the time that it isn't the recent changes page but with more details...) – sarnold Jan 21 '12 at 0:58
  • I think the "noisy and opinionated" guy will be really happy to read your opinion about him, if he comes by this thread. I understand your intention, but maybe try to use other exclusion criteria (or frame it more euphemistically) for excluding someone as "unimportant"! PS: In my frank opinion it is more important that the ACTUAL website is showing hightened participating, not meta. Ideally there is no need for a "meta" discussion (because there is no unclarity in any regard and people do agree). Of course this is only the ideal. – grunwald2.0 Jan 21 '12 at 1:00
  • And before answering your questions (I don't want and can't give an answer on it...), I want to ask you how you objectively measure the participation rate yourself - and which hands-on, specific negative consequences are there resulting out of a "pathetic" (low) meta participation? – grunwald2.0 Jan 21 '12 at 1:03
  • 12
    @grunwald2.0 Just to clarify, I'm the "noisy and opinionated" guy. If you've clicked the link to the badge, it would have been obvious. Sorry if that was misleading. And pathetic was meant in reference to quantity, not quality. – yannis Jan 21 '12 at 1:19
  • 3
    @grunwald2.0 As for the negative consequences of low Meta participation: A relatively small group of users decides on important stuff, like what's on topic and what's off. Isn't that enough? – yannis Jan 21 '12 at 1:55
  • @sarnold Well, yes but no. There isn't a single answer on the 6 questions on the two topics that says "meh". That's what's frustrating, everyone seems to agree (upvotes), but only a very few seem to actually participate. – yannis Jan 21 '12 at 2:05
  • 2
    People with opinions tend to stand out. People who are fine with things often don't bother taking the time to say "I'm ambivalent." :) – sarnold Jan 21 '12 at 2:08
  • 1
    Perhaps it's because people join programmers expecting one thing, and discover another, so although we have a lot of users, there aren't too many that are really dedicated to the site. I joined thinking it'd be a place to get opinions from other programmers on stuff I come across or am interested in, however it's more like the rest of the SE sites: a Q&A board only with limited discussion in chat. – Rachel Jan 23 '12 at 21:13
  • @Rachel Perhaps it is, as Jeff also pointed out in his answer. If that's the case, I'd expect the newer generation of users that joined at the same time as me to be a little bit more involved on Meta. – yannis Jan 24 '12 at 11:10
  • Is your graavitar supposed to represent career development questions on P.SE??? :-) – Adam Rackis Mar 8 '12 at 18:11
  • @AdamRackis No, the avatar is limited to my MSO account... – yannis Mar 8 '12 at 18:14
  • Well that only raises more questions. What is it about MSO that makes you want that gravitar??? – Adam Rackis Mar 8 '12 at 18:20
  • Your bit about the two non-mods is no longer accurate. I count seven current diamonds and one former diamond as of one minute ago. – Pops Mar 8 '12 at 18:44
  • @sarnold I'm ambivalent. – AndrewC Sep 28 '12 at 0:35
  • 1
    @doubleDown Cool. Didn't really mean to give you too much grief over this, my comments were perhaps phrased a bit more crudely than they should have been. Keep on editing ;) – yannis Jul 15 '13 at 8:15
10

I don't believe the two specific issues you mentioned—the Programmers birthday contest and the blog—are indicative of poor meta participation.

I'm active on about a half-dozen sites—ranging from sites older than Programmers to newer beta sites—and from what I can tell, Programmers definitely has one of the most active meta-discussion communities, and it's been getting more active recently as non-moderators such as yourself take more ownership of stuff like general support and moderation issues.

The birthday contest and the blog require three things:

  • Community buy-in: we mostly have this, although it took some time
  • A point person to coordinate the logistics of setting up the contest/blog: we don't have this
  • Stack Exchange, Inc. buy-in: more on that in a second

On Programmers, one of our big problems is that there are a lot of opinions about how things should go, and a lot of "I want this to happen", but not a lot of people effecting change themselves. Issues brought up by the community that could be self-handled (like a host of moderation issues) rarely are, and instead get handled by the two (three until recently) active moderators.

So, in the case of the contest and the blog, it felt like the expectation has been that moderators would do something and they'd magically appear. However, while I think it's fair to say that we think the ideas are great, we really have no desire for the additional workload and be the only ones organizing such things: we were really expecting members of the community to step up.

When we saw that wasn't happening, we asked SE on a number of occasions what exactly was required to get the blog and contest going, but we haven't gotten much: we were pointed to this blog post and told someone on CHAOS was made aware of the contest ideas, but that's been pretty much it.

Without having a clear idea of how to move forward, we've tried to push things forward, but have been met with resistance from the community, who seems perfectly content to continue discussing more ideas rather than moving forward with any one idea.

So we moderators—who aren't really interested in heading these things in the first place—are left with no useful advice to give the community about how to move forward, and thus these things just flounder for months.

So as I said, I don't think it's lack of participation: it's lack of guidance and help setting this stuff up. Some sites—like DIY, Parenting, SciFi, and Gaming—do get that help directly from SE, but I guess we're not lucky enough.

So moving forward, I think we need regular community members—who are really interested in seeing these things happen and know how to set them up—to step up and handle them. I wish I could give advice as to what that means: perhaps SE can shed some specific insights and/or instructions on our meta to let the community know how to move forward.

|improve this answer
  • 1
    Well, obviously, I don't have the slightest idea of what goes on behind the scenes. I attributed the less than enthusiastic support of the blog and the birthday event to poor meta participation, based on the available data. Anyways, I'll keep spamming chat for both the blog and the contest, and every future cleanup, and thanks for taking the time to write all this, I was wondering what the moderators think... – yannis Jan 23 '12 at 1:32
  • @YannisRizos to be honest, I think these things need less bikeshedding and calls for opinions and more doing: there appears to be a reasonable enough consensus to move forward. Unfortunately, we really don't have any insight as to what the "doing" entails at this point. – user149432 Jan 23 '12 at 1:36
  • I agree there's reasonable enough consensus. Was a bit frustrated with doing the [software] cleanup on my own (mods excluded) only to get a discussion after I finished everything. I was following the chat conversation from the start, and all I could think of was meh. I'm guessing we will get some feedback from SE soon enough and a couple of us will take point on the blog and / or the event. Well, I can only speak for myself, really... – yannis Jan 23 '12 at 1:53
  • 2
    Let's get started with a pool of 3 or 4 posts. Should not be a problem if you contact every single person that asked to participate. I'll contribute. – user150926 Jan 24 '12 at 12:52
10

The number of meta posts isn't everything. To take examples from sites I frequent, Unix & Linux is a well-oiled site, remarkably free of conflict; its meta has, if not cobwebs, then at least dust, mainly the occasional retag request. Science Fiction & Fantasy has had a very active meta lately, but most of the participation comes from a few core users. You shouldn't judge metas by their traffic but by whether they meet their goals. I think U&L meta meets its goals because it's there for the very rare issues that come up. Meta.Scifi less so: support requests are answered fine, but policy discussions don't always gather a healthy level of participation. A meta that's dominated by complaints against moderators is not a nice place.

I think there are two main ways of getting the right kind of participation on meta. The main way is to engage people in community discussions. Don't debate whether a close-worthy question is salvageable in comments: vote to close, post one comment explaining your point of view, and if anyone disagrees, move the discussion to meta. If policy issues arise, discuss them on meta, then apply any emerging consensus consistently, referring anyone who objects to the meta discussion (and don't do anything that's hard to reverse if you think the meta consensus may yet change). Basically, be proactive in moderating the site to force people to address the issues with proper argumentation in meta discussions.

A secondary way to engage interest in meta is to run community contests or topic of the time period. Travel often does a region of the day (mostly discussed in chat), Literature has a book club, Gaming and Scifi have promotional grants, Super User and Security have questions of the week…

|improve this answer
  • The number of meta posts is just an indicator, I'm not concentrating the evaluation of meta participation there, it's just the easiest metric I could find. More or less we try to direct every close dispute to Meta, and there's nothing but consistency when it comes to policy. Anyways I'm checking other Metas, to get a better idea of how other communities handle stuff, and will report back when (if) I have some more useful metrics. – yannis Jan 25 '12 at 9:26
6

It could be that programmers is still a bit young of a site to have these kinds of celebrations yet -- it's just over a year old...

http://blog.stackoverflow.com/2010/12/introducing-programmers-stackexchange-com/

... and as I recall we were still working through some fairly divisive teething issues on the site not less than 6 months ago. So perhaps the site is still in its childhood and simply needs more time to grow a healthy, strong community?

|improve this answer
  • That I understand. I'm more interested on how to get people a bit more involved though. I'm seeing more people involved (relatively to their size, of course) on Meta on Code Review and UX. Is there a process to evaluate Meta participation? Even a vague one... – yannis Jan 21 '12 at 1:17
  • 1
    sure go to stackexchange.com and click "more site stats" then sort by meta participation: stackexchange.com/sites?sort=newmetaposts&expand=true – Jeff Atwood Jan 21 '12 at 1:31
  • 2
    4th in users, 5th in new Meta posts. It seems pretty good, so it might as well be that my premise is false. Not entirely convinced though, but good enough. Tips & tricks to get even better? (I'm a little help vampire, I know) – yannis Jan 21 '12 at 1:35
  • Jeff I've updated the question with some fresh data that seem to support my premise. Care to comment? – yannis Jan 23 '12 at 2:22
  • @yannis when counting "users", you really have to drop any users with reputation below 5 as they can't post on the meta anyway... and users with 101 rep are also at the default of cross-network association. – Jeff Atwood Jan 23 '12 at 8:55
  • Oh I'm not going to do that, just scraped the stats page :) I'm speculating that they are about the same per site (relatively to site size), so they don't matter much when relatively comparing the sites. If I find some free time, I will try to get more relevant data via the API, but I think the current data are evidence enough that Programmers Meta participation isn't doing as well as the new Meta posts stat suggests. – yannis Jan 23 '12 at 9:01
  • 1
    @yannis well let's see. Scifi has 101 rep users starting on page 27/101. Programmers has 101 rep users starting on page 286/1064. So you might be right, those ratios are about the same. Eerily close actually! – Jeff Atwood Jan 23 '12 at 9:09
  • Well, that's motivation enough to do a deeper analysis, sometime next weekend... – yannis Jan 23 '12 at 9:26
3

I'd also like to see meta participation pick up. You're examples may be bad, but I think you're right about meta participation on P.SE.

Here are two suggestions I would make, coming from the standpoint of a regular user who took a long time to get understand what meta was for and to interested in it:

Help make more people aware of what meta is

I was not actually aware of meta for quite a while after I started using stack exchange sites. Actually, I was vaguely aware of it, but assumed it was all boring stuff about what was on/off topic, or things like that. Now that I realize what is discussed on meta, I'm quite interested in it :)

As for how we can make people more aware...

  • we could move the section about meta and what it's for closer to the top of the FAQ. I doubt anyone reads our entire FAQ because it's so long, and the meta link gets very lost at the bottom. In addition, I think the description of it is bad.

  • add some kind of description about what Meta actually is on its home page. As it stands now, a user's opinion on meta will be shaped by what questions happen to be on the front page, and if the user doesn't find any of those questions interesting, they're likely to assume the site is uninteresting and not come back

  • link it in comments more frequently. For example, if you are closing a question, provide a meta link with your comment on why the question was closed, and point them to either an existing meta question which explains why their questions was closed (or better yet, one that tells them how to revise their question to get it re-opened), or just tell them they can post a question on meta if they disagree with the closure.

Stop being so aggressive with downvotes

New users don't understand why they're getting downvotes. We've trained them to think of downvotes as negative, so when they ask/answer something and it gets downvoted, they think that they are doing something wrong.

The very first meta question I posted was asking about the possibility of implementing some kind of poll-of-the-day option to P.SE. It was downvoted, and even though it was explained to me that downvotes simply mean users disagree, it still made not want to participate in meta for a year or more afterwards because I felt I had somehow done something wrong.

A much more welcoming solution would be to post an answer/comment disagreeing with the question, and let it gather upvotes.

Basically what I'm saying is, use comments/answers to answer the question, not up/down votes. I'd prefer to use votes to see how much support an idea has, not have it show a skewed sum of people who upvote - people who downvote. I have no idea if a question has 100 upvotes and 100 downvotes, or simply no votes at all.

You may or may not agree with me on this, but I also feel that with smaller communities it is too easy to drive away people with different opinions. If someone posts something contrary to what the active meta group believes, the question will get a lot of downvotes and end up driving the user away. This makes the meta communities smaller and less diverse. This isn't really a problem in SO's meta, where the meta community is large and full of people with different opinions, but I do see it as a problem in smaller communities.

|improve this answer
  • 1
    +1 for your second point. We need to find a middle ground somehow. I have no idea how though. – Adam Rackis Mar 8 '12 at 18:13

You must log in to answer this question.

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