Stories
Slash Boxes
Comments

Dev.SN ♥ developers

posted by NCommander on Wednesday April 02 2014, @08:35AM   Printer-friendly
from the understanding-the-community dept.
We've gotten some incredible feedback regards to the moderation system and the karma system, and trust me, its not going into /dev/null; I'll have a writeup done by the weekend. However, I've noticed something today that made me sit back, and think for awhile. Our community is healthy and vibrant, and we're far more cohesive as a group than we ever were on the other site. Furthermore, our users are significantly more active here than the other site. Almost all of us are from the other site, but there's a huge difference between us and them.

I can sum up the difference in four words: We ARE a community.

While many of us decried the other site calling us an audience, I'm not sure I can say I was a part of the Slashdot community. I read articles, and comments, but I hadn't moderated (or even logged in) on the other site for years. This wasn't always true; I'm UID 700139 on the other site (registered sometime in 2003), and I was fairly active until 2009. Then I stopped. I didn't even post on the Audience Responses post. I've talked to others on IRC, and it turns out I'm not alone; a LOT of people who are active here were permanent lurkers on the other site.

I need to understand why to keep us a community, and to prevent us from just becoming a passive audience. If you're going to post on any story, let it be this one, and tell me your story. We need to know.For this request to make sense, I need to make a distinction between not commenting, and lurking. Lurking is people who have user accounts, but don't sign in, never moderate and never post, even on topics that interest them. They are someone who is completely passive on the other site. Its fine that people comment on every single article; even at my most active on the other site, I posted at best one a month. A lot of people just like to read the comments, and perhaps moderate.

There is nothing wrong with that; those people are still part of the community even if they don't speak often. We've had two stories yesterday that broke 100 comments: Moderation: Discussing !(post^moderate) and OK Cupid Protests Against Mozilla CEO. Looking back at the history, nearly every single article we've run discussing the site broke the hundred comment mark. This is incredible because as of writing, we only have 4007 user accounts total, and slashcode reports seeing 54,620 unique IPIDs* for yesterday.

By chance, Slashdot ran the same article at roughly the same time as we did: OKCupid Warns Off Mozilla Firefox Users Over Gay Rights. This is what made me sit up and take notice. Slashdot does not post their stats publicly, but when DICE acquired Freenet, they posted some rough numbers in the official press release. From that article:

Slashdot, a user-generated news, analysis, peer question and professional insight community. Tech professionals moderate the site which averages more than 5,300 comments daily and 3.7 million unique visitors each month.

As I said before, we don't have a really good idea on the number of unique IPIDs visiting the site, but we do have solid numbers for our daily comment counts. Here's the graph as generated by slashcode for a biweekly period:

Biweekly Comment Count Graph

(due to a quirk in slashcode, the graphs don't update until 48 hours later; our comment count for 04/01 was 712 comments total).

Taking in account averages, we're roughly getting a little less than 10% of Slashdot's comment counts, with a considerably smaller user base. As I said, the OkCupid story made me take notice. Here's the comment counts at various scores between the two sites

         | SoylentNews | Slashdot.org |
---------------------------------------
Score -1 |         130 |         1017 |
Score  0 |         130 |         1005 |
Score  1 |         109 |          696 |
Score  2 |          74 |          586 |
Score  3 |          12 |           96 |
Score  4 |           4 |           64 |
Score  5 |           1 |           46 |
---------------------------------------
Furthermore, I took a look at UIDs on the other site, the vast majority of comments came from 6/7 digit UID posters. Looking at CmdrTaco's Retirement Post as well as posts detailing the history of the other site most of the low UIDs are still around, and are simply in perma-lurk mode.

Here's the rub. If Slashdot is really getting 3.7 million unique visitors per month, and there most popular articles only get to 1000-2000 comments (Taco's retirement, and the Audience Responses post both reached 2k), then Slashdot's readership is passive. Like, insanely passive. Let's assume that the average poster posts 5 comments a month (which is an extremely conservative estimate in my opinion). then out of those 3.7M unique visitors, only one person out of a thousand (1060 to be specific) is posting a comment. That's a horrendous ratio, especially for a site that allows anonymous postings.

I don't think this is inherent to the site itself; if we are getting 100-250k unique users (and I don't think its anywhere close to that high), then our numbers are still drastically better than Slashdot's. I suspect for every 100 users, one is posting, and if not, they're at least moderating or using the site. On average, we float 200-300 logged in users at a time, spiking up to 800-1000 in the evenings. On April 1st, we saw 3842 unique users logged in every day (out of 4007!).

I don't want this site to become a passive audience, I want people to be involved, and active in the site. This doesn't mean posting, but moderating, or at the very least, browsing while logged in. I suspect the vast majority of us were in the perma-lurk mode on the other site before coming here, and I want to know why. Tell me your stories so we can be a community, and not just a website with an audience. Let me hear them loud and clear, and tell me if I'm wrong; let me know if you were one of the most active posters on the other site, and if so, what sense of community did you feel over there.

* - due to the way we use varnish for ACs, the number of unqiue IPID per day is likely far higher it is in actuality. Due to our setup, the backend only sees one AC every five minutes + all logged in users.

 
This discussion has been archived. No new comments can be posted.
Display Options Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 4, Insightful) by kebes on Wednesday April 02 2014, @10:23AM

    by kebes (1505) on Wednesday April 02 2014, @10:23AM (#24805)
    I agree that opportunity is a big part of it. I think it's entirely natural that the "percentage of users who comment" will decrease as the user-base grows:
    - As there are more users and more comments, the opportunity to say something new/different decreases. So a given user is less likely to post.
    - As there are more users and commenters, the pace of commenting rapidly increases. Thus, to get in a comment (and have a chance of it being seen/moderated/appreciated) is more and more difficult (you have to jump on the story as soon as it is posted). If you don't get in 'early', your chances are low. This decreases the likelihood of commenting, because the user feels like it is pointless.
    - As the community grows in size, hostility tends to increase. This is both because for a given percentage of trolls, a bigger community will of course have a large absolute number of trolls. But it's even worse, because in a big community, people feel more anonymous and tend to be less polite. Worse still, trolls tend to gravitate towards large communities so that they can get more attention. The end result is that it decreases one's desire to comment. Even if troll comments are always knocked-down to -1, the community-at-large likely won't read them, but the commenter likely will read them. Even though we all know trolls are idiots, dealing with a bunch of hostile comments (or comments that totally miss the point you were trying to make) gets tiring and causes people to lose interest in commenting.

    It's also worth noting that SN, by the nature of its inception, has created a selection effect: the most motivated people, those most 'passionate' about commenting, were the first to jump over here. As time goes on the demographic will likely shift towards having more lurkers.

    I'm not sure that a decreasing percentage of commenters is in itself a bad thing. It seems somewhat natural. In fact, I think it's necessary to have some subset of people who are lurkers. There's nothing wrong with it (e.g. you may be a lurker on one site but active on another). In open-source software, some users may never contribute code or even bugfixes. But they still play a role (evangelizing, giving the contributors a sense of purpose, ...). It's the same for a website.

    On the other hand, I get the argument that we want the community strongly engaged, and commenting is one obvious measure of engagement (it takes much more effort to post than to just skip). So, I'm all for strategies that make commenting easier and more fun. But I would just caution against trying to maximize a particular metric (like "% commenting"): at the end of the day, the only thing that matters is that we're all enjoying the site.
    Starting Score:    1  point
    Moderation   +2  
       Insightful=2, Total=2
    Extra 'Insightful' Modifier   0  
    Karma-Bonus Modifier   +1  

    Total Score:   4  
  • (Score: 1) by CoolHand on Wednesday April 02 2014, @11:50AM

    by CoolHand (438) on Wednesday April 02 2014, @11:50AM (#24926)

    hmmmm...
    I wonder if some sort of "blind commenting" system could be implemented where for the first x amount of time comments aren't displayed -- no one can see them besides moderators. That would give everyone time to put out initial comments without being intimidated by the number of other posts that may cover the same ground.. Moderators could make sure the best well thought out comments covering the same subject matter rise to the top...

    I dunno... just brainstorming

    --
    Anyone who is capable of getting themselves made President should on no account be allowed to do the job-Douglas Adams
    • (Score: 2) by Common Joe on Wednesday April 02 2014, @03:51PM

      by Common Joe (33) <{common.joe.0101} {at} {gmail.com}> on Wednesday April 02 2014, @03:51PM (#25118) Journal

      Interesting. I don't think I agree with trying this, but since we're brainstorming, I have another crazy idea. Don't show the name of the commenter for a fixed amount of time, but allow moderation and replies. Perhaps make an exception for friends or enemies.

      I don't know if I like my own suggestion, but as you said, just brainstorming.

  • (Score: 2) by etherscythe on Wednesday April 02 2014, @02:57PM

    by etherscythe (937) on Wednesday April 02 2014, @02:57PM (#25085)

    I agree with pretty much all of this. I was a long-time lurker on the other site until I finally made an account, and over the course of maybe a year managed to get a few +2's and +3's and a SINGLE +5. Many times I would continue to lurk because someone had made a comment pretty much exactly making the point I would have, in which case it came down to whether I had mod points to give them or not.

    I would add that there's a large convenience factor involved. The other site has inline expansion of comments, which when I am modding and looking for interesting comments to mod up, is a very good thing. Here, I think we see only the more determined commenters sticking out the rougher experience, as well as already having more of the "frontier" personalities as parent post mentions, due to age and circumstances of this site's genesis.

    Right now there aren't so many comments that I feel the need to trim them down with a visibility threshold, but if we get inline expansion and the site userbase goes up, this will probably change. It will probably help encourage discussion when it's easier to keep track of the threads also, so that the same subtopic doesn't spring up in different places and fracture the discussion.