How do you build a community resource with few barriers while avoiding spam?

Digital Humanities Questions & Answers » Applications, Tools, Formats

How do you build a community resource with few barriers while avoiding spam?

(6 posts) (6 voices)

Asked 5 years ago by Stéfan Sinclair
Latest answer from Shane Landrum

Tags:

Stéfan Sinclair
Admin
As many of you will have noticed, our beloved DHAnswers site has fallen prey to a couple of spamming incidents of late. No one likes that, and we, your humble site administrators, apologize for the noise (that's been exacerbated by the twitter tie-in). We've re-enabled some anti-spam mechanisms that will hopefully avoid the most flagrant cases. Unfortunately, this may also cause some hiccups for legitimate uses.

This post actually is a question: How do you build a community resource with few barriers while avoiding spam? We had disabled Akismet and bb-NoSpamUser because they had caused a few real questions and answers to be suppressed (who knew that DH talk was so risqué? :). We would welcome any suggestions that people might have of strategies to keep barriers low, to keep communication fast (approval is less preferable when you're desperate for a quick answer), and also to keep spam out, especially as it relates to bb-forum. Does anyone have useful experience to share?
Tweet this question
Posted 5 years ago Permalink
Patrick Murray-John

I guess my response for this space, which can be generalized out, is to go with community expectations. As you said, some spam has gotten through, but one of the huge strengths of DHanswers is is the quick response time. I'm willing to accept the price being a few spams here and there. My guess is that most of us have found our own ways of dealing with spam. In short, sometimes the best spaces aren't the most pristine -- hospitals are very clean, but I wouldn't want to live in one.

That said, other online commons might require a more pristine approach -- sounds like more of a social question than a technological one in how they're built and managed.

Posted 5 years ago Permalink
Bethany Nowviskie
Admin

It's true that relatively few spam posts came through in the period in which we turned off filtering, but many spam accounts were registered and we could see the wave coming. On the other hand, early in the DH Answers' lifecycle, it felt to me like a bigger deal that (again, relatively a very few) legitimate posts by first-time users were occasionally getting hung up in our spam filter.

One thing that we might consider is helping the system degrade more gracefully for legitimate users. "Spam" posts were just, from the end user's point of view, evaporating. Would it help if we could give users a little notice when a post goes to moderation, and an avenue to alerting us if they think it was accidental?

It may also be that Akismet is improving enough over time that this won't be a major problem. But to Stéfan's broader question, I'd add: How would you DH Answers users prefer us to handle this?

Posted 5 years ago Permalink
Dorothea Salo
Member

In consideration for all the admins/mods do to keep this space functioning, I would suggest that ease-of-use on the back end is an important desideratum also.

To that end, is there a way that spam detection and removal can be crowdsourced? I would gladly volunteer some clicks.

Posted 5 years ago Permalink
Peter Organisciak
Member

One option if you don't want too many barriers to posting is crowdsourcing the removal of spam with a flagging mechanism. However, the Twitter account certainly makes *any* spam undesirable, even if it is quickly removed by us (though if Twitter has a delay from the RSS feed, it's still an option). Also, DHAnswers does get a high enough volume of posts to really benefit from this.

What about ReCaptcha for new sign-ups (or new OpenID users)? This usually works remarkably well, and legitimate users are only be bothered once. A slightly more aggressive approach would be to force ReCaptchas onto posts with too many links, but I think the basic sign-up captcha would cover most of the problems.

Posted 5 years ago Permalink
Shane Landrum
Member

Replying to @Bethany Nowviskie's post:

I like the idea of notifying users when a post goes to moderation, and I like the idea of automatically green-lighting the posts of anyone who's made more than (3? 5?) nonspammy comments (ever, or in the past X weeks). Like Dorothea, I'd be willing to volunteer a few clicks for spam detection and removal.

Posted 5 years ago Permalink

RSS feed for this topic

Reply

You must log in to post.