SoylentNews is people
SoylentNews
An article posted by Cory Doctorow on Boing Boing http://boingboing.net/2014/09/15/downvoting-considered-harmful.html has interesting insight into moderation:
A study http://cs.stanford.edu/people/jure/pubs/disqus-icwsm14.pdf [PDF] published in a journal of the Association for the Advancement of Artificial Intelligence found that sites that have a "downvote" button to punish bad comments lock the downvoted users into spirals of ever-more-prolific, ever-lower-quality posting due to a perception of having been martyred by the downvoters.
Cory continues: What's more, positive attention for writing good posts acts as less of an incentive to write more good stuff than the incentive to write bad stuff that's produced by negative attention.
How Community Feedback Shapes User Behavior http://cs.stanford.edu/people/jure/pubs/disqus-icwsm14.pdf [Justin Cheng, Cristian Danescu-Niculescu-Mizil, Jure Leskovec]
Why Reddit sucks: some scientific evidence http://www.washingtonpost.com/blogs/monkey-cage/wp/2014/09/09/why-reddit-sucks-some-scientific-evidence/ [Henry Farrell/Washington Post]
So... do you downvote? if so, why? Does this article make you reconsider your down-modding?
[Editor's note: I offer for your consideration and commentary our very own SoylentNews Moderation FAQ.]
(Score: 3, Interesting) by Flyingmoose on Tuesday September 16 2014, @02:28PM
I read a number of forums on the internet that get many times the traffic of Soylent. The way they deal with spam is to have a Report button, which sends a message to the moderator. The moderator can then delete the offending post. I don't think Soylent has so much traffic that it would be a large burden on the moderators, especially since I've seen basically zero commercial spam since I've been here. Some of the other forums I'm on are in industries with lots of spammers, and I still only see a few per week. So maybe the moderation system doesn't actually have to deal with commercial spam, which would make downvoting even more unnecessary.
(Score: 4, Informative) by TK on Tuesday September 16 2014, @02:59PM
There has been commercial spam here in the past. I remember some tool AC generating posts about his linux repair website or something. Before that, when it was just the wiki, spam pages popped up almost hourly for a couple of days. They were meaningless computer-generated text ads for some sort of health drink or something. Very long winded, not even coherent, but they hit a lot of keywords.
Deleting posts seems to be something the staff wants to avoid. Censorship and all that. Those posts get downmodded to -1 pretty quickly in my perception, so I guess that part of the system works.
Lately, it's been pretty pleasant, aside from the occasional troll/flamebait AC posts that usually get posted on any article where race is even tangentially related, but like I said, those are typically -1 by the time I get to them.
The fleas have smaller fleas, upon their backs to bite them, and those fleas have lesser fleas, and so ad infinitum
(Score: 1) by Tanuki64 on Tuesday September 16 2014, @03:20PM
I am not too happy with such an approach. Having moderators, who can delete posts, far to often leads to moderators, who participate in a discussion with the delete button. Dangerous territory.
(Score: 3, Interesting) by TK on Tuesday September 16 2014, @03:31PM
What if there was a "hide as spam" setting that the mods could select. Then instead of "browse at -1" as the lowest threshold, an "include spam" option could be there.
Alternatively, spam posts could show up as "hidden as spam/troll, click to show comment".
The fleas have smaller fleas, upon their backs to bite them, and those fleas have lesser fleas, and so ad infinitum
(Score: 1) by Tanuki64 on Tuesday September 16 2014, @03:37PM
Sounds good to me.
(Score: 0) by Anonymous Coward on Wednesday September 17 2014, @11:13AM
(Score: 2) by TK on Wednesday September 17 2014, @01:22PM
Easy peasy.
The "report" button can be used by anyone. It logs your IP and username (if you're logged in) and sends a message to the powers that be.
These reports are looked at by a mod, who agrees or disagrees with the spam label, and officially marks it as such.
The comment is now hidden, with the message "This comment was marked as spam by $Moderator'sName, click to show comment".
The fleas have smaller fleas, upon their backs to bite them, and those fleas have lesser fleas, and so ad infinitum
(Score: 2) by Common Joe on Thursday September 18 2014, @10:30AM
I like this idea. I'd also like to be able to mark a whole selection of these things as spam. (I've seen them come in batches at a time and I'd rather not waste all 10 of my mod points to mark 10 of 30 messages as spam.) On the flip side, I don't know how you'd control an evil user who starts marking every comment they don't like as spam. That could be a PITA for you guys.
I also wouldn't mind having a troll button. That is more slippery slope, of course. I would hardly use it, but I suspect others might abuse it pretty good.
Merely thinking out loud.
(Score: 2) by TK on Thursday September 18 2014, @02:09PM
Every "report as spam" logs the username and/or IP address of the person who clicks it. If you notice someone is just marking everything as spam, file reports from that user/IP under ignore.
My idea requires reports to be verified by the staff, which is a downside, but it prevents the equivalent of mod-bombing someone's post.
A possible outcome is that people over-use it and reports just start being ignored (or back-logged so long that they may as well be), but then we're back to square one and no worse off than we were before.
The fleas have smaller fleas, upon their backs to bite them, and those fleas have lesser fleas, and so ad infinitum
(Score: 2) by Popeidol on Wednesday September 17 2014, @03:24PM
I was a professional moderator for years, spending a lot of my time responding to reported content. They quickly become used to report people and opinions you don't like and generally dick with the moderators ('hey everybody let's just report every post for 5 minutes and see what happens'). You don't have long to spend on messages, and you tend to miss background information and err on the side of caution.
With crowdsourced moderation, you get fairly quick responses - the first person who scrolls past can handle it while understanding the context. If you make a mistake, somebody will probably fix it while calling you an astroturfing asshat.
Commercial spam mostly comes with scale or site type. If you make a dating or chat site, you get spam incredibly quickly and in quantity. If you throw up a default webform on whatever platform you're using, millions of bots will start to hammer it with their default-form-data. Soylent is a moderate-sized tech community with high rates of skepticism running on a modified 5-year-old unpopular codebase. Spam that arrives here is the result of a personal effort on the part of the spammer. it's an intimate touch.
tl;dr: Report buttons are great for some sites, but not here. We already have a better system.