PostSecret Pulls iOS App Over Abusive Submissions | Hacker News Reader

PostSecret Pulls iOS App Over Abusive Submissions

macrumors.com

19 pointsd_r7 comments

"We even tried prescreening 30,000 secrets a day." and "99% of the secrets created were in the spirit of PostSecret."

Is there a point where even 30,000 quality secrets per day are going to be too overwhelming and better off screened somehow anyway?

I run sites with anonymous submissions and lose a lot of time screening some really offensive crap, but even with 30k decent submissions per year, I'd be keen to filter those pretty strongly. e.g., if there is no expectation from users that they'll see everything decent that they submit, just approve the first x and flag the rest for later or never. Will even the most committed reader have time to go through 30k a day? As far as I understand PostSecret, they aren't really localised and probably not categorised, right?

It has been years since I visited the website, but I remember FMyLife.com simply displaying the "approved" posts to anonymous users. If you register, though, you can rate posts but also view the incoming queue. It seemed to be successful because I never saw spam in the anonymous view and the spam posts in the new queue seemed to already have marks against it. To me it doesn't seem like PostSecret would be any different, but maybe I'm missing something (possibly because it's an iOS app we're discussing).

Also, is anyone aware of any websites crowdsourcing moderation through Amazon's mturk? It seems like that might work and it sounds cheaper than moderating 30k secrets a day in house (but it would probably be less accurate).

Interesting idea to use optionally registered accounts for moderation; although I guess it kind of digresses from the point of anonymous posts.

Interesting idea to use mturk, it would probably be cost effective but the turn-around time may not make it feasible.

On one of my sites, a higher level of moderator has access to edit/delete while basic moderators (any regulars, basically) can "sin bin" a post for review/attention. Seems to work reasonably well in catching the worst posts.

This seems to be a general problem with anonymous posting venues. The amount of spam is often inversely proportional to the effort required to distribute the spam.

I wonder if the app's user community can fight against it by using a simple karma/points rule like on HN or Reddit. Too many down-votes and the post doesn't show up.

Anonymity without moderation will tend toward the profane.

Especially when there's essentially no cost (time/money/etc) to post.

Reply on news.ycombinator.com