Jennifer-Quit
When you manually report something(Spam, profanity, personal info, personal attack, defamation, etc) on the the site, it goes into a queue for review. I review the reported blogs/comments/documents/status updates. If there is prohibited content then just the prohibited content is removed. A reason for edit/moderation is included and the post is approved. If there isn't anything that breaks the rules the post is left as is and is approved.
This is what most sites consider moderation. Many sites I have been on will just reject the post even if 99% of the post is completely OK. At that point you'll either be able to edit the post and re-post or it will disappear completely into the ether. I don't like this approach, as a whole post shouldn't be removed if there is only a small part that isn't appropriate. It's extremely heavy handed. Furthermore, it's not fair to the person posting if they have no idea why their post isn't allowed. If they don't find out they stop posting or get angry and it only gets worse. They should be aware of what they've done so they can avoid it in the future and be a contributing member.
On the other hand there is a set of applications that look together at all posts on the site with the exception of private messages and automatically puts them in my queue for review if something looks suspicious. Flagged posts may have one or more characteristics of possible spam. New members(newly joined or minimal to no participation) are more likely to create spam than an established member. The Jennifer, Elvan's, Giulias, etc. of the community aren't likely to create spam. While spam controls are intended to only target the non-established members, the configuration is very complex and needs to be calibrated to each community to fit the communities needs. As time passes the configuration may need to be adjusted to keep the spam out while still still letting all the great stuff happen. That all being said, a mix of actual spam and legitimate posts are flagged as spam. The spam is removed and you don't see it. All of the false positives are approved and are posted. Right now the ratio of actual spam to real posts is not what I would like to see. I'm confident that this will be fixed when we figure out a better configuration and any discovered bugs are fixed.
Lastly, there is a prohibited word list filter. Another application looks at a list of words and *'s out any words that exist on that list that appear on public posts and private messages. These posts aren't put into my queue for review. The prohibited words simply are *'d out. Most of the words on the list are used for personal attacks or discrimination. This list may morph over time to add/remove words. People should want to be on EX because they feel like they can belong. Many of you live by the fact that EX helped you quit. Helping people feel like they belong is the first step to them feeling like EX was an integral part of their quit. Ensuring language used for discrimination and personal attacks is minimized will reduce likelihood that such events will occur and hopefully will in-turn make EX a better place to be.
So really to summarize. There are legitimate posts that are accidentally getting flagged as spam. I'm simply approving them and though the notification says they're approved by "moderation" they're not really moderated because they're posted as they were created and aren't altered at all. Hopefully that brings some clarification. We're working behind the scenes to fix the spam moderation so it's less disruptive to the community. Improvements will help both you the members and me the Community Manager focus on helping the site being more effective for all. Thanks for your patience.
Mark
EX Community Manager
EX Community Admin Team
