If I had a simple site that required authentication to be able to post questions and submit answers, I’d:
- Perform a captcha and email address verification on sign-up
- Have a report spam/inappropriate link on each user-submission
- Perform back-off algorithms similar to resends of packets for asking for captchas. e.g. Captcha every authenticaton, and then on the submission of the first 50 questions or answers. If none of their submissions were reported by other users as spam, start captcha-ing slightly randomly submissions for another 50, and then remove captchas permanently for that account unless they are reported for spam.
- If more than one of their submissions is reported as spam by multiple users, reset the account’s captcha counters and warn the user. If they are reported a second time by multiple users, lock/ban the account.
I know it’s complicated, but I think it’s a decent compromise. Yes/no?