And we already have a functioning antispam system, and antispam has been mostly separated from ToS, such that most of the spam volume can be handled without bothering ToS. Which I think would count as a specific policy reason why things are lower -- I'm in antispam, and we have a reasonable volume, but it's still small enough that the first round of recruiting plus one more later-added person is reasonably on top of it. (Spam volume has increased since the beginning, but there have been improvements to the system so that it is no longer *quite* the same "Brad, in his dorm room, with BML" system that we inherited. (It's due a revamp, but we can sail on for a while as-is. My counterpart and I are going over antispam applications, but there's low enough volume that it's not been our first priority.) The vast majority of spam is anonymous comment spam, mostly from a few rotating specific recognizable campaigns or methodologies, with a handful of one-offs, some non-spam (mostly obnoxiousness in a few specific anonymous comment communities), and some OpenID spammers.
There have been a very few registered users reported: a mix of legitimate content reported accidentally, legitimate content reported maliciously, legitimate content reported after an apparent loss of friendship, legitimate but mean/dodgy content reported indignantly, legitimate actual users making poor outreach choices, and a handful of gen-u-ine registered-user spammers.
The real spammers are reported to ToS for termination. The poor outreach choices are also referred to ToS in case there are associated complaints, or so that someone can tell them to make better choices in the future. Isolated meanness and other forms for legitimate content are closed without action. There has so far been one case (albeit multipart) of Assorted Dodginess, which was referred to ToS. ToS has courtesy access to the antispam reports in case they need it for any of the things reported to them.
I have previously described LJ's spam problem as "enterprise-level"; I think that a single person could probably handle DW's spam by themselves at this point, if not for the burnout.
no subject
There have been a very few registered users reported: a mix of legitimate content reported accidentally, legitimate content reported maliciously, legitimate content reported after an apparent loss of friendship, legitimate but mean/dodgy content reported indignantly, legitimate actual users making poor outreach choices, and a handful of gen-u-ine registered-user spammers.
The real spammers are reported to ToS for termination. The poor outreach choices are also referred to ToS in case there are associated complaints, or so that someone can tell them to make better choices in the future. Isolated meanness and other forms for legitimate content are closed without action. There has so far been one case (albeit multipart) of Assorted Dodginess, which was referred to ToS. ToS has courtesy access to the antispam reports in case they need it for any of the things reported to them.
I have previously described LJ's spam problem as "enterprise-level"; I think that a single person could probably handle DW's spam by themselves at this point, if not for the burnout.