udosuk wrote:The most tricky for a program/robot to perform this job is to determine whether a post is a spam. Sometimes veteran members might like to post some off-topic stuffs with links provided but that wasn't anything commercially related. At the end of the day I still think a logical human mind works best for this job.
I was thinking of an automaton that let the decision making to
a group of trusted users
the automaton could monitor its mailbox or one particular thread;
trusted users could make pm's/posts that would say something like
"post
nnn looks like spam" (a vote)
once a particular post reaches
some-threshold votes the automaton would take action
the action could mimic the human guidelines, and be stepped,
like pm warning(s) first, then actual post deletion as last resort
post delettion would require admin-like privileges -- don;t know if the current owners would go for that
if not then the last resort could be a pm to realsudoku
is that a bit bucket these days?