thanks for those comments and the information on ‘soft-security’. It sounds like a well considered approach. I will review that information more thoroughly when I have some more time.

With Thanks

Simon

“soft security” empowers communities of people who care to collectively do something about trolling. Trolls in places like facebook are taking advantage of the fact that “power” in Facebook lies largely with Facebook (to ban users, block messages, etc) Soft Security attempts to “devolve power” to the users of the service. Some form of “Soft security” tends to emerge as among smaller social networks like p2pfoundation. For larger “mega” networks, if the empowerment of soft security existed, it would likely stabilize with networks of interest that are numbered somewhere in the neighborhood of 12-150 people.

My conclusions with networks like facebook, myspace, livejournal, orkut, secondlife, etc: the griefers and trolls are inevitable once the network scales up to a certain size. The only sustainable solution is to employ soft security, to allow those who inhabit the corporate spaces to have direct access to collectively maintain the integrity of their spaces. Soft security relies partially on technology. In the case of open wiki communities, it relies on a community-wide ability to reverse/roll-back, change or delete communications. In wiki communities, ID verification is not as important, because anyone can edit anything

But, in online social networks, it could be possible to counteract anonymity through peer review, meaning that network holds data that any one individual or group may affect that attempts to diminish the power of anonymous trolling by showing a reputation, a flag, or a way to both associate a user with community wide undesired actions, but also a pathway that can show the troll how they might direct their energies in helpful, useful, community desired ways. This is not about reputation systems, so much as being about giving users who care some way to do something constructive about a disruption. “Soft security” will technically and in practice look and work differently in different online communities that are engaged in different activities.

I agree that it is a serious problem.

Thanks to Jon for his message. See http://weblogsky.com/ for his excellent blog. I suspect that we need both social and technological solutions, the latter are necessary in terms of automating some of the time-consuming effects of protective governance (such as excluding spammers). I think that next to the positive practices of inclusionary participation, any project also has boundaries, and that these boundaries need some form of defense. Is anyone aware of any literature that focuses on this type of defensive governance of online communities?

Michel

Its the effect of community norms I suspect. – I think the kind of problems I was referring to would tend to occur more in these mega-networks like Facebook. So, its a ‘scale’ thing I suspect. However, by the same token it is therefore a very big and real problem. Identity theft is a serious issue and even a small number of cases can do a lot of damage and have a big influence on public perceptions. – As regards Facebook again, one only has to read the story of what happened to Graham Cluley of Sophos, to appreciate the potential impact of this kind of behavior in a public forum.

http://www.bcs.org/server.php?show=ConWebDoc.20188

Simon

– one is the trolling you deplore, but in an open network, trolls chiefly harm themselves by self-exclusion from the community, at least the ones with healthy norms and a culture of caring

– the other is abuse of open information by employers and other not used to open networks, but it may very well be that this kind of attitude, may actually disappear as the new generations take over …

A third factor is simple literacy, knowing what to post where, which comes through experience.

And as for closed networks, they exist, I’ve been invited to a few, and all of them have been less useful than the open ones in my experience.

At the p2pfoundation, we have a mixture of fully open networks, and rather frequently, invitation only email discussions where people add on the people they feel should be participating … The latter is not meant as an exclusionary policy, but also because not everybody is happy with getting all the discussions …

In the two years of practice, so far no trolling has occured at all, and civility has been maintained. Is that just luck, or the effect of our community norms?

Knock on wood …

Michel