Gathering my thoughts on moderation at TSZ, I found that I really have two OPs to write: one discussing the effects of rules and moderation at TSZ, and another exploring why the moderation — particularly the Guano-related stuff — has those effects. The second topic is by far the more interesting, but it’s the first topic that has the most practical import, so I’ll address it now.
In a nutshell: We’ve already experimented with different levels of moderation at TSZ, and the results are in. Less moderation works better.
This is Lizzie’s blog, so of course any decisions regarding rules and moderation are hers alone to make.
None of us are absolutists about free speech or moderation, as far as I can tell. We all seem to agree that some moderation is necessary — to get rid of spammers or to prevent pseudonymous commenters from being outed, for example. The disputes are over the degree, not the fact, of moderation.
The optimal level of moderation may shift over time as a blog and its commentariat evolve. There is no single guaranteed-for-all-time optimal moderation scheme, and no comprehensive “moderation theory” for deciding ahead of time which scheme is most likely to succeed. Experimentation is essential.
With the preliminaries out of the way, the argument is quite simple, and I’ve already expressed it on another thread:
The ultimate test for any moderation scheme is how well it works in reality, with the actual commenters and the actual moderators. We’ve already run many such (informal) experiments at TSZ, and the results seem clear to me: Changes in the direction of more moderation have backfired, and changes in the direction of less moderation have made TSZ run more smoothly and with more on-topic, substantive discussion.
For all our disagreements, Alan and I both think that the experiment we tried in Lizzie’s absence was a success. If you recall, the policy during that time was to move comments to Guano only upon request (of the “injured” party originally, though that was broadened for some reason to include third parties).
Since that experiment worked so well, and since attempts at increased moderation (like the Wine Cellar, to use a recent example) have failed, why not take the lesson to heart and eliminate moderation altogether, except in the extreme cases where we all agree it’s necessary (spam/porn/outing etc.)?
It fits with the ethos here at TSZ, it minimizes the workload for the moderators, and it eliminates the reason for the numerous and lengthy moderation discussions we have here at TSZ, again and again.
Seems like a worthwhile experiment, doesn’t it?