Somewhere around six months ago, I made the relatively obvious point that spoiler warnings were unencyclopedic and silly, and furthermore were being used to screw up article leads and violate NPOV. This led to a series of events that, over the course of about two weeks, had spoiler tags nearly completely deprecated from Wikipedia and the spoiler policy heavily rewritten to no longer encourage their use.
Six months later, the spoiler debate is still carrying on with the same half-dozen or so people vehemently opposing their removal. These arguments have been presented in every forum imagineable - arbcom twice, an RfC, several deletion debates, the mailing list, etc. The number of remaining forums is growing so slim that people were, in all seriousness, suggesting advertising the discussion on the watchlist sitenotice alongside the arbcom elections. This is, obviously, beyond the pale. Hopefully, the debate is now in its final throws as JzG has deleted the spoiler template following a TfD. Obviously it's on DRV at http://en.wikipedia.org/wiki/Wikipedia:Deletion_review/Log/2007_November_14 (with a breathtakingly bad-faith assuming nomination), but God willing it will stay deleted and this discussion will finally end.
What interests me, though, is the question of how we can prevent this. I've been fighting with the same people over issues with reliable sourcing for well over a year, for instance, and yet those fights still continue despite, seemingly, a substantial shift in opinion away from the former hardline positions (things that included overbroad statements about blogs "never" being reliable sources). [[2004 United States presidential election controversy and irregularities]] has been in need of a dynamite enema since, well, 2004, and has been the subject of an arbcom case, but so far nobody has quite managed to kill the blasted thing and its legion of OR sub-articles.
What is surprising in all of these cases is that it has seemed, to me, at least, that consensus formed for a position quite quickly - spoiler tags were stupid, sourcing guidelines needed to have enough flexibility to not break articles, and the 2004 election controversy articles are abominations. Everybody sane who looked at the situations recognized that. But unfortunately, everybody sane also demonstrated a general lack of willingness to participate in the same debates for months on end. And so the actual discussions have been deadlocks as a handful of tenacious proponents of the losing side continue stamping their feet.
This is a major tarpit, and is one of the ways in which dreadfully stupid things are allowed to profligate. It makes policy formation and the engagement of remotely tricky and nuanced situations a horrid timesink that is unsuitable for sane conduct. So what can we do? How can we streamline our policy formation problems to drive away the policy equivalents of lunatic POV pushers? Again, noting that the usual problems - consensus can change, so forcibly closing debates doesn't work, often contributors who are totally insane on one point are wonderful on every other article they edit, etc. So what can we do?
-Phil