Dan Tobias wrote:
The way I see it, there has been vastly more disruption to Wikipedia coming from attempts to suppress links to sites than has ever occurred by the presence of such links.
Three or four people said this in rapid succession, so let me just say: fifthed.
Will Beback wrote:
The aim of creating some kind of policy or guideline to cover the issue is to give editors a road map of how to handle this type of problem to minimize the disruptions that have resulted from off-site harassment. Simply saying it doesn't exist doesn't help.
No one is saying off-site harassment doesn't exist. But since it's off-site, there is by definition nothing we can do about it directly, at least not within the confines of our Wikipedia sites.
I may be very stupid, but I'm still not seeing why we need more policy here. We've already got WP:NPA, which says that that a link which serves as an attack (my means of pointed reference to an off-site attack) is tantamount to an on-wiki personal attack, and is prohibited. We've already got WP:RS, which works hard to define what a reliable source is, and which says that blogs generally aren't.
So what more do we need? Which elements of the policy formerly known as BADSITES do we need to preserve, and why? What is the cost (in false positives, unexpected consequences, or general inconvenience) of those elements? What is the cost (in terms of potential editor harassment, or other imagined travails) if we don't adopt those elements?
I'm aware of three such elements, are there others?
1. WP:NPA only talks about links which serve as attacks. But we need to ban *all* links to attack sites, even when the links aren't meant to attack, even when they're to pages on the attack sites which aren't attacks.
2. WP:RS only talks about links in article space. But we need to ban links to attack sites from anywhere, including talk and project pages.
3. Removal of any links covered by the policy formerly known as BADSITES should not be subject to the three-revert rule.
(Me, I disagree pretty vehemently with at least the first two of these elements, but the arguments against them have been posted ad nauseam, so I'll not rehearse them here.)