Peter Kaminski wrote:
The best long-term solution would be to work with Google and other search engines to encourage them to efficiently recognize Wikimedia content and reduce rankings of sites mirroring it just for SEO.
I've heard that Google have looked at ways to downrank mirrors, but they don't think it's ethical to just penalise everything with a domain name that doesn't end in wikipedia.org. As far as I'm aware, the most obvious heuristic, "covered in spammy ads", was not considered. Presumably that could risk damaging their bottom line.
Tei wrote:
*cough*But... these sites make a service to wikipedia and the public in general, creating multiple mirrors of articles. If wikipedia is down, you can still read the articles elsewhere, if the articles are deleted, you can read these articles on these sites*cough*
The script in question is a proxy, not a mirror. Look, I can take it down for you right now just by tweaking our squid configuration...
Better?
-- Tim Starling