merriam-webster.com are doing some very funky detection based on Google referrals. Visit a definition page, then a second (e.g. linked from first). You'll be prompted to sign-up. Now to go to Google and search for the word defined on the second page. Google will present a result from this domain it thinks has content, but in the same session that page will be a sign-up challenge without content.

This is quite certainly cloaking content, and it's very simple for a user to get a different page than that served to Googlebot. Naughty, and probably stupid too.

Google+: View post on Google+

This post was first made on the Richard Hearne Google+ profile.