...more recent posts
Wow. Very cool. Google introduces canonical URLs. This solves the problem of losing google juice due to identical (or largely identical) pages being accessible through different URLs. You decide which URL is canonnical, and then put a special <link> tag on the other pages pointing to the cannonical URL. Now google knows to return the cannonical page in search results rather than any of the similar ones, and all your google juice flows to that one page.
The big example would be http://www.example.com and http://example.com. If your site is reachable at both addresses, and both are linked to from external sites, then your rank in google until now has been split between the two (this is simplified a bit, but still.) Now you can decide that you want http://example.com to be the "main" (or canonical) URL, so you just put a <link rel="canonical" href="http://example.com/"> tag in the <head> section of the www.example.com page.
But this is useful in tons of other situations since we're slicing our data up so much these days. Take blogs for instance. A certain post might be accessible from the front page of a blog, from a permalink, from a comment page, from a trackback page, from a search page, from a tag or category page, etc. In the past all of these different views of the "same thing" would be seen as separate by google; now you can pick one view to be canonical. This just really makes sense to me. Thanks google.