i'm not terribly deep into OpenGraph, so there may be bits i'm missing. Still, things seem… oddly familiar.
So, if i understand the whole concept of OpenGraph is to add meta data onto a given web page. So you can take j-random page and add things like, title, site_name, type, url, and description. For example:
<meta property="fb:admins" content="12345678" />
<meta property="og:title" content="My Web Page" />
<meta property="og:site_name" content="Yet Another Blog" />
<meta property="og:type" content="website" />
<meta property="og:url" content="http://example.org/" />
<meta property="og:description" content="Random thoughts and observations" />
<meta property="og:image" content="http://example.org/sample.jpg" />
Of, course, there's no reason you have to provide "accurate" information. You can use pretty much any value you want.
<meta property="fb:admins" content="117415311618936" />
<meta property="og:title" content="Happy Super Fun Rainbow Land" />
<meta property="og:site_name" content="Evilon Astick" />
<meta property="og:type" content="website" />
<meta property="og:url" content="http://evilonastick.com/" />
<meta property="og:description" content="A wonderful place full of happy kittens and overseas pharmacutecals" />
<meta property="og:image" content="http://evilonastick.com/fb/eos_kitten.jpg" />
In fact, should you decide to visit Evilon Astick and look at the source, you'll see just that.
Should you decide to click on the "Like" button at the bottom, and add a "comment", you'll get the following on your Facebook page:
Honestly, this all seems… familiar.
In fact, if you go back and look at that page again, you'll see:
<meta name="keywords" content="sample,evil,sandbox,sock monkeys" />
<meta name="description" content="A playground full of evil things" />
<meta name="DC.title" content="Evil! On a Stick!" />
<link rel="canonical" href="http://evilonastick.com" />
Mind you, most of these ("canonical" is a pretty recent addition). Of course, there's also a reason that these are kind of ignored by most search engines, Spammers have used them to boost they're Search Engine Optimization (SEO) ratings by filling them with all sorts of junk. i honestly can't see where this will be any different. About the only difference is that posts only show up in folks Facebook news streams if they comment, but i don't see that as being a huge hurdle, or really that much more of an incentive. Again, i hope that facebook took precautions against clickjacking. Since they're running the "Like" button inside an embedded iframe, however, i'd be interested in seeing how they might dynamically check the parent opacity.
It's also nice that Facebook is planning on providing some additional information for all the new pages that decide to adopt this new structure, but i can't help think that with 10+ years of, well, existence, there's an awful lot of the web and various "Social Objects" that will never be part of this. i don't know if the current set of search engines and spiders will ever add OpenGraph to the content they're already parsing (one wonders that with the duplicate records, could a site with different data be penalized in some manner if those records differ in some form? Will high traffic sites really want to increase their bandwidth allocations to add the new content?
i'll probably add this for my personal sites, but mostly as a joke. Facebook is still very much a black hole as far as i'm concerned. i get little traffic from it and therefore little benefit. (Since i don't run ads, i consider discussion to be the most valuable element, and there still is no way for me to pull those elements of discussion back out of facebook easily, any discussion held there is lost to anyone without a Facebook account or who is not actively tracking those discussions.)
Wonder how long before OpenGraph becomes just as polluted as HTML Meta tags became?
