May be I am missing something. Isnt Omnilink just detecting the user agent (eg Google/Twitter crawler) and presenting the right content to the crawler? Is there something else to this solution?