Google's search algorithm works on a simple principle: on the web, sites link to other authoritative sources, and the more sites link to something the more authoritative they think it is. That's great if you're dealing with sites that actually follow that rule – as the vast majority do.
But what do you do to make your search engine useful when you have a class of websites that produce almost identical content but which barely ever link to each other? What does Google do about news websites?
They generate new content all the time, sometimes using (or copying) each others' work without acknowledgement – particularly without those links to each others' sites that Google's standard algorithm relies on to measure authority. What do you do with a chunk of the web that doesn't really behave like the rest of the web?
That's the problem facing Josh Cohen, senior business product manager for Google News. He can't win. Since Google News arrived in April 2002, and in its more current form that September, publishers have had a love-hate relationship with it. They complain that it is "stealing" readers by providing a product similar to their own but without producing any content, yet desire prominent placement so that people will come and read their stories by clicking through from it...
Tuesday, October 13, 2009
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.