[hugh3]

My point is that if you want to evaluate quality you have to leave metrics behind and start looking at opinions. And opinions are like assholes in that they're ubiquitous throughout the population.

Of course you can make up metrics based on averages of everyone's opinions, or weighted averages where you weight by the "sophistication" (in some sense) of the person's opinion. But nobody will believe such metrics anyway, in that you'll never catch anybody saying "I hate this game/book/album, but this reliable metric says it's good, so I guess it must actually be good!"

If Farmville is bad, it falls into the "fascinatingly bad" category of things that are derided by professional opinion-makers but still vastly popular. Genuinely bad games, books and bands are a dime a dozen -- check out a publisher's reject pile, an open mike night or a twelve year old programmer's hard disk for examples -- but fascinatingly bad things like The Da Vinci Code or Nickelback or Farmville deserve a lot more analysis.