They need to do something. Neilsen has destroyed more televised art than anything. Yes, they need something to measure who is watching what, and not every show that we love is going to appeal to enough people to justify the expense. I get that. But as far back as the 1960s people were proving impirically that Neilsen was inaccurate in many cases. I refer to Star Trek TOS which may not have been cancelled in 1969 if certain criteria were applied.
Today, Neilsen is even less relevant because of all the different media choices available. A show may air on a network one night, be replayed a few nights later, and then streamed online, and also be downloaded (legally or otherwise). But too often the ratings apply to only the first night. People who hate Enterprise refuse to believe this, but there was plenty of evidence provided in this very forum 6-7-8 years ago that showed Neilsen wasn't showing the full picture when it came to that show's viewership. I'll never forget Connor Trinneer pleading with Starlog readers not to watch the weekend rerun because all Nielsen cared about was Wednesday. And at the time Enterprise was on, the whole idea of watching TV shows online was still in its infancy, as was the notion that many people don't watch shows until they hit DVD now.
Case in point, I'm not planning to watch Fringe this season - I'll buy the DVD later. Yet under Nielsen's decades-old criteria, if I'm not in front of my TV tonight at 8 or whenever, I don't count. Yet, frankly, there's zero difference between me deferring viewing a show for X number of months till the DVD arrives and if I'd videotaped the thing and watched it a few weeks later. And in most respects it's better because I'm paying for the privledge.
The only problem with coming up with Nielsen 2.0 is that I honestly can't think of a truly accurate way to measure viewership without entering into Big Brother territory - tracking what we watch on our DVD players or iPods or computers.