by David Greenberg

In the fall of 1990 I was fresh out of college and interning at The New Republic, taking in the excitement of living in Washington for the first time and working at a leading liberal magazine. Sometime in the early afternoon on Election Day, Sid Blumenthal, then a senior editor at TNR, walked into the office with exit poll numbers. Back then, these numbers were a closely guarded secret, and as he ran down the projected outcomes of gubernatorial and senatorial races, one felt the thrill of getting the inside dope. I picked up the phone--no email back then--and called some political-junkie friends who, like me, were low-level flunkies on the Hill or at media outlets. For the rest of the day we eagerly swapped information. What stunned me was that, as I recall, every single one of Sid's projected outcomes was spot-on. The numbers may have been off a little bit, but every race came out as we expected in the early afternoon. A few years later, as we all came to use the Internet more and more, I realized it would soon be impossible to keep these numbers confined to a few thousand Washington insiders. If I, as an intern, was calling people around the country with the inside dope in 1990, what would happen if someone emailed everyone he knew with the numbers--or posted them on a website? By 2000 Slate began to do just that. In 1998, Slate had refrained from publishing exit poll numbers, but Tim Noah wrote a piece protesting the participation of Internet journalists in the blackout. (His piece, here, bears reading.) Two years later, his arguments prevailed. Slate, along with other websites, began posting the numbers in the New Hampshire primaries and other early presidential races. It stopped for awhile when the Voter News Service, the consortium of media organs that sponsored the polls, threatened legal action. But if Slate was muzzled, other sites weren't, and numbers continued to appear on the Web on Election Day 2000. Moreover, in 2000 the exit polls turned out to be atypically wide of the mark. The Voter News Service, discredited, soon disbanded. A new consortium has replaced it, but the legal threats that VNS wielded evaporated. By 2004, everyone with a website and a friend in Washington or the media was posting exit polls all day long. Unfortunately for those of us who popped our champagne bottles early, they proved unreliable once again. Surely we've now passed a point of no return. Look for exit polls to pop up around the Web today--I'll try to provide them here, though I'm sure most Open U readers frequent other sites that are in much closer proximity to the horse's mouth. I'll be awaiting the numbers keenly. But until I actually hear George Allen's concession speech, I'm going to be keeping the champagne on ice.