A tasty meme war is afire, over polls, who does polling, and those who are polled. It tells us a lot about polls - and even more about how the two major campaigns are seeking to promote their wares via the Web.

Meme, in Web parlance, is a word used to mean an image, idea, phrase, or video that rockets throughout the cyberworld, viralized and reviralized, until "everybody knows about it." (The word is most commonly used for images, such as Dancing Baby, Pepper-Spray Cop, or the rainbows of Double Rainbow Guy. But a meme can be almost anything you can distribute via the Web.)

Memes work their way into the mass consciousness - at least, the communications-savvy mass consciousness that runs the world.

That is why memes are a prime battleground for our current politics. Both parties wage proxy war via memes, slamming and blamming it out in the Internet world.

A great example just occurred, in the Unskewedpolls.com debate, which showed how poll results could be transformed into an epic struggle, waged via memes, to reframe the very meaning of polls.

Meme: During the week of Sept. 24-28, a raft of state and national polls came out, done by all sorts of pollsters in all sorts of ways. In the aggregate, they suggested that President Obama had a five-to-seven percentage point lead in the national popular vote over Mitt Romney. The president's lead was even wider in several of the "all-important battleground states."

So that's the first meme, the great takeaway idea, and the Democratic party and its minions humped it big-time: It's inevitable. Romney's fading, the president picking up steam.

Countermeme: But the GOP and its phalanxes were ready with a countermeme. Enter Unskewedpolls.com. That website, run by Dean Chambers, a Republican blogger and Romney supporter from Virginia, claims to correct polls that Chambers says give too much weight to Democratic votership. Six-percent Obama leads turn into 7.8 percent Romney leads.

(Other sites, such as NumbersMuncher, have made similar suggestions, but Unskewedpolls is having its moment, or meme-ent.)

Instant global brushfire erupted, initially almost all of it blog- and Twitter-based, including a reasoned defense by Mark Blumenthal of the Huffington Post, and a withering, fact-rich critique by Nate Silver at Fivethirtyeight.com. Silver surveyed the history of political polls and found them nearly as likely to overweigh GOP voters as Democrats.

The blogsite Talking Points Memo profiled Chambers, his methods, and the blowback. The controversy reached such moment that it burst into the mainstream, making it into NPR's Morning Edition, Time, the Atlantic, and the New York Times. And it's still there, as it was meant to be.

Counter-countermeme. As of this writing, the consensus from the statistics/math/polling community, both rightward and leftward, is that Chambers doesn't understand how polls are conducted. Pollsters, working their myriad ways, use complex modeling methods to weight their samples and their analyses. So it's not, and never is, simply a matter of how many Democrats and/or Republicans answer your phone calls, never just count 'em and call 'em. And this, too, became a meme, merrily flogged by Democrats.

Ah, but the beauty of memes is that even though they don't last long, while they do they have force, just from the sheer brio of the initial push. So the countermeme Yeah, well, polls are biased toward Democrats thrives, even though the counter-countermeme of Yeah, but that guy doesn't know what he's doing is widely accepted. In fact, the two buzz on together.

That's right: Opposing, contradicting ideas coexist. This isn't reason here; this is what happens. Some would say this gives fact and counterfact a false sense of equivalency.

Too bad. We have meme, countermeme, and counter-countermeme, all coexisting, all battling it out. Who won?

The final counter. The counter-counter-countermeme was a vigorous, frontal assault, as Matt Drudge, one of the godfathers of the blogosphere, joined forces with Rush Limbaugh and other media figures in reminding people that No one really believes polls, anyway. Fox News was very industrious in this regard. An August Rasmussen poll, which found that respondents think media bias is an even worse political problem than campaign contributions, was dredged up. Other polls popped up like destroying angel mushrooms, showing - and is this surprising? - that, yes, people think polls are biased and few people believe polls.

And that is the state of the debate. That final counter is very powerful, by nature of its populist, contrarian, curmudgeonly energy. We like to believe ourselves superior to the professionals, able to swat down their findings just by our native wit. That's why, in Web politics, facts are easy to undermine, and the authority of informed analysis evaporates beneath sheer offhand skepticism. And it's in the parties' interests to encourage that thinking.

It's like soccer: If you can cause confusion in front of the goal, sometimes you can score.

Obama's lead, if any, is small and is likely to tighten or disappear. The anxieties of a close election supercharge the memer's parade, and we'll see a lot more before we're through.