Early last year Howard Dean stormed into the Democratic presidential primary race, his crusade to reform American politics propelled by thousands of small donor, Internet insurgents.

Using the Internet as its fundamental tool, the Dean for America campaign raised in excess of $40 million from approximately 300,000 donors, and persuaded over 600,000 Americans to pledge their support to the upstart political outsider. At the same time, respected members of the national media, fascinated by Dean's innovative operation, prepared to anoint the little-known former governor of Vermont as the Democratic presidential nominee.

"The press became enchanted with the Internet," said Tobe Berkovitz, associate dean of Boston University's College of Communication, "but the media's reaction was all out of proportion to the reality of what was going on."

It wasn't until the campaign faced actual voting in the Iowa caucuses, and the New Hampshire primary, that Dean's Internet phenomenon was subjected to the crucible of presidential primary politics.

A devastating 20-point third-place rout in Iowa, followed by a slightly less damaging, 13-point second-place finish in New Hampshire, both to Sen. John Kerry, revealed significant weaknesses in what was believed to be Dean's unstoppable cyber-stampede to the Democratic nomination.

These very defects, and revelations that the organization had apparently burned through a significant portion of its campaign war chest, ended up costing Dean campaign manager Joe Trippi his job. Trippi's wizardry has been credited with the creation of the Internet fund-raising phenomenon and the campaign's powerful online recruitment program. He resigned Wednesday night after Dean appointed Roy Neel, a Washington insider who was once chief of staff for Al Gore, the campaign's "chief executive officer."

Does this apparent inability of Internet organizing to sustain the campaign of Howard Dean mean that online political organizing and fund-raising is a shooting star with no place in American politics?

Absolutely not.

Studying the Dean operation, one is struck by the depth and breadth of its sophisticated network of Internet functions. The campaign's main "blog" (an abbreviation for "Web log") allows dozens of Dean supporters to communicate online, with one another and campaign headquarters. It is these "conversations" created by the many spontaneous comments posted by fellow Dean partisans that the campaign values so highly.

A recent post on the Dean blog by senior staffer Mathew Gross reflects the campaign's confidence in its Internet operation when he said, "No one had any idea the extent to which the conversation you started would transform our democracy."

What the Internet has brought to political campaigning is just as remarkable a metamorphosis as was broadcast television in the early 1960s. But, the dynamic conversations it enables represent a significant advance over traditional forms of advertising.

As astounding a tool as the Internet is, it lacks the personal and persuasive commitment- building quality a candidate gains by listening to concerned American voters in face-to-face conversations.

Dean's self-evident success compelled the competing presidential campaigns of Sen. John Kerry, retired Gen. Wesley Clark, Sen. John Edwards and others, to develop similar Internet tools to attract thousands of volunteers, raise millions of dollars and initiate comparable Internet-based efforts.

In Iowa there was a critical difference; Kerry meticulously constructed a powerful network of elected state officials. In New Hampshire, Kerry national campaign chairwoman Jeanne Shaheen (a former governor of the state) built a similar structure, grounded in the organizations of experienced local officials. It used online organizing only as a tool to buttress its state organizations.

Mesmerized by their own Internet wizardry, the Dean organization, on the other hand, appeared to forget that politics is about listening -- in diners and church basements -- to the concerns and ambitions of real people. Excited by the virtual conversations on their Internet blog, the Dean campaign failed to appreciate the critical role of effective, local organizing.

The result was a self-congratulatory echo chamber populated by thousands of untrained, highly dedicated Dean partisans. It was a society committed to reinforcing the beliefs of its creators. This organization's inexperience did not prepare it for the predictable media pounding that Dean encountered as the Democratic front-runner.

Unlike John Kerry's local network of experienced elected officials, the Dean campaign chose to create an Iowa volunteer organization drawn from its Internet community. Christened the "Perfect Storm," a group of 3,500 youthful Dean volunteers descended on Iowa, from 11 states, with a mission to blanket the state and to persuade caucus-goers to support their candidate.

The "Perfect Storm" in Iowa was a disaster. It was viewed by observers and Dean volunteers alike as a significant reason for the campaign's disappointing showing.

"They reached the limit of the capabilities of their Internet universe," observed Berkovitz. "They hit their cyber ceiling." While the Internet was able to provide volunteers, it lacked the ability to provide professional campaign experience.

One "Storm" volunteer confirmed this analysis when he told us that he learned that "political campaigning was an art and most of us who went to Iowa had no idea how to go about it. We needed local direction and we did not have it."

With less than a week's time to the New Hampshire primary, the Dean campaign turned to the state network that it had spent two years building to close the gap on the well-organized campaign of John Kerry.

Dean campaign spokeswoman Dorie Clark said: "We're really focusing on New Hampshire natives. Obviously, it works better if they're in-state people." Their second-place finish in the Granite State seemed to confirm her conviction.

The Internet, like previous innovations in politics such as television advertising and direct-mail fund-raising, represents a new generation of political communications. Its conversational nature makes it seem to have the same impact as personal campaigning, but it cannot -- at the end of the day -- replace local organizing.

The lessons learned from Iowa and New Hampshire are that, left to function independently, these Internet tools have a tendency to create societies of insiders and outsiders instead of collaborative communities focused on dynamic political debate and action.

We are observing the emergence of a modern political synergy; a rapid integration of the old and the new, with a consolidation of the best from both. A new form of political campaigning that will combine the technology- based virtual networks with the local and highly interpersonal networks of traditional politics. It is this critical refinement that promises to distinguish new generations of online campaigning.

It is a marriage that looks to surpass the parts that have created it, an innovation that is revolutionizing the very nature of American politics -- a collaboration that could critically refine the political art of winning.