Today is Super Tuesday and we are fully in the thick of campaign season, much of which is unfolding online. It’s so inundated, so a part of our daily lives, that campaigning on and with the web feels like it was always part of the process. But it wasn’t. Decades before the president of the United States was using social media as a platform for belligerence and bullying, there was a time when campaigns didn’t quite know how they would use the web at all. And once they figured it out, campaigning would never be the same. So let’s talk about major elections on the web.

Like many other things, political campaigns didn’t take long to move to the web. It began on the fringes with small experiments no one thought would work. They often exceeded expectations.

Professional wrestler turned longshot gubernatorial candidate Jesse Ventura unveiled his website during the 1998 Minnesota governor race. As a member of the Reform Party, Ventura’s politics were as unusual as his campaign tactics, opting for grassroots support and unconventional ads over political business as usual.

Of all the unorthodox things his campaign tried, having a website wasn’t all that strange. Ventura’s staff collected email addresses for their mailing list and posted daily updates from the campaign trail to the site. From time to time, Ventura would post updates of his own, directly to his voters. A blog of sorts, before that word had made its way into circulation. Having a personal and direct line to voters was a cornerstone of the Ventura campaign, and he used his website as a conduit for that kind of candid interaction.

To the surprise of just about everybody, Ventura won, beating both Democratic and Republican challengers. To this day, he’s the only Reform Party candidate to make it as far as he did. It would be hyperbolic to attribute that success too directly to his online efforts, but that hasn’t stopped plenty of people from trying. In any case, his website couldn’t have hurt.

It was John McCain who turned to the web a couple of years later. During his presidential bid in the GOP primary ahead of the 2000 election, he ran campaign ads on a syndicated network that reached dozens of sites, like Excite and Yahoo. These ads, they were nothing radical, a digital reproduction of their TV and print ancestors. One of the banners, for instance, read John McCain needs your help… click here to find out how.

The ads, however, were unique for one reason. McCain’s campaign was courting Virginians, hoping to gather the ten thousand petition signatures required for eligibility on the ballot there. To target these voters, the campaign matched user lists on popular sites against public voting roll information. These users were served a specific ad with a specific request to sign the petition. It was the first ever targeted digital political advertising attempt. The ads proved successful, and McCain was the first to prove their effectiveness. By the time the 2000 general election rolled around, targeted ads would become the standard.

Since 2000, each subsequent US presidential primary has raised the stakes, building on former success to turn experimental, modest efforts into an intricately constructed machine.

In 2004, it was Howard Dean that picked up the banner. His campaign staff noticed early in his Democratic primary run for president that he was receiving outsized support from Internet users. When MoveOn.org conducted an online primary, Dean brought home 44% of the vote. He was sort of Internet famous, which made sense. Dean preached a progressive, anti-war, anti-establishment message that ran lock step with the egalitarian vision for a new future that sprung from the minds of the mostly affluent, liberal members of the early web (though a decade in, accessing the web still required technical know-how and significant investment). Besides that, Dean was a political outsider reviled by many in Washington and in the Democratic establishment. Web people still liked to think of themselves (however much it was or wasn’t true) as counterculture neo-punks bucking the mainstream; Dean was one of them.

At this point, it was standard fare for campaigns to have a website outlining policy proposals with links to sign up for an email list. If you did sign up you could expect somewhat frequent impersonal messages that read like a cold call from your politician’s phone bank. Ads run online followed the same format as their print counterparts. The web had already become an afterthought, a cheap, bolted on way to secure a few extra votes. Dean’s campaign wanted to go further.

Led by Joe Trippi and Zephyr Teachout, his staff began trying things out on the web. They leveraged sites like Meetup.com to organize local events across the country. They connected volunteers with one another to optimize canvassing efforts. They used a network of blogs to distribute Dean’s message and help raise money. And they leveraged off-line relationships of Internet fans of the campaign by encouraging supporters to reach out with handwritten letters to undecided voters.

As Dean gained momentum, his staff realized that they would need something that ran alongside everything else they were doing. They turned to a group of Drupal developers and Dean supporters who had recently begun calling themselves Americans for Dean. Drupal was still pretty new, launched a few years prior in 2000. It was initially created as a way for eight students at the University of Antwerp to talk to one another. After a couple of years, it evolved from a simple messaging board to more advanced website management software. As one of the earliest examples of a content management system, it empowered users to create their own websites, make changes, and edit content using a graphical interface connected to a database, rather than directly editing HTML files.

But what the Dean campaign had in mind would be the software’s true litmus test. Together they built DeanSpace, an online portal that supported a network of dozens of independent sites connected to a centralized system that could be used to organize events, exchange messaging suggestions, and deploy volunteers where needed. Using DeanSpace, organizers around the country could create their own dedicated websites without needing to code things up. And using the site’s internal tools, volunteers and organizers could discuss tools of the trade, policy, or anything they wanted. All with very little input from the campaign staff. In the words of one of its creators, it marked the the first ever national autonomous, self-organizing, grass-roots campaign network. A string of words that feels like nonsense until you think about it and realize that that’s pretty much how we politick these days.

In creating this network, the Dean staff handed much of the power of their campaign over to a network of essentially strangers. That was the unspoken deal struck, the leeway needed for web users to feel comfortable backing the campaign. There was some centralized coordination, but many decisions were made on the ground or in small groups at meetups or on DeanSpace. The campaign’s platform got split up, divvied out, and filtered through the experience of local groups and individual volunteers. When Dean supporters spoke to others, on the web and offline, they didn’t talk through rigid scripts dictated by the campaign, something you might expect from the canvassing efforts of other candidates. Instead many shared their own feelings about Dean, warts and all, and brought a personal passion to the message.

Intentionally or not, the Dean campaign was the first to adopt the language of the web, and as much as they influenced the digital landscape of politics they were, in turn, influenced by it. Users of the Internet, now as much as then, demand an authenticity that is somewhat mismatched from the vetted and constructed tone of a campaign. We on the web often pride ourselves on being able to see through the bullshit (even if its only through the creation of more bullshit). Dean was a political outsider, so that attitude fit him quite well. And as the campaign enjoyed more and more success, that spirit moved from online into the “real world” and morphed campaign messaging across every platform.

Dean lost the primary to John Kerry. If not for one fateful scream, he may not have. In either case, his digital tactics proved instantly viable and, members of his staff would conclude, reproducible.

As Dean made waves on the web, an actively engaged newcomer was making use of a similar, if not slightly cruder, version of the same suite of tools. When he ran for senator of Illinois in 2004, the same election cycle that saw the launch of DeanSpace, Barack Obama already had a blog. He had a Myspace. He had active BlackPlanet, MiGente, and AsianAve profiles. Like Dean, his anti-war stance and promises of change were appealing to a large portion of web users. It helped propel him to a junior seat in the United States Senate. He gained a small and loyal following, in his home state and across the country, one that would soon mean a great deal more.

The 2008 presidential election was different. It was the beginning of something. After years of trial and error, the tools had grown; the industry had itself a few veterans. Members of Dean’s staff had formed a company dedicated to utilizing these new web strategies to promote Democratic candidates. They found in Barack Obama the perfect candidate to try them out. He was (relatively) young, well-liked in Washington but with a healthy skepticism of the establishment, and a message of a hope that connected with young, and importantly digitally native, constituents.

Obama rounded out his Internet-focused staff by adding Facebook co-founder Chris Hughes. They deployed many of the same tactics used by other national campaigns, first against establishment democrat Hillary Clinton and then in the general election against Republican challenger (and, as we might remember, early web adopter) John McCain. Obama’s campaign gathered email lists and arranged meetups and published daily updates to a blog. They gathered millions of followers across social media accounts. During his presidency, they were early Twitter adopters.

Hughes, however, brought a fresh idea to the table. He helped transform a section of the official Barack Obama website into a full-fledged social network of politically engaged supporters. Located at My.BarackObama.com, users could create a profile, exchange messages, form groups, and a lot of the kinds of things you might expect from a social media platform run by an election campaign. It was ridiculously successful. By the end of the campaign, it would have over 1.5 million users and drive over $600 million in donations.

Yet while the Dean campaign let its grassroots organizers mix in their own personal brand with the message of the campaign, Obama’s staff harnessed that same enthusiasm and digital reach with a much greater focus. They used targeted advertising available to commercial brands to reach potential donors. They followed trends and tried to move the online conversation through a number of surrogates and pundits. They created short, uplifting videos for the YouTube generation. In other words, when it came to digital efforts, they treated Obama like a product.

And to help sell that product, they turned to their army of supporters. Volunteers on MyBO, as the site was affectionately called, were coached in how to take the language of the web to their communities. They were encouraged to mix personal stories with the message of the campaign, to maintain an informal, conversational tone, and to keep their focus on hope for the future. Gone were the stilted recitations of policy proposals and dry biographies. MyBO users flooded YouTube, Facebook, Twitter, and every network you could imagine with their personal takes on pro-Obama messages. If you were standing on the outside, and your primary connection to the world was the web, his support appeared to be insurmountable. It was everywhere and it felt personal and genuine because it mostly was.

Obama triumphed and was elected as the 44th President of the United States. He had support from many places, but the Internet gave him an unquestionable boost. As he made his way into the oval office, Obama brought with him the lessons he learned on the campaign trail. For one, Obama is often called the first social media president. And that may have been inevitable given the timing of his presidency, but just because he was president of the web generation, doesn’t mean he had to be so good at it.

But the spirit of the web mixed with the Obama administration beyond simple platform adoption. One needs only to look at his digital initiatives, MoveOn.org, the USDS, the POTUS Twitter account, and the numerous contests and calls to action the administration issued throughout his presidency. The tone and the pace of the web, which happened to come into maturity and change dramatically in the years of the Obama presidency, began to merge with the executive branch of government.

And that’s important, because by the 2016 campaign (after a more quiet win in 2012 by the incumbent), that tone would take on a greater role at the intersection of our political and online identities. That kind of intersection has happened before. It’s the same impulse that drove us to (at least attempt to) replace mainstream newspapers with the blog. In politics, we got comfortable with politicians, especially those on the national stage, speaking to us in the same raw stream of consciousness you can find on any Twitter feed. We encouraged it. In the simplest possible terms, we wanted a president that could talk like us, to us. The web has been about breaking down barriers and walls since the beginning and it is absolutely no surprise that we aimed, and expected, to do the same to the highest office in the country. The language of the web has a flip side though:a blend of weaponized irony and theatrical spectacle that we go nuts for on the web.

So here I am, over two thousand words into this thing and I’m just now reaching the campaign of our current president, Donald Trump. And there’s a lot I can say about it. We are still, as a nation and in the thick of it, parsing out the effects of how that campaign was run. What I can say is this. Trump adopted the language of the web too. He just found, and embraced, its darkest corners and worst impulses.

That language includes every voice, even the extreme ones; in some cases, especially the extreme ones. And that’s a word that best describes the campaign of Donald Trump. He used the same playbook as the digital campaigns that came before him, but in each instance he took it to its logical extreme.

His campaign team amplified micro-targeted advertising and digital branding and used it to blanket the web with misinformation. He turned to viral, visceral campaign messaging on social networks, working directly with members of Facebook to turn the tide of opinion, instead of relying on much slower, old models of television and print advertising. He adopted the cadence and dialect of online speech on social media to talk directly to his potential voters, often to go off on controversial tirades and rile up bigoted sentiment. On the web and offline, he used the kind of sensationalist, yet candid language that, in the past, has helped politicians demonstrate to voters their trustworthiness. Every major tactic of previous elections were combined and remixed during the Trump campaign.

That is not to say that a campaign like Trump’s was in anyway inevitable. As a historian, it’s my job to trace that through-line of commonalities. I wouldn’t mistake that with a natural progression. There were literally hundreds of factors that led to the Trump presidency. The Internet is a large part of that story. But it’s not the whole one.

What I can say is this. He is, in many ways, a product of the web we’ve created. The web will always be two sides of a coin flip. It lower barriers of communications to create a globally connected world, but it also acts as an amplification of our prejudices. When we drew one side into politics, we also got the other. In Trump we find the language of the web and the tactics of Internet campaigning sharpened to a point and wielded as a weapon. He knows how to say terrible things on the Internet out in the open and hide behind a veil of feigned ignorance, a defense liberally employed by online trolls.

We are beginning to understand something we should have realized a long time ago. There is not a separation between our digital lives and our lives offline. The language of offline and online have become one. The actions of a candidate behind a screen are no less effective or authentic than anywhere else. So when we talk about elections on the web, let’s remove that distinction.