After the events of this summer, a reasonable person could conclude that Dalton Caldwell is the craziest person in Silicon Valley.

Just this past week, he touched off a digital firestorm after a writing a scathing blog post accusing Facebook of trying to bully him into selling his company.

Before that, Caldwell wrote another post criticizing changes in Twitter that prompted such a response he decided to reboot his company to focus on building an "open" version of Twitter's service.

But taking on the two Titans of social media isn't even the craziest thing Caldwell has done. This is: Rather than building his new business on advertising, he wants people to pay to use it.

"I probably am insane," Caldwell said. "But I think there's a place for crazy things."

Yet, this is Silicon Valley, where the line between genius and crazy can be blurred. So don't be too quick to dismiss Caldwell until you hear his whole story.

First, while he's not a household name, Caldwell, 32, is extremely well regarded in startup circles. He's not just a bomb-throwing blogger out to score some cheap pageviews.

He's on his second startup. His first, a music company called Imeem, was started in 2003 and bought by MySpace in 2009 for a reported $1 million. A candid talk Caldwell gave at an entrepreneurs' event about lessons learned and mistake he made with Imeem was a modest sensation. The response convinced him of the merits of being

"The fact that I was willing to be transparent about what happened earned me a great deal of respect," Caldwell said. "That was a defining moment in my career. If I tell the truth, it doesn't actually hurt."

A few months later, Caldwell launched his new venture, Mixed Media Labs, with the idea of testing and creating possible mobile applications. Among the first products to come out was picplz, a mobile application that combined photo sharing and social networking.

Caldwell attracted funding from some notable backers, including Andreessen Horowitz. Partner Marc Andreessen joined his board and the future seemed bright. That is, until another mobile app called Instagram launched a few months later and took off like a rocket.

Andreessen Horowitz had also backed the founders of Instagram when they originally started as a check-in app. When they shifted to mobile photo sharing, the venture firm declined to invest any more, citing a conflict because they were already backing picplz. That caused some second-guessing when Facebook bought Instagram for $1 billion.

"We are excited and enthusiastic investors in Dalton's company," partner Ben Horowitz wrote on his blog in April after his firm was criticized for backing the wrong horse. "We expect great things from Dalton and look forward to another massive return from his new idea."

Still, recognizing that Instagram had won, Caldwell shifted to focus on his next Mixed Media company, called App.net, designed to help people creating mobile apps get them noticed on Facebook.

In June, according to his account posted on Wednesday under the title "Dear Mark Zuckerberg," Caldwell was invited to Facebook to give a demo of the product. Instead, when he arrived, a team of Facebook executives said they were building a competing product, but were interested in possibly buying Caldwell's company and his team, a so-called "acqui-hire." They also implied they would otherwise crush Caldwell's efforts.

"I know all about intimidation-based negotiation tactics: I experienced them for years while dealing with the music industry," Caldwell wrote to Zuckerberg. "Bad-faith negotiations are inexcusable, and I didn't want to believe your company would stoop this low."

The post was an immediate social media sensation. But it also led to potential conflict for Andreessen because of his role on Facebook's board. Andreessen stepped down from App.net and was replaced by Scott Weiss, another partner at Andreessen Horowitz.

Weiss was traveling and couldn't be reached for comment Friday. A Facebook spokesman also said the company had no comments on the blog post.

Shortly after the Facebook meeting, Caldwell wrote another post expressing his disappointment over changes taking place with Twitter's service. While there's still some controversy about what these changes mean, they have been widely interpreted as moves by Twitter to exert more control over its platform with a series of changes that will disrupt the work of many third-party developers who have been critical to the service's success.

These changes appeared to be driven by a need to make Twitter more attractive to advertisers by creating a more uniform experience. Caldwell, in his post, lamented that the company was chasing ad dollars to the detriment of users and developers. Instead, he wished that Twitter would focus on building infrastructure that others would use to build services.

That post got so much feedback that it got Caldwell thinking: Rather than wishing Twitter had built the service he had in mind, he's doing it himself.

So here's his plan: App.net is building a service that offers a similar infrastructure to Twitter, with a few tweaks. Users would pay some amount to use it, and that money would be shared with developers who create applications that attract users to the network. On Friday, App.net began accepting a few users to test an alpha version of the service. Caldwell is also running a campaign to raise money on the crowdfunding service Kickstarters, though not because he needs the money, but rather to see how much interest there is in the idea.

Caldwell doesn't expect the service would ever match the size of Twitter or Facebook. But he thinks there is a potential market for users and developers who don't want their social media feeds filled with advertising.

In response to Caldwell's proposal, Union Ventures partner Fred Wilson said he believes such paid services are quixotic, and that free, ad-based services are superior business models.

"I understand the frustration of certain folks about the commercialization of Twitter, Facebook, Tumblr, and a host of other services," Wilson wrote. "But I would not encourage those same folks to build paid services. I think their goals will be undermined by that choice."

"I think this is disruptive," Caldwell said. "We're trying to invert the economics of Twitter."

Not just Twitter, but really, this idea challenges the entire philosophy that this generation of Web 2.0 companies was built on. Get huge numbers of users with a free service. Then sell those users to advertisers. That is the conventional wisdom. But it's worth noting that companies from Facebook on down are still struggling to prove they can be sustainable businesses. And those doubts only seem to be growing rather than receding after the Facebook IPO.

So while Caldwell's approach is a long-shot, there may also be no better time for a fresh approach to not just the technology, but the business model.

Yes, it's a crazy idea. But sometimes, crazy is what the valley needs to make progress.

Contact Chris O'Brien at (415) 298-0207 or cobrien@mercurynews.com. Follow him on Twitter at obrien and read his blog posts at http://www.siliconbeat.com.