As the number of people using WordPress grew steadily, more and more needed to get done. Matt took a job at CNET, which allowed him to work on WordPress alongside his day job. Ryan was working at Cisco, often 60 or 80 hours a week; he would come home to work on his hobby, WordPress. Apart from time spent on the project, there are all sorts of associated costs that come with running a growing project. Server costs, for example, that increase as the project’s popularity grows.
At the start of March 2005, WordPress 1.5 had seen 50,000 downloads. Just three weeks later, the number doubled to 100,000. To celebrate the landmark, there was a 100K party in San Francisco. On March 22, a group of WordPressers got together at the Odeon Bar in San Francisco.
Jonas Luster (jluster) was at the party. Jonas was an employee at Collabnet, the company behind Subversion. He and Matt had been talking about what a company built around free software should look like. Matt had an idea for an organization called Automattic: an umbrella company that would include several WordPress organizations. The first of these was WordPress Incorporated. At the party, Matt asked Jonas if he’d like to be involved. Jonas said yes.
Shortly afterward, Matt jumped on stage to announce WordPress Inc. — with Jonas Luster as employee number one. By the following morning, a couple of blogs had covered the announcement. A few days later, Matt went on holiday leaving WordPress’ very first employee in charge.
Even as the party was going on, trouble was brewing. In February, a user posted to the WordPress.org support forums asking about hundreds of articles hosted at wordpress.org/articles/. The articles were about everything from credit and healthcare to web hosting. The thread was closed by a forum moderator. Blogger Andy Baio discovered the thread. He contacted Matt to ask about what was going on.
Matt explained to Andy that he was being paid by a company called Hot Nacho to host the articles on WordPress.org and that he was using the money to cover WordPress’ costs. “The /articles thing isn’t something I want to do long-term,” he told Andy, “but if it can help bootstrap something nice for the community, I’m willing to let it run for a little while.” In addition to responding to Andy, Matt re-opened the support forum thread and left a response:
The content in /articles is essentially advertising by a third party that we host for a flat fee. I’m not sure if we’re going to continue it much longer, but we’re committed to this month at least, it was basically an experiment. However around the beginning of Feburary [sic] donations were going down as expenses were ramping up, so it seemed like a good way to cover everything. The adsense on those pages is not ours and I have no idea what they get on it, we just get a flat fee. The money is used just like donations but more specifically it’s been going to the business/trademark expenses so it’s not entirely out of my pocket anymore.
On March 30, while Matt was in Italy, Andy published an article about it on his blog. The reaction from bloggers was bad: no matter what Matt’s intentions, people saw the articles as a shady SEO tactic. A free software project should know better than to work with article spammers. Not only was WordPress.org hosting the spammy articles, CSS had been used to cloak them. If everything was above board, why not do it out in the open? The articles also had an influence on WordPress.org’s hosts, TextDrive, who hosted WordPress for free on a server along with apps such as Ruby on Rails. Their servers were overloaded with hits from the links in the Hot Nacho articles.
With Matt out of town, Jonas dealt with the fallout from Andy Baio’s article on Waxy.org. It wasn’t clear where the line between WordPress.org and WordPress Inc. lay, but Jonas was the only official-sounding person. He fielded the tech community’s ire and spent the next few days putting out fires. Jonas posted on his blog asking for calm, and that Matt be given the benefit of the doubt.
The Hot Nacho articles were perceived as the first move by WordPress Inc. to make money — an indication that Matt and Jonas were’t going to monetize WordPress in an ethical manner. Jonas stressed that the articles had nothing to do with WordPress Inc., but since it wasn’t clear at that time what WordPress Inc. actually was, people naturally lumped them together. By the next day, Andy had updated his post to report that WordPress’ PageRank had dropped to 0/0. WordPress.org had been removed from Google’s search engine results.
The story escalated as major tech resources like The Register, Slashdot, and Ars Technica picked it up. On March 31, Matt posted briefly on his blog to say that he had just learned of what had happened and that he would get online to address the questions as soon as possible. The comments are littered with messages of support from people who would later be central to the project and even from Matt’s rival. Matt also received support on WordPress’ support forums.
On April 1, Matt issued a full response. In it, Matt explains that he was struggling to find a way to support the free software project, and that he felt that his options were limited:
To thrive as an independent project WordPress needs to be self-sufficient. There are several avenues this could go, all of which I’ve given a lot of thought to. One route that would be very easy to go in today’s environment is to take VC funding for a few million and build a big company, fast. Another way would be to be absorbed by an already big company. I don’t think either is the best route for the long-term health of the community. (None of these are hypothetical, they’ve all come up before.) There are a number of things the software could do to nag people for donations, but I’m very hesitant to do anything that degrades the user experience. Finally we could use the blessing and burden of the traffic to wordpress.org to create a sustainable stream of income that can fund WP activities.
Hosting the articles on WordPress.org attempted to mitigate the free software project’s costs. For many people, Matt’s response satisfied their concerns and questions. But for others, it wasn’t enough. If a free software project has to resort to turning its highly ranking website into a link farm, what is the future viability of that project? Aren’t there other ways to support it? Matt had been at the web spam summit in February, which had specifically addressed “comment spam, link spam, TrackBack spam, tag spam, and fake weblogs.” Hadn’t he, at that point, realized that there might be something dubious?
As with most storms on the internet, once the articles were removed and Matt had apologized, the anger subsided. It did, however, have a lasting influence. The experience changed Matt’s thinking about spam. Instead of viewing it purely as something that appears in an email inbox, he now saw it as web spam, which can quickly pollute the web. This influenced his thinking when developing Akismet, Automattic’s anti-spam plugin. And while Matt had learned a harsh lesson about ethical ways to make money on the web, it would return to haunt him again and again. Any time WordPress.org clamped down on search engine optimization tactics, in themes for example, irate community members would bring up Hot Nacho.
Jonas’ first few weeks at WordPress Inc. got off to an explosive start, and over the next few months he continued to work unpaid as WordPress Inc.’s only employee. He got in touch with major web hosts and created partnerships so that WordPress.org could make money by recommending hosting companies. He met with venture capitalists to talk about where WordPress was going.
At an IRC Meetup in August 2005, Matt discussed some of the ways he wanted to see the project supported. Instead of fundraising, he wanted to see a company that could generate revenue. He said that “the goal, eventually, is for it to be the biggest contributor someday, supporting community members to work on WP full-time.” To protect the project from company liabilities, donations would be kept separate. This would ensure that if the company was sold, the project would be safe. In this chat, the “inc.” that Matt refers to isn’t WordPress Inc., but the as-yet-unannounced Automattic. When asked about employees, he said that the only one was Donncha, with Ryan Boren planned for the future. WordPress Inc. itself petered out, disappearing with much less fanfare than when it arrived. There wasn’t money available in WordPress Inc. to pay Jonas a salary, especially when Matt was using his own salary from his job at CNET to pay Donncha to work on WordPress.com.
The wider community was confused about what had happened with WordPress Inc., especially those who had watched Matt announce the company’s launch on stage. In response, Matt posted a page on the WordPress Codex, explaining what had happened:
At the WordPress 100k party in March 2005, I talked about the formation of a “WordPress Inc.” with Jonas Luster as the first employee. That never really got off the ground, I continued my job at CNET and Jonas (who I couldn’t afford to pay a salary) ended up going to work for Technorati. It wasn’t even planned to be announced at the party (since it wasn’t clear logistically how it would happen) but everyone was really excited about it and I had an extra G&T (or two), we all got carried away.
The first movement toward making money to support WordPress was more of a stumble than a step. It did produce some important lessons, however, about how to run a business alongside a free software project. In any attempt to run a business based on a free software project, the company is beholden to two players: the company itself and the community. In terms of the company, investors have to be kept happy and employees have to be looked after. The company needs to make money. On the other hand, it is the community that is invested in growing the free software project, and assuring its integrity and independence. Walking the line between the two can be difficult for any community member, but particularly for a project founder or lead who has multiple reasons for securing the future of both.