As more time goes by, the saga of Matt Mullenweg vs WP Engine continues to devolve. While many will paint it as a good guy / bad guy scenario, the truth is that both sides have made good points alongside bad behavior. I can sympathize with both sides for the (often self-inflected) situation that they’re in, but feel like we need to clarify a few misconceptions coming from the peanut gallery on Twitter/X.
A lot of people have been retweeting something I posted as blanket support of Matt Mullenweg, so I wanted to expand on what I wrote because I feel like the underlying message wasn’t fully received. Here’s my tweet for context, which I wrote in response to an Automattic employee pointing out how much Matt Mullenweg contributes to WordPress:
This was something I realized the more I contributed as well. The wider ecosystem provides a ton of visible value but Matt has been putting an insane amount of resources into keeping the lights on and the engine running. (link)
A few people caught on to the tongue-in-cheek use of the word “engine”, but the key phrase here is actually “insane amount of resources.” Insane. I meant it. I’ll never understand the crowd that believes contributing to WordPress is just lining Matt’s pockets. I mean, in the sense that it helps Automattic raise capital, sure, but WordPress has been helping all of us in the ecosystem make money. Let’s not take that for granted.
I still believe Matt’s intentions in waging a battle of FOSS vs PE are good, but to be clear: I don’t agree with his tactics. Unless some new information comes out- which let’s be honest, is inevitable- he’s definitely entered a scary “ends justify the means” territory. Our agency is one of many who has had their workweek disrupted, and our clients are starting to have real issues from Matt’s decision to cut WP Engine off from WordPress.org. I never would’ve expected Matt to compromise individual WordPress installations and act in such a user-hostile way, knowing that the majority of people affected right now are users, developers, agencies, hosting company employees, and literally everyone except for his enemy, Lee Something-or-other at Silver Lake, who I imagine is still lighting cigars with hundred-dollar bills while WP Engine is widely perceived as the “victim” here.
All that said, what Matt has done is not “against open source” or “misusing the non-profit WordPress Foundation” anything like that. We don’t need to “change the license” of WordPress because of this. He absolutely had every right to do what he did to WP Engine, whether we agree with it or not.
As brutal as it was, by cutting WP Engine off from WordPress.org’s distribution, Matt Mullenweg is making the point that what has made WordPress successful is not just the code, but the seamless distribution of core, theme, and plugin updates that are hosted and deployed by WordPress.org. Matt is reminding us that if everyone was still manually uploading PHP files to update core and plugins, SFTPing into servers, or setting up complicated deployment commands (which, let’s be clear, some of us definitely do)- if those were the only workflows available, WordPress wouldn’t be nearly as popular. One click updates and plugin installations on top of two decades of backwards compatibility? Show me another open source tool as user-friendly as that. WordPress isn’t some developer framework for tinkerers, it’s much more broadly adopted than that.
And to be clear, that distribution network, including the plugin directory where 60,000 open source plugins are hosted, is not itself “open source”. This infrastructure has been subsidized primarily by Automattic. Love it or hate it, it’s Matt’s to do what he pleases with it. If you’ve been relying on one-click core updates and the plugin directory to manage the distribution of open source plugins to/from websites (which of course we all have), then you have been playing in Matt’s sandbox. This sandbox is not “open source WordPress” or the WordPress Foundation. It is WordPress.org, which is a wonderful thing that Matt Mullenweg owns and has stewarded for the last two decades. But it is not “ours”.
Let’s break down “open source” WordPress and, more importantly, what exactly counts as contributions. The way I see it, there’s roughly three types of contributions to WordPress:
- Improvements to the core codebase by writing code, filing issues, testing tickets, designing features, etc.
- The wider “meta” project hosted on WordPress.org, i.e. things that aren’t technically the code of WordPress itself. This includes helping out in the documentation, the official support forums, OpenVerse, Photos, the mobile apps, managing the global distribution of core updates to 43% of the web, hosting the plugin and theme directories, etc. Those last few are big!
- Bringing people to the platform. Teaching WordPress, sharing, learning, writing, building free plugins and themes, building premium plugins and themes, onboarding new users, volunteering for meetups and conferences, and even (ugh) wearing swag. And yes, some of these can make you money, but all are necessary for ecosystem health.
So all three types of contributions are important, but in the community, there’s been a huge imbalance towards the third type because it’s much easier to create and monetize a plugin or a blog than it is contribute code to core. And we all need to make money to support ourselves. I don’t directly monetize my YouTube tutorials. I do them because I enjoy them, but they might help support me financially in other ways by increasing my visibility in the community. The majority of what WP Engine calls “contributions” also fall under that third category, though they definitely employ a handful of core contributors as well.
The thing is, Matt seems to really undervalue people who contribute to the ecosystem in that third way, but the wider community seems to equally undervalue the importance of those first two types of contributions, core and meta.
The reality is that only number one, the core codebase, is thing thing we think of when we say “WordPress is open source”. It’s the thing Matt can never take away. The code running your site is yours forever. He can never shut down your site or make you pay for your code. You do have to host it somewhere though.
Similarly, the code of any plugins or themes running on your site are yours forever.
Everything else, including the directories on WordPress.org that host all those plugins and themes and provide your one-click install and updates? That’s not really part of the “open source” deal. Those are things Matt has freely provided to grow the ecosystem, but he that giveth can take away. And this isn’t just a WordPress issue. We can assume the same single point of failure for any repository or package manager of open source code, from NPM and Composer to Microsoft’s GitHub or even Homebrew. All of these cost money to run, and our industry has grown reliant on them.
To be fair, developers shared their plugins on the directory with implicit trust that their code would be openly distributed to any WordPress site, so it’s a bummer to see that trust broken and those plugins weaponized in this. But it’s not anti-open source. The plugins are still available for manual download. And Noel Tock has persuasively argued that there may be even more incentive in the future for multiple plugin marketplaces from companies like WP Engine, like the one that Automattic has been building at WordPress.com, as fracturing as that sounds.
So where does that leave us?
As I stated in my other post, I disagree with Matt’s approach here. He’s been childish in his interactions with the community1, disingenuous in his presentation of the facts2, and disrespectful of the trust the wider ecosystem has placed in him. All that aside, he is still correct that WordPress might need to make some critical changes to ensure its future success.
I can see the effects the owners are having on WP Engine, but I’m not fully convinced that it’s Matt’s problem to solve. Feels like something the free market would eventually correct for. And the trademark dispute? I’m not a legal expert, but from the outside it feels a bit like going after Al Capone for tax fraud, just finding some place they might have leverage. Because just like Matt, WP Engine is also doing nothing “wrong” here. Their owners are under no obligation to contribute more and treat their customers better, sadly.
Either way, as I laid out before, instead of “going nuclear” I suggested that Matt explicitly clarify his expectations from large companies, outline a reasonable timeline for compliance (publicly, so that customers can have time to “vote with their wallets” before), and engage community support.
How should those changes be enacted? I think Joost deValk has some fantastic ideas to start with. You should read it. Matt should read it. In fact, if I were running WordPress, and I had successful, intelligent business leaders like Joost around who are contributing back into the platform, I would be working with them more often, getting their early feedback on strategies before I go public, and building a board of like-minded individuals to solve WordPress’s problem together.
As of this writing, the most up-to-date insights into the situation come from watching Matt’s interview with The Primeagen. With no preparation and almost no context into the situation, The Primeagen talks to Matt for about 20 minutes before narrowing the conversation down to two questions Matt is unable to answer:
- How long have these conversations been going on with WP Engine? Can Matt prove that the full scope of this conversation- including the “scorched earth nuclear” option he ultimately took- really stretches back eighteen months before WCUS?
- What are the guidelines he’s using to judge whether a company is in trademark violation and thus responsible for mandatory contributions to the project?
I genuinely believe that if Matt had started the entire kerfuffle with these two questions, laid out a clear path for the community going forward, and set forth a strong argument for why WP Engine should meet him halfway, I think the majority of the community would’ve been much more on board with decisions this past week. But it’s very clear that Matt doesn’t see the community as needing to have complete answers or the full picture. He’s not really interested in community support on this one.
There was one other question, though, that Matt did answer: looking back, would he have handled this differently? Sorry, no spoilers here.
There needs to be a balance, and we’re looking at two companies on the complete opposite sides of this spectrum. One the one hand, a company ruthlessly prioritizing profit above all else by focusing on a clear niche, heavy marketing, and heavier cost cutting, with minimal regard for the open source tools its customers use. On the other, a company holding everything together but whose products are seemingly never marketed, not widely popular in the community, with so much money left on the table that’s its almost alarming. Neither is good. Neither feels sustainable for the future of WordPress.
So while I agree WP Engine should act a lot more like Automattic and help out in core, maybe Automattic should act just a tiny bit more like WP Engine and focus on making software that people are willing to pay for.
- https://twitter.com/photomatt/status/1839266415762878939 ↩︎
- In an interview, Matt said that an 8% license fee was fair because WP Engine doesn’t need an R&D budget as “all they do is sell WordPress and WooCommerce to people”. This is just… disingenuous. They sell hosting, infrastructure, security, tooling, and so much more around WordPress. Also, disabling the News widget isn’t “hacking,” it’s a core filter many of us use. And so on. ↩︎