Matt Mullenweg has published an extensive and self-critical assessment of the WordPress project’s state, questioning a process culture that — in his view — paralyzes decision-making, overshadows individual contributors in favor of companies, and has turned Five for the Future into a program generating useless data.
Remember that you can listen to this program from Pocket Casts, Spotify, and Apple Podcasts or subscribe to the feed directly.
Program transcript
Hello, I’m Alicia Ireland, and you’re listening to WPpodcast, bringing the weekly news from the WordPress Community.
In this episode, you’ll find the information from April 13 to 19, 2026.
At WordCamp Asia in Mumbai, a seemingly routine question during the closing Q&A triggered the most intense internal debate the WordPress community has seen in a long time. An attendee asked the panel what the best way was for a company to contribute to the project. Two of core’s most active committers, Peter Wilson and Sergey Biryukov, gave the same answer: sponsor full-time contributors. Matt Mullenweg, following the event remotely, publicly disagreed in real time, transmitting his response through the Automattic communications team. What followed was a week of criticism and self-criticism that put long-unanswered questions back on the table.
The first thread started on Make Core, where Matt published a reflection on a photo of a conference badge from the event that he saw circulating on social media. It showed the label “SELF EMPLOYED” in large letters. That led him to question when the project had started putting company names above any personal contributor data, and whether that shift in emphasis was healthy. His concrete proposal: badges should display the contributor’s WordPress.org username, location, and website — not their employer.
But it was in the core-committers Slack channel where the criticism became more extensive and more pointed. The trigger was a Trac ticket related to the WordPress 7.0 Connectors screen, created by an Automattic employee after a private conversation with a colleague, and merged during the Release Candidate phase without public discussion. An outside contributor had flagged it and sent a notification to Mullenweg. When he reviewed it after returning from WordCamp, he described it as a microcosm of every problem the project carries.
Matt’s diagnosis was direct: WordPress is not being beaten by the competition — it is hurting itself. Nineteen years of sustained growth in the face of critics predicting its failure, and now, according to him, the project has spent years undoing everything that made it successful. He pointed to process overload as the primary cause: participation rules guaranteeing public discussion, broad consensus, and globally accessible meeting schedules — all reasonable in isolation — have together created a culture that makes it functionally impossible to resolve minor issues without weeks of Slack threads and dozens of participants. The accumulation of over 8,000 open Trac tickets, which the project had chosen to hide behind a custom query rather than address, struck him as a perfect example of that dynamic.
He was also critical of the state of WordPress.org, pointing to pages with oversized headers and sparse content, an About page that still doesn’t reflect the switch from MySQL to MariaDB, and inconsistent navigation across sections. And he criticized the AI Connectors screen he himself had pushed for in WordPress 7.0: it took too long to arrive, and when it did, it showed error screens. He compared the time invested to the pace at which Cloudflare had built its own CMS, EmDash, in two months.
In parallel, in the five-for-the-future Slack channel, Matt was equally direct about the program he launched in 2014 to have companies dedicate five percent of their capacity to contributing to the project. The problem is not the concept but the execution: the program was designed to incentivize commitments that are never verified and take no account of whether the resulting activity is aligned with the project’s goals. The data it generates are, in his own words, “worse than useless as we’ve structured the program.”
He identified four specific problems: corporate presence has overshadowed volunteers, students, and contributors working in their own time; pledges are treated as an end in themselves rather than a first step; there is no tracking of actual activity after a pledge is made; and Make teams have lost sight of their purpose in favor of intermediate goals and process dynamics.
Responses within the project were varied. Anne McCarthy of Automattic acknowledged that the pendulum had swung too far toward corporate recognition and opened a proposal to design a standard badge template all WordCamps could adopt. She also supported the idea of reviving the Lead Developer role, whose disappearance she believes has left accountability and decision-making gaps that surface every release cycle. Amber Hinds pointed out that measuring raw hours structurally favors large companies, and proposed measuring the proportion of available hours dedicated to the project — an individual contributing half their time is giving proportionally far more than a company whose hundreds of hours represent only five percent of its total capacity.
Courtney Robertson raised a question that has been circulating since 2024: who actually owns the project’s infrastructure. WordPress.org belongs personally to Mullenweg, and Robertson asked him to state that clearly and in writing so contributors understand that working on that infrastructure means contributing to the project, not enriching a private individual.
Outside Slack, Matt’s diagnosis found considerable agreement in the broader community, though the way he expressed it did not. Several ecosystem figures noted that the tone and style of the intervention fits a familiar pattern: long periods of absence from a topic followed by an extensive, loaded intervention that makes constructive response difficult. The term “Matt Bomb,” coined more than a decade ago, started circulating again.
Mullenweg closed his Tuesday posts with a reflection on Joost de Valk, co-founder of Yoast, with whom he has had a public conflict since 2024 and who was banned from WordPress.org after publicly calling for an end to his leadership as “benevolent dictator for life.” He wrote that he wished he had supported de Valk’s attempts to drive changes in the project more when he had the chance. De Valk, who was following the conversation in the PostStatus community Slack, responded with a shrug emoji.
One of the most recurring problems at WordCamp Contributor Days is that attendees spend the entire session trying to set up their local environment and never get around to actually contributing.
To address this, the team has published the WordPress Core Dev Environment Toolkit, a desktop application available for macOS, Windows, and Linux that sets up a complete WordPress Core development environment with no prerequisites. The process comes down to installing the app, choosing a directory, clicking, and having a cloned wordpress-develop repository ready with a development server running and the ability to make changes and generate a patch ready to attach in Trac.
The tool is designed specifically for the Contributor Day context, where the goal is for someone to go from attendee to first patch in a single afternoon. Organizers can share the download link with participants before the event so they can install it at home on a good connection, as the download is somewhat large.
Directly in line with the criticism he launched the previous week about the project’s excess of manual processes, Matt Mullenweg posted on Make Meta an open call: any contributor is invited to propose WordPress.org processes that could be automated.
The community’s responses clearly map where the most obvious bottlenecks are. Plugin review is the most frequently mentioned: with over 500 weekly submissions and first-review wait times exceeding two weeks, some propose letting AI do the first pass while human reviewers focus on edge cases, appeals, and security reports.
Other proposals cover community event approvals, where the process can also stretch for weeks; automated first responses in support forums; Trac ticket triage, including detecting duplicates, requesting missing information, and archiving inactive tickets; automatic generation of weekly team summaries from Slack and GitHub logs; and fully automated theme directory review with no human intervention.
Mullenweg’s response was brief but clear: “get ready, we’re going to have a lot of Wapuus running around.”
The Community team has announced the global sponsors for the events program covering the period from Q2 2026 through Q2 2027. Three organizations are involved: Automattic, with its Jetpack and WordPress.com brands, and Hostinger as Global Leaders, and Woo as Regional Powerhouse.
Their support covers the operational costs that make WordCamps and Meetups possible throughout the year: venue rental, catering, audiovisual equipment, licenses for more than 685 active groups worldwide, and event insurance. Organizers of 2026 events must include these partners’ logos on their websites and can contact each company to confirm which specific brand will represent them at their event.
And finally, this podcast is distributed under a Creative Commons license as a derivative version of the podcast in Spanish; you can find all the links for more information, and the podcast in other languages, at WPpodcast .org.
Thanks for listening, and until the next episode!
Leave a Reply