On June 17, 2026, the Society of Motion Picture and Television Engineers made an announcement that should not have been news. SMPTE, the 110-year-old organization that defines the technical standards behind everything from film reel dimensions to digital cinema projection, declared that its entire catalog of more than eight hundred standards documents would now be freely available to anyone, anywhere, without a paywall. The announcement was treated as a significant event. People who had paid hundreds of dollars for a single PDF celebrated. Developers who had pirated standards documents to do their jobs admitted it in public. And the top comment, from a well-known engineer, captured the reaction that most readers felt: “I don’t understand why any standards body would consider not doing this as a default.”

That comment is worth sitting with because it reveals an assumption that is common among software engineers but almost completely alien to the world of formal standards organizations. The IETF, which produces the standards that run the internet, has always made its documents freely available. You can read the HTTP specification, the TCP specification, the DNS specification, and every RFC ever published without paying a cent or creating an account. This was a deliberate choice made in the early days of the internet, and it is one of the most important decisions in the history of computing. The IETF understood that a standard nobody can read is not a standard. It is a proprietary specification with a marketing label.

SMPTE came from a different world. It was founded in 1916, when the motion picture industry was young enough that different studios used different film formats, different projection speeds, and different mechanical standards for everything from sprocket holes to reel canisters. A film that played perfectly in one theater would not fit the projector in the next town. SMPTE’s first standard, adopted in 1917, fixed the dimensions of 35mm film and set the frame rate at sixteen per second, later revised to twenty-four when sound film arrived in 1927. These were infrastructure decisions disguised as engineering decisions. By agreeing on a common format, SMPTE made it possible for any theater to play any film, which created a national distribution market, which created the movie industry as we know it. The standard was the platform, and the platform was free to use. The document that described it, however, was not.

For most of its history, SMPTE sold its standards documents the way a publisher sells books. You wrote a check, you waited six to eight weeks, and a paper copy arrived in the mail. When the internet arrived in the mid-nineties, SMPTE was already eighty years old, and the institutional muscle memory of selling documents was so deep that nobody seriously questioned it. The PDF replaced the paper, the website replaced the mail order form, but the transaction remained the same: you paid for access to the specification of how things work.

This model was not unique to SMPTE. It is the default in the standards world. ISO, the International Organization for Standardization, charges 133 Swiss francs for a single document. The C and C++ language standards cost approximately three hundred dollars each from ANSI, which is why virtually no programmer has ever read the actual C standard — they read the freely available final draft, which is identical in content but published before the paywall descends. IEEE, which manages the 802 family of standards covering WiFi, Ethernet, and Bluetooth, paywalls most of its specifications. Building codes in the United States — the electrical code, the plumbing code, the structural code — are all locked behind paid access. The National Electrical Code costs $170. If you are a homeowner who wants to know whether your DIY wiring job is legal, you pay $170 for the privilege of finding out what the law requires you to do. The law requires you to comply with a document you cannot read without paying the organization that wrote it. This arrangement is not an accident of history. It is the business model.

The justification for paywalling standards has always been that standards cost money to produce. Committees must meet. Drafts must circulate. Engineers must spend years arguing about edge cases in windowless conference rooms. Someone has to pay for the coffee and biscuits. The money has to come from somewhere, and if it does not come from selling the documents, it must come from membership dues, which would make the organization dependent on the very companies it is supposed to regulate neutrally. This argument has surface plausibility. It falls apart the moment you look at where the money actually goes.

The engineers who write the standards are volunteers or employees of member companies who donate their time. They get a room to sit in, power for their laptops, and the aforementioned coffee and biscuits. The money from document sales goes to administrative staff, building rent, taxes, and the salaries of people whose job is to manage the document-selling infrastructure. The system is a machine for converting volunteer engineering labor into administrative payroll, with the standards document serving as the tollbooth.

There is a darker explanation for the paywall, and it is not about funding at all. Standards bodies are often formed by oligarchies of incumbent industry players who have decided that mutual interoperability would benefit their businesses. They have no interest in lowering the barrier of entry for new competitors. A startup that wants to build a device compatible with a standard must pay for the right to read the specification, and must pay again for the right to certify compliance, and must pay legal fees to navigate the patent licensing regime that often accompanies the standard. Each of these tolls is individually modest — a few hundred dollars for a PDF is not going to stop a funded company. But collectively, and combined with the legal uncertainty around patent encumbrance, they create a moat. The moat is not high enough to stop a determined competitor, but it is high enough to stop an open source project, an independent developer, or a hobbyist who just wants to understand how the thing works.

When the only people who can read the standard are those willing to either pay or pirate, the standard stops functioning as a coordination mechanism and starts functioning as a filter. It filters out everyone who cannot or will not navigate the tollbooth, which means it filters out exactly the people whose participation would make the ecosystem healthier: newcomers, outsiders, and people building things the incumbents did not anticipate.

The damage goes beyond individual developers. When a standard is paywalled, people who cannot or will not pay do not simply give up. They reverse-engineer. They study sample files and captured network traffic and output from existing implementations, and they build something that looks like it works. These reverse-engineered implementations almost always contain subtle errors because the original specification contains edge cases that sample files do not exercise. When two such implementations attempt to interoperate, the errors compound. The standard becomes a suggestion rather than a specification, and the ecosystem fragments not because anyone wanted fragmentation but because the cost of reading the specification was higher than the cost of guessing.

This is not a theoretical concern for SMPTE standards. SMPTE timecode, the system that synchronizes audio and video in professional production environments, is one of the most widely implemented standards in the world. Every video editing program, every digital audio workstation, every broadcast switcher, and every professional camera implements SMPTE timecode. If those implementations diverge because the people who wrote them were working from secondhand summaries rather than the actual specification, the result is not a cosmetic bug. It is a frame of audio that does not match the corresponding frame of video, which in a professional context means a ruined take, a missed cue, or a broadcast that goes to air with the presenter’s lips moving half a second after their voice.

SMPTE’s announcement is not just an isolated act of generosity by one organization. It is part of a broader modernization effort that includes adopting GitHub for version control, transitioning to structured HTML-based authoring, and implementing an automated publishing pipeline that streamlines document creation and review. These are not just tooling changes. They represent a philosophical shift. GitHub-based workflows mean the standards development process becomes visible. HTML-based authoring means the standards become machine-readable in ways that a PDF locked behind a login wall never was. Automated publishing means the friction between a committee decision and the world being able to read it drops to near zero. A 110-year-old organization is adopting the tools and culture of open source software development, and if that sounds remarkable, it should. The open source movement spent three decades proving that transparent, collaborative, publicly visible development produces better technical output than closed-door committee work. Standards bodies are finally noticing.

The question that follows naturally from SMPTE’s decision is who goes next. IEEE’s 802 standards are the foundation of modern networking. 802.3 defines Ethernet. 802.11 defines WiFi. 802.15.4 is the basis for Zigbee and Thread, the low-power wireless protocols that run smart home devices. These are not industry-specific standards like SMPTE timecode. They are the operating system of the physical internet. Every Ethernet port on every device you own implements 802.3. Every WiFi chip implements 802.11. These specifications should be as freely available as the TCP specification, because they perform the same function at a different layer of the stack. The fact that they are not is a legacy of institutional inertia that has persisted long past any credible justification.

The building code situation is worse because it involves the law. When a government adopts a building code as regulation, it is making compliance with that code legally mandatory. In a functioning democracy, citizens should be able to read the laws they are required to follow without paying a private organization for the privilege. The fact that you must pay the National Fire Protection Association $170 to read the electrical code that your local government has adopted as law is a quiet scandal that has persisted for decades because the people affected by it — homeowners, small contractors, DIY renovators — are not an organized political constituency. They do not have a lobbyist. They just pay the $170, or more likely, they do the wiring without reading the code and hope the inspector does not notice.

There are signs of movement beyond SMPTE. The European Union is considering reforms to the standardization system, motivated in part by the recognition that standards referenced in EU legislation must be freely accessible for the legislation to have democratic legitimacy. The covid pandemic created a precedent when several standards organizations temporarily made their medical device standards freely available to accelerate the production of ventilators and protective equipment. The fact that these standards could be made free overnight when there was political will to do so revealed something important: the paywall was never a technical or financial necessity. It was a choice, and choices can be unmade.

The IETF model — free access to all standards, funded by meeting fees and corporate sponsorship rather than document sales — has now been validated by thirty years of the internet’s existence. The internet did not collapse because anyone could read the RFCs. It grew faster than any technical infrastructure in human history. The SMPTE model, slowly being abandoned, represented the opposite theory: that standards are intellectual property to be monetized, that restricting access maintains quality, and that the people who need the standards can afford to pay for them. That theory has now been tested for over a century and the result is in. The paywall did not make the standards better. It made them less widely implemented, less accurately implemented, and less relevant to the people who build things on top of them.

SMPTE’s decision matters not because SMPTE standards are the most important standards in the world — they are important to a specific industry, but they are not Ethernet or WiFi or the C language. It matters because SMPTE is old enough, established enough, and respectable enough that its decision to abandon the paywall model deprives every other standards organization of the excuse that open access is a naive idea from internet kids who do not understand how real industries work. SMPTE is a hundred and ten years old. It defined the film formats that shot Citizen Kane and the synchronization protocols that broadcast the moon landing. When an organization with that history decides that selling PDFs is a worse strategy than giving them away, the rest of the standards world loses its cover.

A standard is a public good. Its value increases with the number of people who implement it correctly, and every person who cannot read it is a person who might implement it incorrectly or not at all. The cost of producing a standard is fixed and sunk once the document exists. The marginal cost of distributing one more copy is zero. This is the economic definition of a public good, and public goods should be publicly available. The fact that it took a hundred and ten years for a standards organization to rediscover this principle is not a sign of progress. It is a sign of how far we drifted from the obvious.