 I know how engineers get excited about standards compliance, and I'm worried about how they'll respond to this script. It's making me kind of antsy. After years of drought, New Year's Eve brought a record-breaking downpour to the San Francisco area. The city was immersed in 140 millimeters of water in a single day, resulting not just in the flooding of homes and businesses, but cascading failures of other important civic systems. Transit, highway and road closures, electrical grid faults, emergency shelters packed to capacity with evacuees. It wasn't exactly Katrina, but it wasn't pretty. Catastrophes draw our attention to the many interlocking, complicated, and sometimes fragile technologies that make it possible for us to have a normal day. The San Francisco sewer system is, happily, out of sight and out of mind for most of the city's residents most of the time. But watching coursing rivers of rainwater pooling into streets, we start to wonder about things like, how big are those underground pipes? What sort of downpour is the wastewater system designed to handle? What other systems are built assuming it's continued operation, and do they have contingency plans for when those pipes get overloaded? The word infrastructure is a good catch-all term for these sorts of essential civic services, but its definition isn't extraordinarily precise. Infrastructure is the basic physical and organizational structures and facilities needed for the operation of a society or enterprise. Yes, that describes sewers, roads, and power plants as you'd expect. But considered more broadly, there are other things that fit that definition that you might not think of off the top of your head. Like there are radio waves that are essential to the operation of San Francisco. A solar flare that briefly interrupted cellular service would result in a fair amount of chaos and a breakdown of other facilities, as many essential repair and maintenance technicians are dispatched via cell phone. Examples and algorithms, like air traffic control procedures, can cause incredible damage and disorder if they break down. Even something as simple as the cultural preference for driving on one side of the road versus the other is, technically speaking, an organizational structure necessary for the continued operation of society. None of this is to say that the dictionary definition requires us to pedantically include these things when we're talking about infrastructure, but in the context of a disaster or catastrophe where we're forced to take a hard look at all the stuff that has to keep working for society to keep working, it can be helpful to conceive of infrastructure more broadly than is the power still on. Standards and testing form an important part of that extended infrastructure, the maintenance schedule of your sewer system, the chemical treatment specification for your tap water, and the legally mandated width of your roads are just as necessary as the physical structures themselves. Some of those standards are set by government bodies, some are set by professional organizations, some are set by standards agencies, organizations that are dedicated to creating rules to provide a common technical basis for various global industries. The International Organization for Standardization, or ISO, the letters are iOS, they're ISO, don't worry about it, it's very silly, is the international authority that defines the official protocols and specifications for an incredible diversity of stuff, from JPEG compression, ISO 18181, to freight shipping containers, ISO 668, to paper sizes, ISO 216, and tens of thousands of other indispensable technologies at every level of your society. ISO crafts these standards to ensure that wherever you are in the world, any new implementations of that tech satisfy certain requirements, things like safety, functionality, and reliability, that are compatible with other existing tech, so your multi-million dollar public works, work with your other works. But ISO is a private organization, non-profit, but private, they charge fees for access to the vast majority of their standards, sometimes up to thousands of dollars, and many previously free standards have been locked behind their store's paywall. Nations that don't want to or can't afford to pay ISO membership fees are no less subject to those rules, at least if they want to sell their goods to nations that require standards compliance, which means their economy, their industry, and their future is kind of at the mercy of whoever's writing them. ISO's membership manual explicitly states, if you are not a full and active member of ISO, you are letting others decide how standards are set and under what conditions your country participates in international trade. The paywall model also causes problems for businesses. Many laws require companies and their products to conform to ISO standards, but do not supply access to those standards. This can lead to the bizarre situation where your company might be violating the law, but you can't be totally sure unless you pay ISO to see what legal requirements you might not be satisfying. Of course, these documents can be incredibly useful for coordinating industries, and take a great deal of time and effort to produce and validate. One might argue that requiring people to pay for access helps to support the efforts of standardization, to get even more of the world on the same page. There are three problems with this argument. First, although some of the money goes towards keeping ISO running and paying their staff, the technical experts who actually write the standards do so on an unpaid volunteer basis. Someone might spend years exhaustively researching and thinking through all the necessary specifications for their field of expertise for the good of all, only to turn around and have to hand ISO a few hundred bucks for a licensed copy of the completed version. A number of previous contributors to ISO's library have started a petition to remove the paywall, asserting that it's in both the public's and the organization's best interests to dispense with the paid model. Second, many standards, even the legally mandatory kind, are of questionable value. ISO 9000 is by far the organization's number one seller, a set of management techniques ostensibly designed to ensure and improve quality. ISO 9000's certification is a long and expensive process that has to be re-upped every three years by an approved accreditation consultancy. But if you want to participate in certain EU trade agreements, it's requisite. So what's in this very expensive document that's so important that millions of companies worldwide pay for both the standard and regular audits? Documentation of, and adherence to, a decision making process. If you think that sounds underwhelming, you're in good company. Many critics have accused ISO 9000 of being a sort of cool kids club for companies that can afford it rather than a meaningful mechanism for industry coordination or quality assurance. There's no compelling evidence that adhering to the standard has any appreciable effect on a company's performance or the quality of its products, but the roster of certified companies grows every year. What choice do they have if they want to do business? Third, because ISO standards are written by people, there are certain inevitable cultural, political, and societal values that are bound up in their authorship, and because they get embedded in infrastructure all over the world, sometimes those values can be reproduced in harmful ways, on a scale that can hamper, undermine, or even rot entire societies. In Confessions of a Recovering Engineer, civil engineer Chuck Morone notes how the specifications for road construction in the U.S., a set of metrics and guidelines ostensibly justified on grounds of safety, actually result in a great deal of unnecessary danger for Americans who have to use those roads. The standards implicitly prioritize the safe speed that vehicles can travel and mandate certain features on roads to maximize that value, large shoulders, wide lanes, unobtrusive pedestrian crossings, time traffic lights, and so on. That sounds reasonable, Ray. The committee that originally composed the standards wanted drivers to get where they're going quickly and safely, and built their civil engineering Bible around that principle. Who could argue with that? But if you start to interrogate those requirements and what they actually do, they start to look less like a virtuous mandate and more like a set of expensive trade-offs. High vehicle speeds are considerably less safe for pedestrians and cyclists who have to share those roads with drivers. Prioritizing unobstructed flow of automotive traffic makes walking or cycling very inconvenient as a means of getting from one place to another. Worse yet, there's evidence that building roads to minimize distractions and interruptions from going fast in a straight line simply results in less attentive, faster driving, which is actually more dangerous for motorists. There was a steep spike in vehicular fatalities during COVID restrictions in the US, despite there being fewer cars on the road. Barone suggests that without traffic congestion to get drivers to slow down and pay attention to their surroundings, the designs optimize for faster car traffic coaxed otherwise safe drivers into a false sense of security at high speeds with predictable results. Also, traffic lights increase maximum speed but reduce average speed and freeway metering doesn't work. Please read the book so I'm not the only one ranting about this to my friends and family. Barone's observations highlight how even standards that are built to achieve commendable goals must carry an implicit set of values and A DNA replicated over and over in every new bit of construction, along with all its shortcomings and unintended consequences. Nobody would argue that getting drivers where they're going safely is a bad thing, but the ways you define and implement that objective say a lot about what you care about and what you don't care about quite so much. ISO standards aren't exempt from those implicit value judgments. Although ISO 9000 has some allowances for employee feedback, so long as there's some well documented internal process that's rigorously followed and checked, there's nothing in the requirements that forbid a business from treating its employees like garbage. ISO 14001 is a standard for environmental management, helping organizations identify, manage, monitor, and control their environmental issues in a holistic manner. I'm sure you'll be relieved to know that ExxonMobil is ISO 14001 certified. To combat the problem of civil engineering specifications that either don't address or actively undermine the needs of the people affected by them, Morone recommends that civil engineers take a less active role in prescribing what sorts of civic improvements should be built in communities, behaving more as facilitators of the community's stated goals than dictators of the one true path. Rather than looking at a city plan, consulting the tables in the civil engineer's handbook and saying, you need bigger shoulders on your roads, Morone suggests actively listening to the problems the individuals in the community face and addressing those problems directly. If someone complains that they have to cross a busy road every day to get to the bus stop, figure out a way to manage traffic to give them better access. If bikers feel unsafe at a particular intersection, build them a bike path away from the road, or make the intersection friendlier for bikes. These sorts of fixes don't just alleviate the minor concerns of the individuals who need to use this sort of infrastructure every day. They create new channels for access and movement, contextualized by the actual values and needs of the people who live there, making the whole town a nicer place to live. Obviously there will be conflicts of those values that need to be resolved, but Morone argues that the starting point for policy should be the explicit needs of the users, not some arbitrary, abstract, a priori good as dictated by civil engineers who have never seen the city they are improving. From this perspective, the ISO requirements that govern our cities and infrastructure have some important limitations. Although ISO makes some allowance for public commentary channeled through national representatives, the universality of their application kind of prohibits any sort of consideration for the specific circumstances in which they appear and function, let alone the diverse values of all the people they might conceivably affect. That's not to say that ISO requirements are a bad thing by any stretch of the imagination. I'm an engineer for God's sake. More than most people, I'm sensitive to the fact that in order for anything to work with anything else, there has to be some external pressure for uniformity. Without organizations like ISO, DIN, ANSI, and so on for prescribing and throwing their weight behind standards, many important technologies would be impossible. But there's something troubling about the way impartial paywalled ISO requirements are baked into the most fundamental operations of our towns and cities, with their unstated implicit values carried along for the ride. The power of standards to compel certain outcomes isn't always subtle. Corporate interests have put their thumbs on the scale more and more deliberately in recent years. Strong arming ISO committees to carve out advantages for their business by flooding them with partisan volunteers. The tactic former ISO governor, Martin Bryan, has called standardization by corporation. Microsoft recently engaged in this practice to fast track the approval of ISO 29500, the standard linked to Microsoft's proprietary open XML format, forcing the work of other technical committees working on things like nuclear waste disposal and textile manufacturing onto the back burner, to get open XML codified as quickly as possible. After all, once it's part of the standard, the industry will be pressured to migrate to open XML, and pay Microsoft to use it. In his resignation letter, Bryan suggested that standardization by corporation is becoming impossible to combat, and it's anyone's guess how long ISO will be able to keep those efforts at bay. As climate change promises to pound infrastructure around the world with worsening weather-fueled catastrophes, it's worth reflecting on the role that standards agencies will play in dictating the protocols and requirements of that infrastructure. Whether the unstated values and strategic omissions are fostering positive outcomes for the people who are most affected by them, or if they're just synchronizing industrial activities in a way that prioritizes conformity, conflict aversion, and control over welfare. Standards do a lot of good stuff. I'm glad that there were passionate nerds dedicated to fostering interoperability, safety, and some sort of baseline functionality for the stuff I use every day, from cameras to cars. But if their work remains inaccessible for anyone who doesn't want to or can't pay for it, if it's only representative of the values of the authors and whoever can wield influence over them, I don't think it's doing as much good as it could. Do you think ISO standards should be made freely available to anyone who wants to use them? Do you think pressure for universal design principles should be balanced against the specific needs of the people who live under them? Please leave a comment below and let me know what you think. Thank you very much for watching. Don't forget to blow up a subscribe while I share, and don't stop thunkin'.