Daniel sent us this one, and it's a good one. He's been thinking about how standard-setting bodies work — the DSM, industry working groups, any committee charged with updating a big shared framework — and he's pointing out something true: from the outside, these processes look like they exist mostly to generate paperwork nobody reads, buffering egos rather than producing anything useful. But he's not buying the cynical take. His question is, what's actually happening inside these processes? And what ideas have been floated to make them feel less opaque and bureaucratic from the outside, even if the slowness is a feature, not a bug?
This is the kind of question where the answer is genuinely more interesting than the complaint. Because the complaint writes itself — committees are slow, bureaucrats love reports, everyone's protecting turf. But the inside story of how a working group actually produces a standard is almost never told, and when you look at it, it's surprisingly rigorous.
Let's be honest, the DSM is the perfect case study here, because nothing says "committee-designed" quite like a diagnostic manual that took fourteen years between editions and involved over a hundred and fifty people.
Actually more than that. The DSM-5 revision involved over four hundred clinicians and researchers across thirteen work groups, and the whole process ran from roughly nineteen ninety-nine to two thousand thirteen. So fourteen years is if anything an undercount — the formal work groups kicked off in two thousand seven, but the planning conferences started years earlier.
Four hundred people. That's not a committee, that's a small town.
That's exactly the point. When you have four hundred experts from different subfields, different countries, different theoretical orientations, you can't just get them on a Zoom call and hash it out. You need infrastructure. You need a process that ensures every voice gets heard, every objection gets documented, and every decision has a paper trail.
Let's unpack that infrastructure, because I think this is where the outside view completely breaks down. People imagine a working group as twelve people in a conference room arguing about commas. But the actual machinery is more like... what, a legal proceeding crossed with a scientific peer review?
That's not a bad description. The DSM process, for instance, has this formal structure where each work group is chaired by two people — one psychiatrist and one psychologist usually — and they're responsible for reviewing the entire literature in their domain. We're talking thousands of papers per group. They have to produce systematic reviews, meta-analyses where possible, and field trial results. The American Psychiatric Association requires that every proposed change be accompanied by a written justification that cites the evidence base. It's not just people sitting around sharing opinions.
That's the inside. The question is about the outside perception. And I think the gap between those two things — the rigorous internal process and the "this is all bureaucratic nonsense" external view — that gap is partly a communications failure and partly a structural inevitability.
Let me push on that. What do you mean by structural inevitability?
I mean that any process designed to be thorough and evidence-based will, by its nature, produce outputs that look bureaucratic. The documentation that proves you were careful is the same documentation that makes your eyes glaze over. You can't have the rigor without the paperwork, and you can't have the paperwork without looking like you're just generating paperwork.
That's actually a really important point. It connects to something I was reading about how standard-setting organizations think about this problem. The International Organization for Standardization — ISO — has explicitly grappled with the perception that their process is impenetrable. They have over twenty thousand standards now, and the average person interacts with dozens of them daily without knowing it. The ISO six hundred forty-four standard for freight containers, the ISO two hundred sixteen standard for paper sizes, the ISO three thousand one hundred standard for risk management. These things shape the physical world, but the process that produces them is almost invisible unless you're inside it.
When people do glimpse it, they recoil. Because it looks like a factory for producing documents nobody asked for.
But here's the thing — the documents are the product. A standard is literally a document. So when you say "they're just producing paperwork," it's like saying a bakery is just producing bread. That's the output.
Okay, but the bakery analogy cuts both ways. If the bakery's process involves seventeen committees to approve the shape of each loaf and a five-year review cycle for the recipe, someone's going to ask whether all that overhead is actually improving the bread.
And that's where the tension lives. Let me give you a concrete example of how this plays out in practice. When the DSM-5 was being developed, one of the most controversial proposals was to eliminate the bereavement exclusion for major depressive disorder. Under DSM-IV, if someone had just lost a loved one, you weren't supposed to diagnose depression for two months, even if they met all the criteria, because grief was considered a normal response. The DSM-5 work group proposed removing that exclusion.
This became a huge fight, right?
And here's why it's relevant to the question about process. The work group didn't just wake up one day and decide to change this. They reviewed something like fifty to sixty studies on bereavement and depression. They held multiple rounds of public comment. They published their rationale in multiple venues. They brought in outside consultants. And still, when the change was announced, the public perception was that a small group of psychiatrists had just decided to pathologize grief over lunch.
Because the process wasn't legible from the outside. You had all this rigor happening behind closed doors — well, not closed, but effectively invisible — and the only thing that broke through was the headline: "Psychiatrists Say Grief Is Now a Mental Illness.
And this gets to the core challenge. The process was designed to be thorough and evidence-based, and it was. But it wasn't designed to be legible to journalists, patients, or the general public. And those are two different design goals that sometimes conflict.
Let's talk about the ideas people have explored for bridging that gap. Because the prompt isn't just asking us to diagnose the problem — it's asking what's been tried.
There are a few categories of solutions that keep coming up in the literature on this. One is radical transparency — put everything online in real time. Meeting minutes, draft documents, dissenting opinions, the whole thing. The idea is that if people can watch the process unfold, they'll understand why it takes so long and why certain decisions get made.
Has anyone actually tried this?
The Internet Engineering Task Force — the IETF — is probably the best example. They've been doing open standards development since the eighties. Their mantra is "rough consensus and running code." All their working group discussions happen on public mailing lists. Anyone can join, anyone can comment, and the entire decision-making history is archived and searchable. It's about as transparent as a standard-setting process can get.
Does it work? Does the public actually understand what the IETF does?
Most people have no idea what the IETF is. But within the technical community, the process is highly regarded and considered legitimate. So transparency works for building trust with the people who are paying attention. It doesn't necessarily solve the problem of public perception more broadly.
Transparency helps with legitimacy among stakeholders, but it doesn't automatically make the process feel less bureaucratic to outsiders. If anything, watching the sausage get made might confirm people's worst suspicions.
There's a second approach that's more about narrative — essentially, embedding the process in a story that people can follow. The World Health Organization tried something like this with the ICD-11 revision. They created an open web platform where anyone could submit proposals for changes, and they published regular updates on what was being considered and why. The idea was to make the revision feel like an ongoing conversation rather than a black box that occasionally spits out a new edition.
Did it work?
The platform got something like ten thousand proposals from health professionals in over a hundred countries, which is impressive. But the actual decision-making still happened in committees that most people couldn't follow in detail. So you had this veneer of openness on top of a process that was still fundamentally closed. And when the final product came out, it still felt like it had arrived by fiat to anyone who hadn't been tracking the platform closely.
That's the thing about transparency, right? It only works if people actually look. And most people don't have the time or the expertise to follow a multi-year standards process. So you're transparent to an audience of basically nobody.
The stakeholders who care deeply — clinicians, researchers, industry players — they do follow. And for them, transparency is valuable. But the prompt is asking about the broader public perception, and I think you're right that transparency alone doesn't solve that.
There's a third approach I want to throw out, which is what I'd call "process compression" — designing the governance so that it's still rigorous but has fewer visible layers. The idea being: if you can't make the layers interesting, at least have fewer of them.
This is where things get tricky, because the layers exist for a reason. Let me give you an example from the DSM world. After the DSM-5 came out in two thousand thirteen, the APA realized that the fourteen-year revision cycle was too slow. Science was moving faster than the manual. So they switched to a living document model — DSM-5-TR in two thousand twenty-two, and ongoing incremental revisions rather than waiting for a DSM-6.
Which sounds great in theory. Continuous improvement, no more decade-plus gaps between editions.
But the challenge is that the governance structure for incremental changes is actually more complex, not less. You need a standing committee that can evaluate proposals on an ongoing basis. You need criteria for what counts as enough evidence to justify a change between editions. You need a process for public comment that's faster than the old multi-year cycle but still meaningful. The APA created a steering committee and a series of review committees, each with specific domains, and they meet regularly to evaluate new evidence. It's not fewer layers — it's just layers that operate on a faster cadence.
The attempt to reduce bureaucracy actually produced more bureaucracy.
It produced more infrastructure. And whether that infrastructure feels like bureaucracy depends on whether you think the work it's doing is necessary. The APA would argue that having a standing committee to review suicide risk assessment tools, for example, is not bureaucracy — it's due diligence. But from the outside, it looks like another layer of meetings and memos.
This is the paradox at the heart of the question. The things that make a process rigorous — documentation, review, multiple rounds of input, formal decision-making procedures — are the same things that make it look bureaucratic. You can't separate the substance from the appearance.
You can try. And there are some interesting experiments in that direction. One approach is what I'd call "decision-centric communication." Instead of publishing the full process, you publish clear, accessible explanations of specific decisions at the moment they're made. The UK's National Institute for Health and Care Excellence — NICE — does this reasonably well. When they issue a new guideline, they publish not just the guideline itself but a lay summary that explains what changed, why it changed, and what evidence supported the change.
Instead of trying to make the whole process transparent, you make the outputs explainable.
And I think this is actually the most promising direction, because it acknowledges that most people don't want to follow a multi-year process. They want to understand the decisions that affect them. So you invest your communication energy at the decision points, not in documenting every working group call.
Even that has limits, right? Because some decisions are technical and can't be boiled down to a two-paragraph summary without losing what made them rigorous in the first place.
And this is where I think standard-setting bodies need to get more comfortable with saying "this is complicated, and here's where you can go deeper if you want." The lay summary is an on-ramp, not a replacement for the full documentation. But too often, organizations either publish nothing for the public or they publish a press release that oversimplifies to the point of distortion.
Let's talk about a specific reform proposal that's been floating around in the DSM context. There's been discussion about creating what amounts to a "living systematic review" infrastructure — essentially, a continuously updated evidence base that feeds into the diagnostic criteria. The idea being that instead of a committee deciding every five or ten years what the evidence says, you have an ongoing process of evidence synthesis that anyone can look at.
This is actually being explored in a few domains. The Cochrane Collaboration has been doing living systematic reviews for a while now — they update reviews as new evidence comes in rather than waiting for a fixed review cycle. And there's a group called the Hierarchical Taxonomy of Psychopathology — HiTOP — that's trying to build an evidence-based alternative to the DSM's categorical system using a similar approach. Their whole model is that the classification should emerge from the data continuously, not be decided by committee periodically.
How's that going?
It's fascinating work, but it hasn't replaced the DSM, and I don't think it will anytime soon. The challenge is that clinical utility and research validity don't always align. A data-driven dimensional model might be more scientifically accurate, but if clinicians can't use it to make treatment decisions, it won't get adopted. And that's where the committee process actually adds value — it balances scientific evidence with practical considerations.
The committees aren't just a necessary evil. They're doing work that a purely algorithmic approach couldn't do.
A committee can say, "The evidence supports this dimensional model, but for clinical purposes we're keeping the categorical threshold because it maps to treatment guidelines and insurance reimbursement." That's a judgment call that requires human deliberation. It's not something you can automate.
Which brings us back to the perception problem. The very thing that makes committees valuable — their ability to weigh multiple types of evidence and make judgment calls — is also what makes them look political and arbitrary from the outside.
And I think the honest answer is that you can't fully solve this. You can manage it, you can communicate better, you can be more transparent, but you can't eliminate the tension between deliberative rigor and public legibility. They're in tension by design.
Let's pivot slightly and talk about what other industries have tried. The prompt mentions "any organization charged with updating an industry-wide standard," and I think there are lessons from outside clinical psychology that are worth looking at.
The world of technical standards is actually full of interesting governance experiments. The W3C — the World Wide Web Consortium — has a fascinating process for web standards. They use a formal maturity ladder: working draft, candidate recommendation, proposed recommendation, and finally W3C recommendation. Each stage has specific exit criteria, and the whole thing is designed to ensure that standards are implementable before they're finalized.
Is this process legible to outsiders?
To web developers, yes. To the average person, no. But I think the W3C would argue that's fine — their primary audience is implementers, not the general public. And that's an important distinction. Not every standard-setting process needs to be legible to everyone. The question is whether it's legible to the people who need to use it and trust it.
That's a fair point, but the DSM is different. The DSM affects millions of people who receive diagnoses based on its criteria. It's not just a technical standard for clinicians — it shapes how people understand their own mental health. So the public has a legitimate stake in the process, and the perception problem matters in a way it might not for, say, a standard for bolt thread specifications.
And I think the APA has recognized this, which is why the DSM-5 process included a public comment period and why they've made efforts to engage patient advocacy groups. But the engagement still tends to happen at the input stage — "tell us what you think" — rather than at the deliberation stage, where the actual decisions get made.
That's the crux of it, isn't it? Public comment is easy. Letting the public watch you weigh conflicting evidence and make trade-offs is hard, both logistically and politically.
It's hard because deliberation requires candor, and candor is risky when everything you say might end up on Twitter. If I'm on a DSM work group and I say, "I'm not sure the evidence for this change is strong enough, but I'm worried about the political backlash if we don't make it," that's a legitimate deliberative comment. But if it gets quoted out of context in a news article, it looks like I'm admitting the process is political.
Which it is, to some degree. Science doesn't happen in a vacuum, and standard-setting involves value judgments. The question is whether you can acknowledge that openly without undermining the credibility of the whole enterprise.
This is where I think the most interesting reform proposals live. There's a school of thought that says standard-setting bodies should be more explicit about the values they're weighing, not less. Instead of pretending the process is purely scientific, acknowledge that it involves trade-offs between sensitivity and specificity, between research validity and clinical utility, between continuity and innovation. Make the value judgments explicit so they can be debated openly rather than buried in technical language.
That's the "admit it's political to make it less political" argument.
And there's some evidence it works. The National Academy of Medicine has been moving in this direction with some of its consensus studies. They now include explicit "values statements" that lay out the principles guiding the committee's decisions. It doesn't eliminate controversy, but it makes the basis for decisions more transparent.
Let me throw a different example at you, because I think it illustrates something important. The Montreal Protocol — the international treaty that phased out ozone-depleting substances — is widely considered one of the most successful standard-setting processes in history. It's not a clinical standard, but it's a global agreement that required buy-in from governments, industries, and scientists.
What's interesting about the Montreal Protocol is that it was designed to be adaptive. The original agreement had built-in mechanisms for updating the phase-out schedules as new scientific evidence came in. The technical panels would meet regularly, review the latest data, and recommend adjustments. It wasn't a one-and-done negotiation — it was an ongoing process that could respond to new information.
Did the public understand this process?
I'd say the public understood the outcome — "we're saving the ozone layer" — without understanding the mechanism. But the mechanism itself was trusted because it was tied to visible results. The ozone hole stopped growing and started shrinking. That's a kind of legibility that doesn't require understanding working group procedures.
Results build trust, and trust makes the process feel less bureaucratic.
Yes, but that's a slow solution. You have to deliver results over decades before the trust accumulates. And for something like the DSM, where "results" are harder to measure than parts-per-billion of atmospheric chlorine, the trust-building timeline is even longer.
Let's talk about one more reform idea that I've seen floated, which is what I'd call "process narrators." The idea is to embed journalists or science writers within the standard-setting process, not as members of the committee, but as observers who produce accessible running commentary on what's happening and why.
I love this idea in theory. In practice, it runs into confidentiality problems immediately. A lot of committee deliberation is sensitive — not because anyone's doing anything wrong, but because preliminary discussions can be easily misinterpreted. If a working group member says, "Let's explore eliminating this diagnosis and see where the evidence leads," and that gets reported as "DSM Committee Considering Eliminating Depression," you've created a firestorm over something that was just an exploratory question.
You'd need some kind of embargo system, where the narrator has access but agrees not to report on preliminary discussions, only on decisions once they're firm enough to be explained.
Which brings us right back to the decision-centric communication model we talked about earlier. At some point, all these reform ideas converge on the same insight: you can't make the deliberation itself fully public without damaging the quality of the deliberation. So you focus on making the outputs explainable, and you build trust through consistent, honest communication over time.
I want to push back on that slightly, because I think there's a middle ground that's underexplored. What if you made the deliberation itself more structured in a way that made it more legible? I'm thinking of something like the Federal Register in the US — when an agency proposes a new regulation, they have to publish the proposed rule, the evidence supporting it, and a response to every significant comment they received. It's still bureaucratic, but the structure makes it possible to trace how a decision was made.
The DSM process actually does include elements of this. The APA published extensive documentation of the field trial results, the literature reviews, and the rationale for major changes. The problem is that this documentation runs to thousands of pages and is written in technical language. So the structure is there, but it's not accessible.
The question becomes: can you make structured deliberation accessible without dumbing it down?
I think you can, but it requires investment in what I'd call "translational infrastructure." You need people whose job is to take the technical documentation and produce versions that are accurate but readable. And most standard-setting bodies don't budget for this. They see it as a nice-to-have, not a core function.
Because the core function is producing the standard, not explaining it.
And until that changes — until communication is seen as part of the standard-setting process rather than an afterthought — the perception problem will persist.
Let's bring this back to the DSM specifically, because I think there's a unique dimension here that doesn't apply to, say, ISO standards for shipping containers. The DSM deals with mental illness, and mental illness is contested territory. There are genuine philosophical disagreements about what counts as a disorder, whether diagnostic labels help or harm, whether the whole enterprise of psychiatric classification is valid. So the process isn't just trying to build consensus among experts — it's operating in a domain where the very legitimacy of the enterprise is questioned.
That's a crucial point. The ISO committee for freight containers doesn't have to deal with activists arguing that the whole concept of a freight container is oppressive. The DSM committees do face that kind of fundamental critique. And that makes communication even harder, because you're not just explaining your process — you're defending your existence.
Which probably makes the committees more defensive and less transparent, because every internal disagreement becomes ammunition for critics who want to discredit the whole thing.
And you can see this dynamic playing out in real time with the debates over the DSM-5. The bereavement exclusion fight wasn't just about the evidence — it was a proxy for deeper disagreements about the medicalization of normal human experience. When the stakes are that high, committees tend to circle the wagons, and transparency suffers.
What's the bottom line here? If someone asked you, "What should a standard-setting body do tomorrow to make its process feel less bureaucratic and convoluted," what would you tell them?
I'd give them three things. First, invest in decision-centric communication. Don't try to make the whole process transparent — make the decisions explainable at the moment they're made. Publish lay summaries alongside technical documentation. Hire people who can translate between expert and public audiences.
Second, make your values explicit. Don't pretend the process is purely scientific when it involves trade-offs between competing goods. Say what you're optimizing for and why. People are more willing to accept decisions they disagree with if they understand the framework that produced them.
Third, design for adaptive iteration. The DSM's shift to a living document model is a step in the right direction, but it needs to be paired with communication infrastructure that makes the ongoing changes visible and understandable. If you're going to update criteria every year, you need to explain every year what changed and why. Otherwise you've just made the black box spin faster.
I'd add a fourth, which is: acknowledge the limits of legibility. Some things are complicated for good reasons, and no amount of communication wizardry is going to make a multi-year deliberative process feel like a breezy read. Being honest about that — saying "this is complex, here's the short version, here's where to find the long version" — is better than pretending you can make it simple.
That's well put. And I think it connects to something deeper about how we think about expertise and democracy. There's a tension between the idea that important decisions should be made by experts using rigorous processes, and the idea that those decisions should be understandable and contestable by the public. You can't fully resolve that tension. You can only manage it honestly.
Managing it honestly means, among other things, not pretending that a working group of four hundred people is ever going to feel like a town hall meeting. It's going to feel like a bureaucracy, because it is a bureaucracy, and some bureaucracies are actually doing important work.
The phrase "bureaucracy" has become a pejorative, but the original meaning was just "rule by officials" — and in a complex society, some things need to be ruled by officials who know what they're doing. The challenge is making sure those officials are accountable, transparent where possible, and honest about where transparency has to end.
That's the note I think we should end on. The process is bureaucratic because it has to be. The question isn't how to eliminate the bureaucracy — it's how to make it trustworthy.
Now: Hilbert's daily fun fact.
Hilbert: In the nineteen eighties, a team of Soviet zoologists working in the Aral Sea basin published a widely-cited paper proposing that bats navigated not by echolocation but by detecting minute fluctuations in the Earth's magnetic field through specialized hairs in their nostrils. The theory was abandoned in nineteen eighty-nine when a follow-up study revealed the "specialized hairs" were actually carpet fibers from the lab's flooring.
...right.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop. If you enjoyed this episode, leave us a review wherever you listen — it helps. We're at myweirdprompts.I'm Corn.
I'm Herman Poppleberry. See you next time.
