{"blyg":"0.4","id":"2j6p37emqxxe89emxz6vknnqxh","kind":"thread","origin":"https://blyg.pioneeringspirit.xyz/","page":"t/2j6p37emqxxe89emxz6vknnqxh/","author":{"name":"PAtwater","url":"https://blyg.pioneeringspirit.xyz"},"created":"2026-11-17T17:37:42Z","updated":"2026-11-17T17:50:56Z","version":1,"content_md":"# More on permitting \n\nEzra Klein has a big piece on permitting in the NYT that’s making the rounds. It’s a lot of analysis and rhetoric regarding intra-democratic party ideology and factional posturing. But the part I found both funny and sad was the opening. Over the 2000s, there’s been so many bills that were The Biggest X since half a century or whatever. The Stimulus. The IRA. The Trump tax cut. Many were in impact and also really analysis prior to implementation much smaller. \n\n> When I say it this way, it sounds like such an easy call: The Bipartisan American Affordability and Jobs Act of 2026 is the largest reimagining of the federal permitting process in a generation. It would make it far easier to build new energy infrastructure of all kinds: wind and solar farms and interstate transmission lines and, yes, natural gas pipelines. It would make it easier to build much else, too: rail and transit infrastructure and affordable housing and broadband internet.\n\n> The bill represents Democrats and Republicans deciding that the fight over what gets built should play out in elections and markets rather than in the bowels of administrative agencies and courts. It should be driven by what people want, in public, rather than what they can block, often in private. I do not expect, in every case, that that would lead to outcomes I like. But leaving the system unchanged is a guarantee of calamity. The status quo has left us completely unable to build the energy system we need.\n\nThe bill focuses as I understand on creating strict timelines and increasing the capacity of federal agencies to process permits. It also creates a process for developing new data standards and digitizing permitting. Those tools strike me as kludges in the last era of the modernity machine. Ezra kinda gets this as he acknowledges that a 400+ page bill will not do necessarily what the analysis-industrial complex thinks it will do. \n\nBut more deeply the bill doesn’t really change the fundamental approach to permitting in the US. There will still be the patchwork quilt of state and local permits to get through. My hunch is that if this passes the DC crowd will slowly realize that it wasn’t enough. The sense making machine took around a decade and a half to realize that the infrastructure provisions in the stimulus and then the IRA were not doing what they intended. There is much more sludge slowing things down beyond the formal policy.\n\n> Federal permitting laws were adopted in the late 1960s and early 1970s, a period of intense cultural change that ushered in a new era of skepticism about big government projects. That skepticism spurred wide-reaching changes in judicial doctrine, bureaucratic norms, and substantive laws. Permitting reform may be necessary to spur more infrastructure development, but it is unlikely to be sufficient without sustained attention to the dense network of laws and practices that have built up around that cultural change. https://nicholasbagley.com/assets/papers/costs-of-federal-permitting.pdf\n[[18e7j8r86s1t6n1zvehbamn3r2 | “Very good sentences: Federal permitting laws were adopted in…”]]\n\nChanging that of course is a start but as the digital revolution blurs the line between url and irl it’s worth asking. What is a permit really? What are the existing information flows? The friction points? The coordination costs? What beyond the political economy of interests vested in development and a status quo are involved? There is a seismic shift in the operating mechanics of government underway… bureaucracies were created as information processing machines! More to come. \n\n### Longer thoughts on permitting in the context of the dusk of the modernity machine \n\n![[34n81narp9mcs1znydn9wdr0bd]]\n\n","content_html":"<h1>More on permitting</h1>\n<p>Ezra Klein has a big piece on permitting in the NYT that’s making the rounds. It’s a lot of analysis and rhetoric regarding intra-democratic party ideology and factional posturing. But the part I found both funny and sad was the opening. Over the 2000s, there’s been so many bills that were The Biggest X since half a century or whatever. The Stimulus. The IRA. The Trump tax cut. Many were in impact and also really analysis prior to implementation much smaller.</p>\n<blockquote>\n<p>When I say it this way, it sounds like such an easy call: The Bipartisan American Affordability and Jobs Act of 2026 is the largest reimagining of the federal permitting process in a generation. It would make it far easier to build new energy infrastructure of all kinds: wind and solar farms and interstate transmission lines and, yes, natural gas pipelines. It would make it easier to build much else, too: rail and transit infrastructure and affordable housing and broadband internet.</p>\n</blockquote>\n<blockquote>\n<p>The bill represents Democrats and Republicans deciding that the fight over what gets built should play out in elections and markets rather than in the bowels of administrative agencies and courts. It should be driven by what people want, in public, rather than what they can block, often in private. I do not expect, in every case, that that would lead to outcomes I like. But leaving the system unchanged is a guarantee of calamity. The status quo has left us completely unable to build the energy system we need.</p>\n</blockquote>\n<p>The bill focuses as I understand on creating strict timelines and increasing the capacity of federal agencies to process permits. It also creates a process for developing new data standards and digitizing permitting. Those tools strike me as kludges in the last era of the modernity machine. Ezra kinda gets this as he acknowledges that a 400+ page bill will not do necessarily what the analysis-industrial complex thinks it will do.</p>\n<p>But more deeply the bill doesn’t really change the fundamental approach to permitting in the US. There will still be the patchwork quilt of state and local permits to get through. My hunch is that if this passes the DC crowd will slowly realize that it wasn’t enough. The sense making machine took around a decade and a half to realize that the infrastructure provisions in the stimulus and then the IRA were not doing what they intended. There is much more sludge slowing things down beyond the formal policy.</p>\n<blockquote>\n<p>Federal permitting laws were adopted in the late 1960s and early 1970s, a period of intense cultural change that ushered in a new era of skepticism about big government projects. That skepticism spurred wide-reaching changes in judicial doctrine, bureaucratic norms, and substantive laws. Permitting reform may be necessary to spur more infrastructure development, but it is unlikely to be sufficient without sustained attention to the dense network of laws and practices that have built up around that cultural change. <a href=\"https://nicholasbagley.com/assets/papers/costs-of-federal-permitting.pdf\">https://nicholasbagley.com/assets/papers/costs-of-federal-permitting.pdf</a>\n<a href=\"https://blyg.pioneeringspirit.xyz/f/18e7j8r86s1t6n1zvehbamn3r2/\">“Very good sentences: Federal permitting laws were adopted in…”</a></p>\n</blockquote>\n<p>Changing that of course is a start but as the digital revolution blurs the line between url and irl it’s worth asking. What is a permit really? What are the existing information flows? The friction points? The coordination costs? What beyond the political economy of interests vested in development and a status quo are involved? There is a seismic shift in the operating mechanics of government underway… bureaucracies were created as information processing machines! More to come.</p>\n<h3>Longer thoughts on permitting in the context of the dusk of the modernity machine</h3>\n\n<blockquote class=\"blyg-transclusion\" data-blyg-id=\"34n81narp9mcs1znydn9wdr0bd\" data-blyg-version=\"1\">\n<h1>The Lost Leverage of Legibility</h1>\n<h2>Legibility theater, permits and potholes in the dusk of the modernity machine</h2>\n<p>Recently I filled a page of notes on what it takes to do routine infrastructure work in Southern California, the kind of job that replaces an underground asset under a busy street. When the work crosses a couple of cities and a Caltrans right of way, it starts long before anyone touches a wrench. Someone has to research every jurisdiction involved, call 811 and DigAlert, prepare drawings and draft a preliminary traffic control plan that puts an inspector from each party on site. Neighbors and city councils have to be notified, and community outreach has to happen before the permits are even requested. Working around the clock means a noise variance, and a noise variance can mean knocking on doors alongside the LAPD.</p>\n<p>Then the drawings go through a couple of rounds of review. When the comments from different agencies conflict, someone calls a meeting to reconcile them. Sometimes the fix is as simple as moving a sign a few feet so that it falls into a different jurisdiction. Restoration requirements differ within a single city and by council district. Staff turn over in the permitting departments while all of this is underway, and the equipment has its own clock.</p>\n<p>None of this is secret, and almost all of it is written down somewhere. But how to actually build increasingly resembles digging into the Restricted Section of the Hogwarts library, involving spelunking into a mixture of folk wisdom, rules of thumb and practical experience about the coordination involved to dig a common piece of infrastructure. There's a banality to how strange everything has gotten.</p>\n<hr>\n<p>A century ago, legibility was a lever. In the early days of the New York Bureau of Municipal Research, reformers counted the potholes the city had actually fixed rather than trusting the reports of the patronage machine. That simple count mattered because it sat outside the chain it measured, and anyone could check it. A machine that claimed to be paving the streets could be confronted with a ledger showing that it was not. The Bureaus, plural, went on to help design the architecture of modern public administration.</p>\n<p>Venkatesh Rao's World Machines project offers a useful frame for why that count worked so well. In <a href=\"https://contraptions.venkateshrao.com/p/the-modernity-machine-iii\">The Modernity Machine III</a>, which Rao notes was drafted with ChatGPT from his first two installments and a group brainstorm, the argument is that the modernity machine optimized not for truth or progress but for legibility, the ability to render &quot;people, land, goods, time, belief, and violence&quot; countable and interoperable at scale. Venice with its double-entry bookkeeping was an early legibility engine. The modern state was its chief operator. For roughly four centuries, making the world countable was the surest way to change it, because the count was wired to a lever: a budget, a permit, a ballot or a front-page story.</p>\n<p>The same essay explains why that wiring frays. Once the machine is complete, more law increases rigidity without legitimacy, and more planning amplifies unintended consequences. The machine does not collapse. It saturates, gets patched in increasingly fragile ways and slowly hands the work of coordination to something else.</p>\n<hr>\n<p>I've started calling what that saturation looks like from the inside <strong>legibility theater</strong>. The apparatus of readability keeps running, often at enormous expense, but it has come loose from any downstream action. Reports are filed to show that reports were filed.</p>\n<p>Roads are the most vivid case I know. Around a single pothole there is now a cacophony of measures. Cities score their streets on the Pavement Condition Index, a 0 to 100 rating that the U.S. Army Corps of Engineers developed and that ASTM has since revised again and again. The standard itself is modest about what it is. It describes the index as the collective judgement of pavement maintenance engineers and an indirect measurement of structural integrity, not a replacement for measuring ride quality or structural capacity. Yet the number has been reified and threaded into funding formulas, and an entire consultant industrial complex exists to produce it.</p>\n<p>The pavement score is only the beginning. Under SB 1, California's Road Repair and Accountability Act, every city and county must submit an adopted project list and then an annual project expenditure report to the California Transportation Commission to stay eligible for road maintenance funds. The program is called accountability, and the paperwork is real, yet the Commission reviews those project lists and expenditure reports only for completeness. Around those reports sit contractor performance measures for the crews doing the patching, deliverables from the consultants doing the engineering design, DigAlert tickets and excavation permits, 311 complaints and now sensor data from Waymo.</p>\n<p>Some of these streams are better than others. A 311 complaint measures who complains, and Waymo's cameras resolve the street in finer detail than a windshield survey ever could. But a better-resolved picture of a crack still does not tell anyone which repair held, which contractor did good work or whether the money bought smoother streets. In my own public records requests to cities around the country on how they track and manage potholes, the divergence in data formats and practices was striking (<a href=\"https://blyg.pioneeringspirit.xyz/t/5qdnzp1mfv2pt25gtkwwv4tbap/\">Protocolizing Potholes</a>). For a long time I thought that was a data problem. I now think it is a protocol problem, particularly in terms of how layers and layers of process obscures feedback loop. Nothing closes the circuit from &quot;we measured&quot; to &quot;we fixed the right street, and it held.&quot;</p>\n<p>I wrote earlier this year that when public works slow down, the feedback loop between idea and structure stretches out over years, and ritual creeps into the gap (<a href=\"https://pioneeringspirit.xyz/real-engineers-deliver-results\">Real engineers deliver results</a>). Legibility theater is that ritual at institutional scale. Premodern states were blind because the world went uncounted. Late-modern states are blind because the world is counted constantly, for upstream readers, with no loop back to the people who act.</p>\n<hr>\n<p>Why does the loop break? I see three reasons, and they compound.</p>\n<p>The first is noise without feedback. Each reporting regime accretes because it serves a particular upstream reader, whether that is a funder, an auditor or a legislature. Each one makes sense on its own. Together they create a fog in which the operator, the person who actually has to decide which street to fix next, gets less usable signal than the Bureau clerk had with a notebook.</p>\n<p>The second is a broken actuator. Even accurate information needs someone positioned to act on it, and that position has eroded. Jurisdictions fragment until moving a sign ten feet changes who has authority. Mandates get enforced by shame, as in a recent federal permitting bill whose main remedy for agencies that ignore new data standards is a public list of the non-compliant (<a href=\"https://blyg.pioneeringspirit.xyz/t/4gt40pfw716gqvs9yvzh74fqaw/\">data standards thread</a>). Staff capacity thins out. When the Government Accountability Office looked at geothermal permitting on federal land, it found some Bureau of Land Management offices accepting email while others wanted wet signatures, handbooks in draft since at least 2022 and one Nevada office 60 percent vacant (<a href=\"https://blyg.pioneeringspirit.xyz/t/7qgzb0c5wdr21q60m9nxx6113c/\">Rewilding Permitting Protocols</a>). On a typical infrastructure job, there is usually someone who is ostensibly responsible for integrating everything, the project manager. But the project manager holds the whole picture in their head precisely because no system holds it for them.</p>\n<p>The third is incentives. Motivation matters, but beneath it sits a harder fact. The reporting chain often works perfectly well for its actual customers. A funder can accept a report that demands no response because it proves the grant was administered without creating accountability for outcomes. A consultant can produce a pavement score that nobody acts on and still get paid for the next one. Over time the measure gets captured at the point of collection by the people being measured. The Pavement Condition Index is what becomes of the independent pothole count once the count has been absorbed into the chain it was meant to check. Organizational sociologists have a name for the gap between what institutions say and what they do, which Nils Brunsson called organized hypocrisy. Legibility theater is its data-age cousin.</p>\n<p>Rao has a vivid term for this condition. In <a href=\"https://contraptions.venkateshrao.com/p/the-divergence-machine\">The Divergence Machine</a> he describes late modernity as the zombie persistence of the modernity machine, high in energy and low in intelligence, and he observes that almost everything around us is late modern. That matches my experience of public works. The machine still moves an enormous amount of money and paper. It just no longer learns.</p>\n<hr>\n<p>The usual response is reform. Pass a bill that sets a deadline, add an ombudsman, require one more report on whether the previous reports were filed. I have come to think that most of these small and moderate permitting reforms consume a huge amount of political will for modest impact, because they are aimed at the documented protocol rather than the one that actually governs the work.</p>\n<p>I would rather we rewild before we reform. Rewilding means finding out what is really happening before deciding what to fix. A protocol that lives in habit, like whether a field office takes email, can be changed with a handbook, a tracker and a decision about who is allowed to say no. A protocol that lives in a political settlement needs a different kind of work and a lot more time. We need reorientation, an oblique angle on a problem that has defeated a generation of frontal assaults.</p>\n<p>Those new oblique perspectives needs institutions of its own. One model I keep returning to is the Arc Institute in biomedical research, which gives its <a href=\"https://arcinstitute.org/model\">core investigators</a> fully funded, renewable eight-year appointments and supports <a href=\"https://arcinstitute.org/about\">innovation investigators</a> at partner universities who keep their posts while serving as bridges. Imagine something similar for state capacity: a network of applied researchers embedded in specific places, funded for long enough to trace how a pipe actually gets replaced or how a street actually gets paved, and free to publish what they find. A researcher could replay ten or twenty past jobs to see which rules drive the most delay, which is close to the approach I am excited to collaborate on with a mechanism design researcher at NYU CUSP for cross-jurisdiction permitting. That is the opposite of the high-modernist attack vector, a single grand reform launched from the capitol. It looks more like the Bureaus of Municipal Research, plural, updated for an age of cheap analysis.</p>\n<p>The strongest reason for optimism is that we may not need to rebuild the old machine to do this. Rao argues that the divergence machine, which has been running as the world's dominant logic for about twenty-five years, relies on what the modernity machine already built without trying to perpetuate it or destroy it. Common protocols are exactly that kind of inheritance. If cities shared a few simple protocols for how they record street work, AI-accelerated analysis could answer common questions rigorously across many places at once. Are we getting good value for money on street repair? Which treatments hold up, and under which contractors? Those questions do not require a single canonical metric handed down from Sacramento.</p>\n<p>There is precedent for this kind of correction. In 2013 an electronics engineer named Mats Järlström worked through the formula traffic engineers had used to time yellow lights since 1959 and found that it ignored drivers slowing down to turn. Oregon's engineering board fined him for practicing engineering without a license. He won in federal court, and the Institute of Transportation Engineers later adopted his corrected formula. That is legibility working the way it should, with open math, outside critique and a standard that actually updates.</p>\n<p>I don't think the answer is a better single number to replace the Pavement Condition Index. The public problem is fixing the potholes. What matters is rough consensus and running code applied to streets: many people looking at the same shared records, arguing about whether the work happened and whether it held, and acting on what they find. Accountability can come from that diversity of dialogue rather than from a high-modernist metric that grows more byzantine and opaque each time it gets embedded in another funding formula.</p>\n<hr>\n<p>In Rao's account of the modernity machine's birth, Don Quixote marks the end of the medieval world, a knight who &quot;behaves impeccably within a dead symbolic system, and chaos results.&quot; I keep thinking we already have our Quixote. He is not a man on a horse. He is a project list, adopted at a regular meeting of the council, submitted on time to Sacramento and reviewed for completeness. He is a pavement score that rises and falls with great ceremony while the pothole on my street stays exactly where it is. Every ritual is performed correctly, and the street never gets fixed.</p>\n<p>The Bureau clerk didn't need a better ritual. He went out, counted and asked whether the work had actually happened. That question was the lever all along, and it still is. The tools for asking it now fit in a pocket, ride on the roof of a robotaxi and run on protocols any city could adopt, if we can summon the collective courage.</p>\n<p>The Bureau clerk with a notebook had leverage because the count was independent and the public could read it. We will not get that leverage back by building a bigger notebook with the oft floated dream of better performance metrics dashboards. We might get it back by building many small ones, in many places, connected by protocols simple enough that the people doing the work can see their impact and course correct as needed.</p>\n<hr>\n<h3>A note on AI assistance</h3>\n<p>This piece was drafted with Claude. It worked from my blyg posts, my Bot Garden notes on permitting and an outline we developed over a long evening of back and forth, which included a round of critique from vgr_zirp, the reconstructed Ribbonfarm voice I like to spar with. Claude also read Venkatesh Rao's World Machines essays and looked up the details on the Pavement Condition Index and SB 1 reporting. The arguments and the stubbornness about potholes are mine.</p>\n<hr>\n<p><em>This is partly why I ran for my local water board. A board seat is a strange place to study legibility theater, since so much of a public meeting is theater in the literal sense. But it is also one node in the kind of network I'm describing, a place where the reports arrive and someone can ask what they actually tell us.</em></p>\n<p><em>Suggested tags: legibility, permitting, protocols, world machines</em></p>\n\n</blockquote>\n","content_hash":"sha256:e716e631beb695cca497731285f238a3851b04b79d89f1bcdb697980c132e338","media":[],"transclusions":[{"id":"34n81narp9mcs1znydn9wdr0bd","version":1}],"stub_of":{"origin":"https://blyg.pioneeringspirit.xyz/","id":"34n81narp9mcs1znydn9wdr0bd","version":1,"cited":{"source":"Pioneering Spirit (Blyg)","excerpt":"The Lost Leverage of Legibility Legibility theater, permits and potholes in the…","url":"https://blyg.pioneeringspirit.xyz/t/34n81narp9mcs1znydn9wdr0bd/","retrieved":"2026-11-17T17:50:56Z"}},"changelog":[{"version":1,"at":"2026-11-17T17:50:56Z","note":null}]}