<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Yung Codes on Medium]]></title>
        <description><![CDATA[Stories by Yung Codes on Medium]]></description>
        <link>https://medium.com/@yungcodes?source=rss-3c36767d8f45------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*Q6KKUAFAxwSVtT7AdIHRpg.jpeg</url>
            <title>Stories by Yung Codes on Medium</title>
            <link>https://medium.com/@yungcodes?source=rss-3c36767d8f45------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 00:51:52 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@yungcodes/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Crazy About Methane]]></title>
            <link>https://medium.com/@yungcodes/crazy-about-methane-c132e1d9da46?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/c132e1d9da46</guid>
            <category><![CDATA[waste-management]]></category>
            <category><![CDATA[renewable-energy]]></category>
            <category><![CDATA[climate-change]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Wed, 04 Nov 2026 21:08:24 GMT</pubDate>
            <atom:updated>2026-11-04T21:08:24.369Z</atom:updated>
            <content:encoded><![CDATA[<p>Alright, so my new job is about landfill and our favorite greenhouse gases: methane.</p><p>You know, CH4?</p><p>To find out how deadly is your trash and practical things society can do — read on below.</p><h3>Why is Methane a Baddie</h3><p>Once methane reaches the atmosphere, its average atmospheric lifetime is about a decade (~12 years). Reaction with hydroxyl radical (OH) ultimately turn methane into CO2 and water.</p><p>Despite its short life, in most greenhouse gases reporting protocol, the global warming potential (GWP) of methane is reported as GWP-100 (the trailing number 100 stands for 100 years span), dimmering its rapid climate worsening potential.</p><p>In hallmark places like New York City, even using GWP-20 for methane was recently repealed to “stay consistent” with global reporting. In short, when using the muted GWP-100 metrics, methane is about ~27 times worst than carbon dioxide whereas GWP-20 metrics indicate at least <strong>80 times worst</strong>. In most climate action plans and GHG reporting, you’ll see the 27 number more often.</p><p>What this consequentially leads to is that:</p><ul><li>Natural gas just look cleaner on paper</li><li>Methane gets discounted by about 2/3</li><li>More “progress” are made on paper</li><li>Near term warming is muted to doesn’t matter</li></ul><p>Research shows that every extra fraction of a degree brings more heat waves, droughts, coral reef die-off, and a higher risk of hard-to-reverse changes like melting ice sheets. Cutting methane is the fastest way to slow warming over the next few decades. Methane traps a lot of heat but only lasts about a decade in the air, so reducing it pays off quickly. Impacts you and I can see in our life time.</p><p>If you need to remember anything, just know this — methane is a very important greenhouse gas with a <strong>substantial (80x) </strong>equal-mass pulse of carbon dioxide. We are running out of time. We can see the climate change impacts when we reduce methane in our lifetime.</p><h3>Renewable Natural Gas for Good</h3><p>Based on a <a href="https://www.energy.ca.gov/sites/default/files/2023-08/CEC-200-2023-010.pdf">California Energy Commission consultant report on Renewable Natural Gas</a> (RNG), Californians use around ~12,500 MM therms in 2020 and forecasted 2030 usage to be 13,274 MM therms. RNG potentially have the production of 1000~6000 MM therms (10% to almost 50%), with sources from solid wastes, agriculture (livestock waste), wastewater treatment plant, or other).</p><p>My primary interest highlights that RNG from landfills have the potential to displace 6 to 15% of the non-power-plant natural gas consumption in California.</p><p>I know my decarbonization friends may think natural gas is going away, but it’s not going to happen very soon. CPUC estimates that across the 3 major IOUs had about $27 billion of gas infrastructure investment that had not been yet recovered from customers through depreciation.</p><p>Yeah, natural gas distribution networks aren’t going to disappear overnight.</p><p>Plus, with reliability is a foundation to climate resiliency, it’s not enough to be efficient and renewable. We need energy resources that can be distributed like a battery.</p><p>A giant battery to lower the duck curve peak demand? Guess what? Landfill gas <em>kind of</em> works like one.</p><p>The current Long-Term Gas Planning Rulemaking (R.24–09–012) states CPUC is examining both gas-system decarbonization and <em>opportunities</em> for decommissioning gas infrastructure as gas demand declines. <a href="https://www.cpuc.ca.gov/news-and-updates/all-news/cpuc-sets-biomethane-targets-for-utilities">SB1440 authorized CPUC</a> to adopt biomethane procurement targets or goals for gas utilities to displace some fossil fuel natural gas.</p><p>In the similar spirit, CARB’s 2022 Scoping Plan explicitly retains a role for low-carbon gaseous fuels. Its modeled pathway includes<em> biomethane blended into pipelines in the 2030s </em>and renewable hydrogen blended into the fossil-gas system, while dedicated hydrogen pipelines serve some industrial clusters. CARB describes renewable natural gas and hydrogen as options during the transition away from fossil fuels.</p><h3><strong>Let’s Talk Affordability</strong></h3><p>Some ways of cutting emissions actually save money, while others cost more up front.</p><p>Fixing gas leaks and capturing gas from landfills is relatively cheap. The saved gas can even be sold, and households don’t have to change anything. Making homes more efficient, like better insulation, also lowers energy bills. The expensive part is replacing gas furnaces, cars and power plants with cleaner options, which means big upfront costs that often show up on top of increasing utility bills while you’re already trying to use less.</p><p>Carbon pricing programs also raise fuel prices, though they can give money back to households. The political problem is timing. The costs hit your bill now, while the benefits come later and are harder to see: fewer wildfires and floods, lower insurance premiums, cleaner air and less illness.</p><p>Waiting has its own costs too, because climate damage and volatile fossil fuel prices already raise what people pay. The smartest approach usually starts with the cheap, fast fixes like methane leaks, protects lower-income families during the switch, and sets realistic timelines so costs don’t spike all at once.</p><p>That’s the urgency behind maximizing renewable natural gas, but ask anyone on the street this or even an energy specialist at a bar, no one would even know how much their own community’s landfills could have been saving them money.</p><p>It doesn’t have to be all or nothing to decarbonize.</p><p>We enjoy the reliability of gas system when our electric grid goes down. That mean many Californians and I can boil water for my family during a catastrophe. That is survival for some families.</p><p>Look, I’m not defending the fossil fuel industry. The fact is California has one of the most extensive natural gas distribution networks in the United States, representing a significant investment in existing energy infrastructure.</p><h3>So Where Are We At, California?</h3><p>Data analyst at heart, so I sought to understand how much methane are we getting from landfill and why are <strong>most landfills in California</strong> burning off methane rather than to capture it?</p><p>Based on Cal Recycle SWIS Database, there are about 378 landfills. Around 139 are active landfills — meaning waste is still being buried and while the other are closed (still generating landfill gas, but no longer accepting new waste). Half of the waste management operators are private companies like Republic Services, Edco, and etc.</p><p>Out of all the landfills in California, about 78 shows up in the<strong> voluntary</strong> EPA’s LMOP Database with more in-depth data. These are the ones that have some kind of energy recovery system.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*H0Nmjue1R2qCgjcfYHol6g.png" /><figcaption>Source: Yours truly and <a href="https://view.officeapps.live.com/op/view.aspx?src=https%3A%2F%2Fwww.epa.gov%2Fsystem%2Ffiles%2Fdocuments%2F2024-09%2Fopprjslmopdata-detailed.xlsx&amp;wdOrigin=BROWSELINK">EPA LMOP Database </a>(pulled Sep 2026)</figcaption></figure><p>That graph is showing most landfills in California still burns off methane than capturing them for energy recovery. The biogas market is likely struggling against strict air compliance standards and overcoming first cost capital. However, if you look at the stock prices for many of the waste companies in the past 10 years, they’re going up.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/884/1*gDoo_pulMRtVpq6qwNRQtA.png" /><figcaption>ChatGPT comparison of Total Shareholder Return between the major waste management companies and S&amp;P500 benchmark.</figcaption></figure><p>Investing in trash management a decade ago put you at a better place compared to the S&amp;P500 — imagine that!</p><p>From the <a href="https://www.cpuc.ca.gov/industries-and-topics/natural-gas/renewable-gas">CPUC Renewable Gas proceeding</a>, the agency proposed to reduce the procurement target in roughly half due to cost containment concern, environmental integrity, and realistic procurement.</p><p>In other words, things are still moving slow, even on the demand side.</p><p>Public funding for a public agency to establish LFG to Energy is barely there, if at all. So, it’s a niche industry run by lawyers and bureaucracy — not technologists. There are some challenges with navigating California’s <strong>complex network of environmental regulation compliance</strong> because landfills gas all starts out as — frankly— <strong>dirty</strong>. Not to mention the NIMBY-ism is still alive and strong in this sunshine state. So, currently, maybe we are at the mercy of private equity and lawyers.</p><p>Perhaps it should stay that way.</p><p>Don’t come at me; I’m still forming my opinion on that because there can be more financial structure and incentives already in place to overcome the high capital cost investment and compliance barriers.</p><p>Except the private equity and lawyers would probably point fingers at the regulators. I can see both sides. It’s no small task to hold hands and move forward with a consensus.</p><h3>What’s Next?</h3><p>The planners out there know about SB1383 and how California aims toward zero waste by diverting all organics to compost. After all, plastics doesn’t really produce methane gas (unless you have a special plastic-fast-eating bacteria). How does that that affect renewable natural gas production?</p><p>Good point — all the policies are aiming toward a circular economy and zero waste ideally. In theory, eventually, our landfills will just be composed of un-recyclable plastics, and my job won’t exist by then. As long as there are cellulose (think paper, wood), there will be natural bacteria that eats them and produce methane.</p><p>Heck, if the whole country goes vegan and there’s no animal farms to produce agricultural waste — I’ll be happy to be out of a job and transfer my skills elsewhere.</p><p>Meanwhile, there’s still plenty of trash around. It’s not just a dump and forget anymore. We know our resources are finite, so why not make use of the stuff we have and buy some time for our children?</p><p>I’m still on my learning journey in the world of solid waste. This was the first step to understand the broad overview of current policy trends and what-are-other-people-doing kind of thing. I felt like I barely scratched the surface.</p><p>Stay tune for the next episode.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c132e1d9da46" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Everyone wants the Easy Button. Who wants the recipe for it?]]></title>
            <link>https://medium.com/@yungcodes/everyone-wants-the-easy-button-who-wants-the-recipe-for-it-0ecab34147f2?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/0ecab34147f2</guid>
            <category><![CDATA[codes-and-standards]]></category>
            <category><![CDATA[building-code-compliance]]></category>
            <category><![CDATA[political-philosophy]]></category>
            <category><![CDATA[easy-button]]></category>
            <category><![CDATA[public-policy]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Wed, 15 Jul 2026 17:51:34 GMT</pubDate>
            <atom:updated>2026-07-15T17:51:55.044Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Tackling the invisible pain points of codes &amp; standards. A think-out-loud.</em></p><p><em>I debated against publishing this story that has been sitting in my draft for a month because part of it made me feel a little too personal.</em></p><p><em>Then, I remember, this is an ode to my future self. A picture may be worth a thousand words, but it can’t tell all the stories in my head at this point in time.</em></p><p>In the last month, I crawled out of my cozy hole and spent more time with my peers.</p><p>That’s just the things about codes and standards. People — like users, inspectors, designers, contractors, and plan checkers— uphold these rules. They don’t magically happen in the forms buried in a portal nor behind computer screens.</p><p>So into the world I went.</p><h3>What Everyone Wants</h3><p>SimBuild, BayREN quarterly forum, Electrification Summit, and a small gathering of codes &amp; standards powerhouse — all these engagements were very inspiring in their own ways and made the work feel a little more real. I got a taste of the local regional perspectives, statewide perspectives, and even a small taste of national perspectives — all in a few short weeks.</p><p>With AI being more prevalent in these discussions, the desire is still the same: <strong>everyone wants an easy button</strong>.</p><figure><img alt="Picture of an easy button." src="https://cdn-images-1.medium.com/max/1024/1*mPuK-OT67zPn5wN8k_cM-w.jpeg" /><figcaption>Source: Shopping on the Amazon, Staples Easy Button</figcaption></figure><p>This was the premise of the “app” era and now the “Large Language Model (LLM)” era.</p><p>We need lower barriers and less frictions to quickly deploy software tools to enable building hardware (i.e. buildings) <strong>faster</strong>, <strong>more affordably</strong>, and <strong>without compensating for quality</strong>.</p><p>That sounds nice.</p><p>However, in the real world, you can only pick two if you’re lucky.</p><h3>Why “Easy Button” for Policy is Hard</h3><p>The higher you go in policy, I feel like problems around conflicting goals and objectives do make things harder to focus. It’s something I heard a few months back from the Third-Party Implementer Conference or something like that.</p><p>That’s one of the unavoidable side effects in unity and <strong>finding optimization</strong>. You have to do trade-offs. The cost of not doing something is immaterial and invisible. To add on to the complexity, cost changes and compounds with time.</p><p>If you optimize for certain cost, <strong>it will be like a moving target over time</strong>. This, lady and gentleman, is the basis and crux of energy efficiency policies. This means adoption is justified when resources constraints raise demand, but it also means it’s subject to politics.</p><p>Whoever said easy buttons come with efficiency made a foolish assumption. In public policy, easy buttons often require an <strong>extraordinary amount of effort </strong>to create.</p><p>California has a strong history of energy efficiency work and policy. The pipeline is fairly established in a way. Policies are developed upon the success of pilot projects and the emerging techs within a vertical integration pathway.</p><h3>Pause: What is “vertical integration”?</h3><p>The vertical integration pathway starts with: emerging technology (proof of concept) to program implementation (mini-scale up for early adopters) then codified in standards (apply to every body).</p><figure><img alt="Infographic showing California’s energy efficiency innovation pipeline in four stages: (1) Research &amp; Emerging Technologies, (2) Demonstration &amp; Market Development, (3) Codes &amp; Standards Development, and (4) Building Energy Standards. Each stage includes representative activities and California examples, illustrating how new technologies progress from research and pilot programs to adoption into Title 24 and Title 20, delivering long-term energy savings and compliance benefits." src="https://cdn-images-1.medium.com/max/1024/1*FkItHNSbdBrZF2UYhxJP9w.png" /><figcaption>In this case, a picture does worth a thousand words. Source: me + my assistant</figcaption></figure><p>The reoccurring problem OR (if you’re an optimist) also the most opportunities for easy buttons happen at <strong>the jump to scalability. </strong>That’s where I’ve been marinating my thoughts in.</p><p>Making a one-size-fit-all solution work across the most diverse State ranked 4th in the world’s economy is no easy task. I felt like answers weren’t as obvious and intuitive. I found myself questioning my own’s lacking in understanding and having a deeper depth to convince me to really believe in that future.</p><p>So, here’s my thinking exercise of the whole thing. Unfiltered. Unedited (kind of).</p><p>The key questions are: what characteristics can make up easy buttons and how to make them work happen in this kind of complex California settings.</p><h3>The Recipe Behind Easy Buttons</h3><ol><li><strong><em>The General Will and the Institutions</em></strong></li></ol><p>That’s where it all really starts.<strong> A thought.</strong></p><p>Wouldn’t it be easier if…</p><p>Wouldn’t it be better if…</p><p>Wouldn’t be faster if…</p><p>Wouldn’t it make more sense if…</p><p>Ideas, on paper, are cheap. Ideas built on consensus are expensive.</p><p>In my Vietnamese culture, we have a saying.</p><p>“Một người mười ý, mười người trăm ý.” This translates to — one person, ten ideas; ten people, a hundred ideas.</p><p>The difficulty isn’t finding solutions. It’s that when you have hundreds of stakeholders, you suddenly have <strong>multiple hundreds (more likely thousands) of perspectives</strong>, incentives, fears, and definitions of success.</p><p>Imagine all that noises.</p><p>Try to meditate with the TV on and ten people yapping in the back.</p><p>One of the first challenges are to find <strong>the solutions that are worth scaling collectively</strong>. In many cases, the effort required to bring people along exceeds the effort required to build the thing itself.</p><p>The alternative is allowing a small group of people to decide what’s best for everyone else. That has been the case. However, it doesn’t mean the rest of us will comply because of the inherent problems of small-group-of-expert-led solutions are limited visibility, blind spots, values misalignment, paternalism, and error amplification.</p><p>Even with everyone benefits from cooperation, Thomas Hobbes notes that people may fail to cooperate because they cannot trust others to do the same. So, to make good ideas happen, we need<strong> institutions</strong> to ensure mechanism exists to align behaviors and scale agreements.</p><p>Easy buttons aren’t just about technical solutions. Those good ideas need to transcribe into the <strong>general will </strong>and <strong>institutions</strong>. Especially the silent majority and institutions that implement the laws.</p><p>Good examples are:</p><ul><li>Seat Belt Laws</li><li>Smoking Restrictions</li></ul><p>Working backwards, my hypothesis is that solutions will be more effective correlating with broader range of support from the general will.</p><p><strong><em>2. Keep-It-Simple-Stupid (KISS) of Scalability</em></strong></p><p>To scale, the easy buttons need to be <strong>stupid-proof</strong>.</p><p>Fail-proof is a politically correct alternative, but to any architect of the social infrastructure, you need to address the stupid heads-on.</p><p>Information is plentiful. With codes and standards, we’re talking about thousands of pages that aren’t naturally for most people to memorize and comply to when their day to day is about building things with hands and tools. Fundamentally, that’s already an invisible human behavior barrier to codes and standards implementation. So what can we do?</p><p>The hourly rates for a human to review technical specs only get higher and higher, requiring people to become more and more specialized. We want to minimize tax burden and maximize consumer protection. Those two goals don’t necessarily align.</p><p>For example, to maximize the value of a permit, the public servants in codes and standards, notorious with subpar pay than private industry, need to have tons of experience and be better than their private counterparts.</p><p>Think about this for a second.</p><p>Structurally, as our society becomes more capitalistic, it weakens the fabric of public servants to provide effective consumer protection. Any easy button to be developed in this kind of environment will need <strong>to scale</strong> to carry specialize experience and consistently apply across large AND small bureaucracies.</p><p>Don’t get me wrong. I’m not saying there isn’t super smart and talented people in government work. In fact, public servants&#39; qualifications matter even more and should require high standards because taxpayers demand high-quality service.</p><p>I’m saying easy buttons need to <strong>be intentional in how it addresses the very basic frictions, root causes, and invisible roadblocks from human behaviors</strong> rather than just technical symptoms at the surface. Otherwise, your “stupid” problems will balloon to no ends.</p><p>The most effective solutions might not be the most technically sophisticated and it needs to be stress-tested — not by the best of us. That isn’t intending to be demeaning.</p><p>A speaker at the Electrification Summit said it more elegantly— focus on the <em>customer experience</em>. The very basic values that forged the system in the first place need to be the <em>first </em>thought, not a historical thought.</p><p><strong><em>3. Time is Money. How can we hack time?</em></strong></p><p>Lastly, the seasoning of easy button ingredients involves<strong> time</strong> —how fast can you implement?</p><p>Elon Musk spoke to in shrinking your 10 years goals to accomplish in 6 months&#39; time. This requires outside of the box thinking with incredibly intention and concentration on the problems.</p><p>This is more enabled in private and start-ups because time is boosted by capital and the promises of profits. In public settings, the rewards are less materialistic or monetary.</p><p>In public work, the strongest driver is the <strong>preservation of human lives </strong>that accelerate implementation. This is the very first thing I learned from our Deputy Public Works Director as an intern at the County of San Diego. He’s right. This is a noble objective for public service that should get the utmost time and attention.</p><p>A perception that plagues the Energy Codes is that it’s a nice to have. However, as the cost of resources rise over time inevitably, the impact of a slow death will hit us before we know it. In the case of data centers and water quality, it’s here sooner than we think.</p><p>Because I’m an analyst at heart and optimistic, here are few successful examples in the context of public service:</p><ul><li><strong>New York City CompStat (1990s)</strong> — William Bratton asked what if we knew where crime was occurring this week instead of next quarter, so NYC created CompStat that predicted weekly crime rates, geographical hotspot mapping, accountability meetings, and rapid deployment. The results reduced crime much faster than traditional practices from police departments to act on the last quarterly data.</li><li><strong>FEMA after Hurricane Katrina</strong> — FEMA reform after the historical disaster included pre-positioning supplies, pre-negotiated contracts, and building incidental command systems. This preparedness minimizes the decision-making that can occur before a disaster and enable faster responses.</li></ul><p>When it comes to energy, we’re racing against a clock that most people can’t see. The benefits of efficiency — and the costs of inefficiency — are often invisible, making urgency difficult to communicate.</p><p>Recently, I’ve been listening to the book “Outlive”. The author said there’s a fast death and a slow death. The worst consequences of bad energy resources management feel like a slow death, but when it strikes, it happens quickly. Kind of like global warming and cancer, something we created ourselves and unable to stop when it ripened. Yuck.</p><h3>Learning to Cook in the Dark Ourselves and Intentionally Use AI to Solve Our Problems</h3><p>Being the master of none, what I learned about growing anew in each role is the <strong>ability to adapt</strong>. To adapt in a new environment means learning to solve problems <strong>when you know you don’t know what you don’t know, </strong>willingness to do an <strong>ungodly amount of iterations</strong>, and <strong>putting in the work</strong>.</p><p>When we get to the edge of the unknown, I truly don’t think AI is going to solve it for us (more likely to controlling us — the problem). Maybe Yann LeCun’s world model might prove me wrong.</p><p>What I suspect is that reaching the frontier may require compromising some of the guard rails we put in place to protect ourselves. After all, this superintelligent machine we’re trying to build to outdo our makers is ultimately a tool. And like any powerful tool, it’s a bit like a knife.</p><p>You don’t give knives to children. Children who don’t know what they don’t know.</p><p>The hardest pill to swallow whether you’re on AI camp or not — to make easy buttons meaningful, the maker has to <strong>put in the work </strong>and solve problems in the dark. You are still in control of the micro-decisions and know where to fix it.</p><h3>A Few Hypotheses</h3><p>Beyond the characteristics of easy button, my hypotheses to enable easy buttons in the first place require us to address the <em>invisible barriers</em> that make scalability hard in implementing easy buttons. What’s the point of having great solutions without the systematic structures to scale them?</p><p><strong>1. Fundamental policy needs to shift away from a “what if, cost-effective” policy. Free energy for all. Period.</strong></p><p>This will shift the incentives to innovate and the incentives to scale. Painfully at first, no doubt, but more sustainable than a cost-effective policy. Yes, it’s fundamental to the Warren Alquist Act. Those goals were written 50 years ago. What does it take for our civilization to grow?</p><p>I know your first reaction might be — how dare you even suggest that? Are you crazy? Energy takes resources. How can it be free for all exist in capitalism?</p><p>Look — in our current civilization, productivity drives the economy. <a href="https://unchartedterritories.tomaspueyo.com/p/mountains">Global trends show that cooler climate and more comfortable climates correlates to higher productivity</a>.</p><p>For a capitalistic country — what’s a very compelling case for enabling free energy for all.</p><blockquote>The modern objective is to aim for true independence from energy cost burden and to enable humanity to strive for richer experiences in life.</blockquote><p>The strategy that informs the tactics. Otherwise, tactics are just noises, per the famous Sun Tzun.</p><p>Cost effectiveness is a tactic, not a strategy anymore. When it becomes less relevance where there is more emphasis on first cost and survivability, we need the flexibility to change tactics to resonate with the modern world.</p><p>If it’s any proof, Abundance, the book, is a trending for a reason.</p><p><strong>2. Establishing Trust.</strong></p><p>There has been a perception on the tug-o-war pull between energy efficiency and affordability.</p><p>The beef is not only exclusive with energy standards. The sentiment is more generalized with <em>any</em> building standards. It’s just the ones perceived to be most complex are kind of the first targets.</p><p>At the surface, to the policy makers that developed these efficiency standards, this highlighted a lack of trust likely rooted from the lack of understanding. The premise of any energy efficiency measure is that it needs to be cost effective (i.e. the benefits outweigh the costs). Some may argue the weakness of this calculation method is simply inaccuracies to reflect of the real world, not including other incidental cost and process cost that can tip the scale. These arguments raise doubts to the cost effectiveness results and the attitudes toward the adopted standards.</p><p>In addition, affordability centers around first costs- not limited to incremental cost and including other compliance related costs as well.</p><p>A typical homeowner can get sticker shocks when an ECC Raters quoted $700 to test the system after already paying $5,000 to install an HVAC system. Add the sticker price of permitting (let’s say $250), the inspector coming out and telling the contractor to rework, the contractor giving the homeowner the receipts of the change order — everyone is in a predicament of being unhappy. This is the way it often works in construction. The all-in cost expectation is almost never the same as the final bill, so it creates more opportunities for emotional turmoil and ill perceptions to form.</p><p>The value position of energy efficiency seems pale in comparison for the average Jill. Even to economist and policy analyst who question the relevance of cost-effectiveness snapshot, they can be swayed to make assumptions that such benefits may not be consistent or always credible.</p><p>An immediate solution to this is establishing stronger feedback loop system between what the market is doing and design the standards more efficiently. It will slowly establish trust and take a lot of resources to make it happen. To make aggressive progress with little resources, that’s where we can really use creative solutions or unconventional designs to establish trust.</p><p>All in all, <strong>trust is the butter to implement easy buttons</strong>. It’s a nonstarter otherwise. Without trust between agencies and institutions, we fight against each other instead with doubts and indecisions. We won’t get anything done.</p><p><strong>3. Efficiency needs a bigger marketing budget and strategy.</strong></p><p>In my years as a Senior Project Manager in military constructions, I see the same problems over and over when there’s a lack of understanding. It’s not just educating the contractors or your potential bidders, because you’re not truly resourced to do so. You mitigate through creating very clear expectations and channel their energy to solve dealbreakers.</p><p>Like how Apple sells “creativity” instead of computers and how Turbo Tax sells “convenience”, not a tax software. Sometimes, painting a picture isn’t really telling your story to people but influencing people to tell the story themselves.</p><p>Energy standards compliance’s single value proposition that resonates across contractors, designers, building departments, homeowners, utilities, and the commission is probably not energy savings, climate goals, or even cost savings.</p><p>The closest <strong>universal selling point</strong> is:</p><blockquote><strong>Get it right the first time.</strong></blockquote><p>The value proposition works regardless of whether someone is motivated by money, time, workload, risk, or policy.</p><p>California especially established a pre-rulemaking and rulemaking process to adopt new standards, providing stakeholders the last-ditch efforts to voice their concerns. This rigor for certainty needs continual discipline, appropriate institutional support, and the attention it deserves.</p><p><strong>Confidence with less wasted time. </strong>That’s how reduction in the burden of work happen. Compliance occurs when it’s undisputed, butter through trust and institution.</p><p>Furthermore, the compliance infrastructure needs to look like a duck and walk like duck. The cost to not getting right should be clear — not clear as mud.</p><p>The reputation of a policy is as important as its implementation resources.</p><p><strong>4. No Band-Aids. Period. Period. Period.</strong></p><p>As a system thinker, I absolutely understand why band-aids happen and loath the number of resources directed to something not truly meaningful for a quick dopamine hits. I have kids who ask for ice cream- it’s a quick fix to get their compliance, but is it what I really wanted? No.</p><p>Humanity is kind of a reactive species. We, as a collective, are short sighted. The one who have the foresight and the long-term outlook even when it’s truly the best for the collective — they don’t always win. I see these symptoms reoccurring across small to large organizations.</p><p>Band-aids hurt us in the long term. If it’s not good enough, then it doesn’t deserve to scale. It happens in implementation anyway because quick wins matter to people.</p><p>Easy buttons give those impressions, but we ought to be careful. This clip below describes the downfall of symptom management perfectly.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.tiktok.com%2Fembed%2Fv2%2F7471870511560166711&amp;display_name=tiktok&amp;url=https%3A%2F%2Fwww.tiktok.com%2F%40mmmovie_edits%2Fvideo%2F7471870511560166711%3Fis_from_webapp%3D1%26sender_device%3Dpc&amp;image=https%3A%2F%2Fp19-common-sign.tiktokcdn-us.com%2Ftos-maliva-p-0068%2Fo4fqcAJDnCahcxgFEAAWIAEWEB2xfspQ6RIqCq%7Etplv-tiktokx-origin.image%3Fdr%3D9636%26x-expires%3D1784304000%26x-signature%3DqWuSjebDRscyTxIbLCLyCM92GRs%253D%26t%3D4d5b0474%26ps%3D13740610%26shp%3D81f88b70%26shcp%3D43f4a2f9%26idc%3Duseast8&amp;type=text%2Fhtml&amp;schema=tiktok" width="340" height="700" frameborder="0" scrolling="no"><a href="https://medium.com/media/6901610cbce3f913a8ebddd27b66e15c/href">https://medium.com/media/6901610cbce3f913a8ebddd27b66e15c/href</a></iframe><p><strong>Other solutions </strong>— Comment if you have reactions or other thoughts to solutions.</p><h3>Closing with the Thinking Compass</h3><p>With great power, comes great responsibility. As I am sitting here consuming resources where <a href="https://www.youtube.com/watch?v=zV9v0RIE-iI&amp;pp=ygUVYnVzaW5lc3MgaW5zaWRlciB0b2Z1">someone else on the planet has to shovel plastic trash into the furnace to make tofu for their livelihood 7 days a week and ingesting the microplastic in our trash with it</a>, I feel privileged and the responsibility of this cost for comfort. Even if we fail, we ought to keep trying and collect data along the way.</p><p>At the end of the day, we share a single truth in public service is to <strong>make</strong> <strong>life better for others</strong>. In the best of our abilities. What we can in our control. Until our time is up. Perhaps that’s what drawn to me to public service to afford that as a mission.</p><p>Maybe it’s just the story I tell myself to keep me motivated.</p><p>I hope that this story does it for you.</p><p>In some small ways.</p><p>Even if you’re an echo.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FEfYtMLe7gqI%3Fstart%3D24%26feature%3Doembed%26start%3D24&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DEfYtMLe7gqI&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FEfYtMLe7gqI%2Fhqdefault.jpg&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/9b11552b2ae03b2ba214c59fffa2917f/href">https://medium.com/media/9b11552b2ae03b2ba214c59fffa2917f/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0ecab34147f2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Will Energy Codes survive WWIII?]]></title>
            <link>https://medium.com/@yungcodes/will-energy-codes-survive-wwiii-e6db53ba8c14?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/e6db53ba8c14</guid>
            <category><![CDATA[energy-efficiency]]></category>
            <category><![CDATA[energy-code]]></category>
            <category><![CDATA[geopolitics]]></category>
            <category><![CDATA[predictive-history]]></category>
            <category><![CDATA[petrodollar]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Tue, 12 May 2026 07:00:18 GMT</pubDate>
            <atom:updated>2026-05-14T21:05:02.692Z</atom:updated>
            <content:encoded><![CDATA[<p>Lately, <a href="https://www.youtube.com/watch?v=BTJGr78-zyw">geopolitical analyst Professor Jiang </a>has been getting things right. In 2024, he predicted Trump would win, that the US would start a war with Iran, and that United States will restart the draft. He was right on all 3 accounts. His framework — the slow unraveling of the petrodollar, the weaponization of the US dollar through Russia sanctions, the rise of a Eurasian trading bloc that doesn’t need America to intermediate — is no longer a thesis. It’s happening in real time.</p><p>I’m not writing to debate his predictions. I’m writing because of what they imply for something almost no one is talking about: the buildings we live in, and the codes that govern how they’re built.</p><h3>The Signals You Might Have Missed This Month</h3><p>Last month in April, The U.S. Department of Housing and Urban Development (HUD) and the U.S. Department of Agriculture (USDA) have <a href="https://www.nahb.org/blog/2026/05/hud-rescinded-mandate">rescinded a 2024 federal rule</a> that required new homes financed through FHA or USDA-backed loans to meet the latest model energy codes. In the same month, the DOE released its proposed <a href="https://www.energy.gov/documents/doe-fy-2027-volume-2-cmei">FY2027 budget</a>.</p><p>The headline number — a 10% increase to $53.9 billion — sounds like growth. It isn’t. Nearly the entire increase flows to the National Nuclear Security Administration, nuclear weapons and defense programs. For everything else, the budget cuts non-defense DOE programs by 11%.</p><p>The Office of Energy Efficiency and Renewable Energy — the federal home of building energy codes, weatherization assistance, and state energy programs — has effectively been dissolved. Its successor, the Office of Critical Minerals and Energy Innovation, faces a 63% budget reduction from the programs it absorbed. Building technologies: nearly zeroed out. The Weatherization Assistance Program, which reduces energy costs for low-income households: eliminated. State and community energy programs: gone.</p><p>Just last week in early May, the Department of Energy (DOE) released a <a href="https://www.federalregister.gov/documents/2026/05/04/2026-08646/updating-and-improving-the-methodology-for-assessing-affordability-and-cost-effectiveness-of">public request to updating and improving methodology for assessing affordability and cost effectiveness of building energy codes</a>. Whether that effort can fully achieve its goals remains an open question, as the underlying construction cost data is rarely transparent and often too inconsistent to produce fully precise results.</p><p>The administration’s framing is “energy dominance.” The practical effect is the systematic defunding of the institutional infrastructure that makes energy efficiency policy function at a national level.</p><h3>Why Energy Efficiency Is Politically Vulnerable</h3><p>Energy efficiency has always been a “yes, if” policy.</p><p>In military construction, <a href="https://stg.wbdg.org/dod/ufc/ufc-1-200-02">UFC-1–200–02</a> — the Unified Facilities Criteria governing how the Department of Defense builds — frames energy efficiency as “do this, if practical” and “do this, if it meets life cycle cost effectiveness.” That conditional framing reflects standard economic policy logic: optimize under stable conditions.</p><p>But what happens when economic stability ends?</p><h4>How We Got Here</h4><p>During World War II, the <a href="https://www.nps.gov/articles/000/material-drives-on-the-world-war-ii-home-front.htm">War Production Board</a> banned all nondefense construction in April 1942 and put strict limits on any alteration of existing residential buildings. When the ban lifted, the pent-up housing demand was enormous — and speed became the only metric.</p><p>Energy codes didn’t exist yet in that era — they weren’t established and implemented until the 1970s. The oil embargo and the series of blackouts in New York spurred the birth of building energy codes. As a result, we’re seeing homes with <a href="https://www.construction-physics.com/p/looking-at-energy-use-in-us-residential?utm_source=chatgpt.com">70%+ of efficiency today compared to pre-1960s level</a>.</p><p>Energy efficiency policy made sense and saved fuel when crisis set the terms. Politics followed. In 2015, the National Association of Home Builders (NAHB) strongly supported the energy codes and its transparent process in communicating cost effectiveness before adoption. In 2025, NAHB implied that energy codes lead to housing unaffordability. So how did we get here — to a place where energy efficiency has bipartisan concurrence to be deprioritized? Why did the narrative change 180 degrees between <a href="https://nationalmortgageprofessional.com/news/53830/nahb-testifies-congress-push-cost-effective-energy-measures">2015</a> to <a href="https://www.nahb.org/blog/2025/09/energy-subcommittee-testimony">2025</a>?</p><p>Having gone through a build myself in 2025, I can’t say that the compliance cost alone from the California Energy Codes would actually be significant enough to stop any investor to build as long as there is a <strong>strong demand</strong> (and there is!) despite of the sunshine tax. So, what’s the <em>real</em> other motivation?</p><h4>This Isn’t Just a Fuels Fight</h4><p>At the surface, the latest model codes have been increasing in stringency, while being fuel-neutral, in two notable differences: electrification and heat pumps.</p><p>Many of my colleagues naturally wondered whether winding back or pausing energy codes indicates a fight between oil/gas and electrification? Between fossil fuels and heat pumps?</p><p>To them, I say: look at Japan. California and Japan are neck to neck in global GDP ranking. Japan has approximately 127 million people whereas California has about 39 million people. Residential adoption of electric heat pumps climbed 38% in 2024 alone, driven by policy incentives and rising gas price volatility. Daikin, Mitsubishi, and Panasonic — all Japanese companies — collectively dominate the global heat pump market. Japan essentially invented the modern heat pump industry. Having lived there for four years, I saw heat pumps everywhere, all-electric in most homes.</p><p>Yet Japan still buys natural gas. Their <a href="https://www.enecho.meti.go.jp/en/category/others/basic_plan/pdf/7th_outline.pdf">7th Strategic Energy Plan</a> identifies a 2040 energy mix where renewables make up 40–50% of power generation, nuclear 20%, and thermal power — including gas — between 30 and 40%. Gas isn’t being phased out that fast. It’s being positioned as a flexible backup.</p><p>Japan didn’t choose between affordability, energy efficiency, and economic pragmatism because efficiency is foundational in their culture. They are especially sensitive to limited resources. The question isn’t whether America can afford to build efficient buildings. It’s whether we can afford the confusion between a fuel fight and a <em>financial structural problem</em>.</p><h4>The Petrodollar Trap</h4><p>Professor Jiang’s argument is that the petrodollar <em>stability is ending</em>. More specifically, the petrodollar’s purchasing power — which has quietly subsidized America’s ability to run deficits, borrow cheaply, and maintain consumer purchasing power — is eroding. The Russia sanctions of 2022 were the inflection point: by weaponizing dollar access, the US broke the foundational promise of political neutrality that made the dollar worth holding. This reflects in gold’s and silver’s prices as China sought another currency. According to Jiang, every country is watching this moment and making moves.</p><p>When the economy destabilizes, cost-effectiveness calculations — the entire analytical basis for energy efficiency policy — raises doubts. <a href="https://www.nahb.org/advocacy/top-priorities/building-materials-trade-policy/how-tariffs-impact-home-building">Tariffs</a>, <a href="https://raiven.com/blog/covid-19-has-damaged-the-electrical-supply-chain/">supply chain disruptions</a>, <a href="https://iranwarcost.watson.brown.edu/">energy price volatility</a>, and currency pressure make long-term payback calculations increase uncertainties to the roof. And when the analytical basis seem nonintuitive, so does the political will.</p><p>The efficiency mandate gets reframed as an overly complex ideology. A bureaucratic burden. An unaffordable distraction.</p><p>We’re already seeing this reframing happening even in the “progressive” states:</p><ul><li><a href="https://newbuildings.org/the-false-economy-of-weaker-energy-codes/">AB130 in California pausing the Energy Codes update for housing in 2028</a></li><li><a href="https://newbuildings.org/the-false-economy-of-weaker-energy-codes/">Hawaii’s Governor singing the 15th Emergency Proclamation Relating to Affordable Housing, </a>and</li><li><a href="https://newbuildings.org/the-false-economy-of-weaker-energy-codes/">New York pausing All-Electric Buildings Act enforcement</a>.</li></ul><p>Energy codes get lumped in with activities that incurred unnecessary debt — technical debt for being overly complex, opportunity cost to more housing supply, and affordability barriers to builders and investors. They all sound like they make sense at the surface, but the data is still unclear.</p><p>Part of why it’s unclear is that data isn’t static. People feel the economic symptoms in real time: <a href="https://energybadboys.substack.com/p/why-energy-efficiency-wont-save-you">bills rising even as efficiency improves</a>, and the growing cost of integrating renewables. When utilities shift infrastructure maintenance and upgrade costs into fixed charges for ratepayers, the payback on efficiency investments looks weaker against today’s fuel prices. But that’s a pricing structure problem, not an efficiency problem. The fundamentals don’t change — wasting less is good for everyone. And fuel prices can spike violently in a short period, as any geopolitical conflict reminds us. The fight over resources can happen any day.</p><p>At the last Implementer conference hosted by the CPUC, a panelist made a point worth paraphrasing: there are frankly <em>conflicting goals</em> that make implementation difficult when the rubber meets the road. This tension isn’t unique to the EM&amp;V world — it plays out across the efficiency sector at large.</p><p>We want affordability <em>and</em> resilience. We want efficiency <em>and</em> reliability.</p><p>First costs and long-term implementation goals don’t always align. Until the right thing to do is also the cheapest thing to do upfront — like LED lighting eventually became — policy intervention will always be necessary to bridge the gap while markets catch up.</p><h4>The Real Tea: A Structural Contradiction</h4><p>Jiang argues that America’s National Defense Strategy is trying to maintain the petrodollar. If American monetary dominance structurally requires global oil consumption to continue — not just American consumption, but global consumption — then American domestic energy efficiency policy is quietly undermining its own National Defense Strategy.</p><p>This is why rebranding energy efficiency under the umbrella of energy security is not really going to be very effective.</p><p>Every solar panel installed in Germany, every heat pump deployed in Japan, every EV sold in China is a transaction that didn’t require dollars. At scale, that erodes the petrodollar’s foundation. Every building that uses 40% less energy is a small vote against the system that funds American prosperity. Not intentionally. Not consciously. But structurally.</p><p>This is more than just the oil lobbyist. If the US financial system was architected around oil dependence, that architecture created powerful structural incentives against energy efficiency — <em>incentives that don’t need anyone to be villainous to operate</em>.</p><p>And here’s the worst part. If Jiang’s prediction is correct about the petrodollar that is quickly eroding regardless of what American energy efficiency policy does — we would be sacrificing American building performance and energy independence to protect a system that’s already broken. That is the <strong>worst</strong> possible trade.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5TE1ySqvvi_1FW4cL8bCGw.png" /><figcaption>A meme from reddit, recreated with ChatGPT5.5</figcaption></figure><h3>How Building Energy Codes Actually Collapse</h3><p>If I allow myself to be a cynic for a second, codes don’t die in a single dramatic vote. They die a slow death and erode through institutional failure:</p><ul><li>AHJs — Authorities Having Jurisdiction, the local officials who enforce codes — become under-resourced and overwhelmed. Enforcement stops not by policy but by capacity and implementation.</li><li>Emergency executive orders suspend “regulatory burdens” in the name of mobilization.</li><li>Supply chain breakdowns make compliant materials unavailable, and builders substitute whatever they can find.</li><li>Political pressure reframes efficiency mandates as elite concerns during a survival moment.</li><li>Federal preemption overrides state and local standards entirely.</li></ul><p>Each of these mechanisms can happen under the scenario Jiang describes. Several are already visible in the DOE FY27 budget proposal.</p><h3>What We Lose When Codes Collapse</h3><p>The stakes are not abstract.</p><p>Buildings permitted during a code collapse window lock in 50 to 75 years of excess energy demand. That is not a rounding error — that is a generation of households paying permanently elevated utility bills, in a period when energy costs will already be rising. We’d also loose expertise, the time to innovate, and technological advantage.</p><p>Grid fragility worsens at precisely the moment when military resilience demands the opposite. The Department of Defense’s own energy security strategy depends on reducing military installation dependence on the commercial grid. That strategy is undermined if the civilian buildings surrounding every base are drawing maximum load from an overwhelmed and stressed system. Efficient buildings are part of the load reduction that keeps the grid lean and can even be flexible (with Vehicle2Grid, backup power, and island-mode). We lose what could have been — a strong roll-out of the most resilient and efficient buildings.</p><p>And the irony that should trouble every defense strategist: energy codes are more strategically important during wartime, not less. But they are also most politically vulnerable during wartime, especially concurrently with the financial incentives to global dominance.</p><h3>The Discipline We Actually Need</h3><p><strong>America needs the discipline of energy efficiency, not just the performative acts of compliance theater.</strong> I’m not saying building energy codes are perfect, but the intent has always been clear to moving toward grid resilience and a sustainable future. That intent still matters.</p><p>At the national level, the direction to the energy code discipline seems to be about <a href="https://newbuildings.org/the-false-economy-of-weaker-energy-codes/">simplification</a> in the name of affordability- a simple payback return vice the traditional cost benefits analysis. This is alarming because it is prioritizing short-term gains first and undermine the underlying intent and long-term outcome.</p><p>Remember that the buildings we permit this year will still at least be standing in 2075. The codes we enforce — or fail to enforce — will determine whether those buildings are assets in a constrained future or liabilities we cannot afford.</p><p>That choice is being made right now, in budget line items few people are reading and permit offices few people are watching. According to Nora Wang Esram, the CEO at the New Building Institute, <a href="https://newbuildings.org/the-2030-iecc-whats-decided-whats-at-stake-and-where-we-go-from-here/">the window</a> to affect decisions in the future of energy codes is happening right now. This shouldn’t be driven by narrow set of voices within the code development process, nor should it be overshadowed by the ditch to preserve the petrodollar. It affects all of us. It affects the places we leave behind to our children.</p><p>I think my first call of action was to engage and pay attention, but it’s hard and unrealistic. We all have responsibilities and distractions — frankly, we don’t have time. There is urgency to make things work (streamlining) — in speed <em>and quality</em>. We can’t compensate the latter without a plan to make up for it. Let’s work together, lift each other up, and get this right.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e6db53ba8c14" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[5 Lessons I Learned About Analysis Paralysis (After Overthinking Everything)]]></title>
            <link>https://medium.com/activated-thinker/5-lessons-i-learned-about-analysis-paralysis-after-overthinking-everything-32b5e070cf83?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/32b5e070cf83</guid>
            <category><![CDATA[mindset]]></category>
            <category><![CDATA[analysis-paralysis]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[overthinking]]></category>
            <category><![CDATA[overstimulated-moms-club]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Thu, 07 May 2026 04:32:41 GMT</pubDate>
            <atom:updated>2026-05-07T04:32:41.461Z</atom:updated>
            <content:encoded><![CDATA[<p>As I get older, I think more deliberately about how to make my effort go further. There’s less time, and far more competing for attention (thanks to doom scrolling — and the little humans). Something about being over stimulated really propel the instincts to get things done fast and at the same time, overthinking everything. Does that makes any sense?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*rbOntH5607TQq__vW59YJg.gif" /><figcaption>My hamster mind is werking.</figcaption></figure><p>I’ve seen organizations move fast and others move painfully slow. Both, in their own ways, fall into their own analysis paralysis — either by <strong>over-investing in low-impact work</strong> or by <strong>getting stuck in internal stalemates</strong> that prevent any real progress. It’s a hard pill to swallow to recognize complacency in analysis paralysis and how complacency can be very expensive when it compounds over time.</p><p>In my current research work, slow shipping time appears like analysis paralysis. On the other hand, rigor and quality frankly take time and a lot of discipline.</p><p>I’ve drawn from my own work experience, along with the thinking of people much smarter than me (via podcasts and talks), to shape how I approach analysis that can avoid that false impression of analysis paralysis and the bothersome negative side effects.</p><p>Here’s what I’ve learned — and a reminder to my future self:</p><ol><li><strong>Have strong objective(s)</strong>. The work that is most aligned to the objective gets priority. Things that don’t align well to the objective gets deprioritize. The stronger your objective is (and less vague), the more effective it is to identify what things align to your primary objective.</li><li><strong>Sequence matters. </strong>Sometimes, you give the audience <strong>what they want first</strong>, then you can give them what they need.</li><li><strong>Paint the picture about the counterfactuals </strong>(a fancy word frequently used for the alternative path if otherwise) <strong>to build conviction.</strong> I’m seeing this recommended a lot in more formal academic work.</li><li><strong>Knowing when to stop</strong>. Your inner Type A instinct will push you to do more — x, y, and z. Pause and ask: does the audience actually value that additional effort? Often, they won’t say what they want explicitly right away. Reduce the <em>communication tax early </em>will make it easier later.</li><li><strong>Ship fast and show the process</strong>. There’s value in letting people see how decisions are made. Sharing the “making of the sausage” builds trust, reinforces your thinking, and <em>trains your audience</em> through repetition. In marketing, 7 times is a magic number in paying for ads to influence. People rarely remember what the polished outputs are about — they remember the experience.</li></ol><p>When I feel like I’m overthinking and spinning my wheels with cogitative overload, I’ll go back to these principles. None of this is particularly new — but it’s easy to forget or distracted from the next thing.</p><p>— — — — —</p><p>Thanks for reading. This is me thinking in public — testing, refining, and sometimes being wrong. If that’s interesting to you, clap, subscribe and follow along.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=32b5e070cf83" width="1" height="1" alt=""><hr><p><a href="https://medium.com/activated-thinker/5-lessons-i-learned-about-analysis-paralysis-after-overthinking-everything-32b5e070cf83">5 Lessons I Learned About Analysis Paralysis (After Overthinking Everything)</a> was originally published in <a href="https://medium.com/activated-thinker">Activated Thinker</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Codes & Standards Programs as Products?]]></title>
            <link>https://medium.com/@yungcodes/codes-standards-programs-as-a-products-8ff7deb01df7?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/8ff7deb01df7</guid>
            <category><![CDATA[energy]]></category>
            <category><![CDATA[codes-and-standards]]></category>
            <category><![CDATA[public-programs]]></category>
            <category><![CDATA[government]]></category>
            <category><![CDATA[building-code-compliance]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Fri, 10 Apr 2026 04:36:59 GMT</pubDate>
            <atom:updated>2026-04-10T05:11:14.439Z</atom:updated>
            <content:encoded><![CDATA[<p>Rethinking How We Design Public Programs</p><p>Since the <a href="https://pda.energydataweb.com/#!/documents/4295/view">last CPUC EM&amp;V stakeholder meeting</a>, something has been quietly marinating in my brain from the Codes &amp; Standards presentation— and now it’s all I can see. What I remembered someone said:</p><p><strong>What if Codes &amp; Standards programs should be designed more like products?</strong></p><p>At first, I tried to simplify the distinction. A product has a beginning and an end. A program doesn’t.</p><p>Easy enough.</p><p>But the more I think about it. It doesn’t quite feel enough or clicked it for me yet. And so, I went down the usual rabbit hole — searching, reading, thinking — and landed on a more useful distinction:</p><ul><li><strong>Product design</strong> focuses on <em>user experience</em></li><li><strong>Program design</strong> focuses on <em>systems of coordinated interventions</em></li></ul><p>A program provides levers and doesn’t try to control the entire ecosystem. A product is how people actually experience that system.</p><p>Think of it this way:</p><ul><li>In a transportation system, <strong>a program</strong> may incentivize riders to use public transit than individual cars by taking the funds from parking passes revenue and support the public transit system.</li><li>An easy-to-use train app or ticketing system<strong> </strong>is a <strong>product</strong>.</li></ul><p>One tries to influence the ecosystem. The other determines whether anyone can actually use it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/498/1*qkFqgYSB6-FvkLwNs7qdoA.gif" /><figcaption>POV: Program Managers be like…</figcaption></figure><h3>What Do Strong Programs Look Like?</h3><p>When I think about strong programs in the last 10–20 years, a few stand out:</p><ul><li>Energy Star</li><li>Apple App Store</li><li>Open Government Data initiatives</li></ul><p>These programs feel very different on the surface — but structurally, they share something important. A strong program:</p><ol><li><strong>Creates the conditions for many products to emerge</strong></li><li><strong>Maintains a strong gravitational center</strong> (clear governance, consistent rules)</li><li><strong>Enables feedback loops that actually influence behavior</strong></li></ol><p>That last point is a bonus.</p><p>With the accessibility and abundance of data in our modern world, most programs collect data. Strong programs use data to adapt themselves. They don’t just report metrics — they feed insights back into the system, shaping future products, decisions, and behaviors. This helps them to autocorrect or to scale without falling apart.</p><h3>Where Does the Energy Codes Programs Fit?</h3><p>Now let’s bring this back to something more concrete: <strong>California’s Title 24 Energy Code.</strong></p><blockquote>Is it a program or a product?</blockquote><p>At a high level, the Energy Codes ecosystem looks like this:</p><ul><li>The California Energy Commission (CEC) sets new standards every 3 years, often adding stringency via new measures or optimizing how compliance occurs</li><li>Local jurisdictions (AHJs) enforce compliance</li><li>Designers, contractors, and verifiers implement requirements</li></ul><p>Compliance is demonstrated through:</p><ul><li>Prescriptive or performance paths</li><li>Documentation (forms)</li><li>Third-party verification</li></ul><p>Support comes from multiple layers:</p><ul><li>Training and outreach (e.g., EnergyCodeACE, CPUC funds programs through IOUs, RENs, CCAs for C&amp;S support)</li><li>Some Outreach &amp; Education, Software Support (Vendors, CEC)</li><li>Market actors executing in practice</li><li>Manufacturers provide products that meet minimum code requirements</li></ul><p>This is a <strong>decentralized system</strong>. Like how our cities and local governments are built from.</p><h3>The Products of the Energy Code</h3><p>If you zoom in, the Energy Code ecosystem isn’t just regulations because it’s not like “you shall do this”. It’s also “you shall do what performs well”. To determine that and to verify that, you need products. It produces <strong>products</strong>.</p><p>It isn’t just an intervention that forces the masses to do the right thing because it becomes the products that touches many market actors to interact with and utilize in order to produce an acceptable building to California’s society.</p><p>Some of the most important ones:</p><ul><li>Energy modeling tools or compliance software</li><li>Verification protocols like forms</li></ul><p>And especially… <strong>forms. </strong>Yes — forms. Everyone hates them because they do feel intimidating or even overwhelming to the untrained eyes.</p><p>Digression — Much of the efforts of compliance software in standardizing many inputs is to enable streamlining, but it also creates by product where “compliance modeling doesn’t mean performance modeling”. To a normal person, that doesn’t mean anything. To an energy modeler, it means you prove to your client that they have or are or are going to buy the best design because true performance lies in another energy model — completely separate from the compliance model. Sometimes, it’s a necessary evil that you’d do both.</p><p>Back to forms. Yes — forms. Forms are not just paperwork. They are <strong>data products</strong> that:</p><ul><li>Touch nearly every market actor</li><li>Structure how compliance is demonstrated</li><li>Influence design decisions (especially if you have a target of x% above code)</li><li>Create entire workflows (and industries) for design team who works on California projects</li></ul><p>In many ways, these products <em>are</em> the interface to the program. From ongoing conversations with building departments and other market actors — and from my own time working within the system— I question how you can design any program if the program performance is largely a function of how well these primary products perform in practice.</p><p>Which raises a deeper question:</p><p><strong>Could the system survive without the core products?</strong></p><p>And if not —</p><p><strong>Are we designing critical “programs” that are more like products intentionally as products?</strong></p><p><strong>If the core products are weak</strong>, then implementation of external programs will compensate by adding more training, more guidance, more oversight, and more documentation. This doesn’t fix the issue — it shifts the cognitive and operational burden onto the end users. Overtime, this creates fatigue, workarounds, and inconsistent compliance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/716/1*GQRB5Jjv2Q0ylJIUpzBi-A.png" /><figcaption>We’re staying Positive Pancy.</figcaption></figure><p><strong>Dare I ask, have we put the appropriate resources toward the core products?</strong></p><h3>The Cost Problem (and Why It Matters)</h3><p>At some point, every product and program conversation lands in the same place:</p><p><strong>Cost.</strong></p><p>Right now, across both federal and state conversations, two themes dominate:</p><ul><li>Streamlining</li><li>Affordability</li></ul><p>In Codes &amp; Standards, cost is not straightforward.</p><p>Incremental cost:</p><ul><li>Isn’t static</li><li>Isn’t uniform across the State</li><li>Isn’t fully observable or transparent</li></ul><p>It’s a <strong>distribution hidden inside a competitive market</strong>. There’s an entire discipline in construction just to estimate it.</p><p>So, when we ask:</p><blockquote>“Is this cost-effective?”</blockquote><p>It’s easy to disclaim that we are working with a lot of incomplete data and a lot of assumptions. It’s easy to assume people make mistakes.</p><p>To prove anything or to do this right, it rigorously takes a lot of resources and time — not a natural investment. Sometimes, we trust the experts to take the risks and determine what is good enough. And when time isn’t available, we fall back on something else: what’s mostly agreeable<strong>. </strong>Until you have to prove it to someone new.</p><p>Therefore, I’m not sure really sure how entirely productive it is to argue about cost scientifically before it becomes a political debate. I can be wrong. You can be wrong. We can both be wrong. Does it matter because <strong>the public good value</strong> doesn’t make it to the consumer while we’re busy trying to prove a point?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*2NaSTaC6lHrlGRvI_6RGnQ.gif" /><figcaption>The technical debts got interests! In the form of fatigue and debates.</figcaption></figure><h3>Beyond the Textbook</h3><p>This is where I think Codes &amp; Standards programs hit a wall.</p><p>If we rely only on textbook methods:</p><ul><li>We move too slowly</li><li>We oversimplify complex systems</li></ul><p>But if we abandon rigor:</p><ul><li>We lose credibility</li></ul><p>We lose either way. So, the real challenge is: <strong>How do we operate scientifically in a system full of ambiguity?</strong></p><p>This is where a product mindset might help.</p><p>Because product design:</p><ul><li>embraces iteration</li><li>relies on feedback loops</li><li>prioritizes real-world usage over theoretical completeness</li></ul><p>It doesn’t wait for perfect information. It <strong>learns by doing</strong>. So, I get it. Why product design thinking makes sense as the next step. However, every program should really evaluate whether they behave more like a program or a product. Even if you have a program in your name — that doesn’t mean you’re not a product.</p><p>Beyond that, there are <strong>scale </strong>and <strong>equity considerations</strong>. To me, equity means not unintentionally creating hardship for even a small group. We need to really stand by the tradeoffs that we accept today because they become the conditions people inherit tomorrow.</p><p>I’ve seen this across different countries:</p><ul><li>Some optimize purely for cost</li><li>Others invest in durability and long-term livability</li></ul><p>And the difference shows — not just in buildings, but in communities.</p><h3>Final Thoughts: A Different Kind of Question</h3><p>Maybe the real question isn’t:<br> <em>How do we make compliance cheaper or faster with product thinking vice program thinking?</em></p><p>But:<br> <em>What kind of world are we trying to build — and are our programs (interventions) and products (user experience) designed to support that?</em></p><p>Because even after a disaster — even after a tsunami — some buildings will still stand. And those structures will define what’s possible next for the generation that remains.</p><p>— — —</p><p>Thanks for reading this far. I hope it was slightly entertaining. I’ve been leaning a bit philosophical lately, but more practical posts are coming — things like mini vibe code projects, explore, and learn from, especially in data analytics arena (where I also count my-end-user-experience-with-AI). Aiming for a post a month.</p><p>If you’d like to see that, clap, comment, and/or subscribe!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8ff7deb01df7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How To Deal with AI Risks in Public Service]]></title>
            <link>https://medium.com/@yungcodes/how-to-act-on-ai-risks-in-public-service-a327dffeb858?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/a327dffeb858</guid>
            <category><![CDATA[public-service]]></category>
            <category><![CDATA[public-servants]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[government]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Thu, 26 Feb 2026 05:15:39 GMT</pubDate>
            <atom:updated>2026-03-07T00:12:07.465Z</atom:updated>
            <content:encoded><![CDATA[<p>It’s been a little jarring how AI have been rapidly growing. If you’re not an AI company now or 5 years ago, you better think about how it can integrate into your workflow because it’s going to affect your bottom line. If you’re a start-up, you might <a href="https://www.shovels.ai/blog/february-2026-newsletter/">end up being an AI-company</a>.</p><p>Like my mother said, “the dose makes the poison”. These days, it feels like <a href="https://finance.yahoo.com/news/openais-gpt-5-3-codex-123124816.html?guccounter=1&amp;guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&amp;guce_referrer_sig=AQAAADIQWbGZHP4cNHY1UK-DYeAZVX0DSj5oxXjvlIqU1qJ8FoJXW01E0bwqJymJFEUKczmneWz18gtRf0z6gdtSaiwMXW6JeTc-2a9ov6i9Z2g9-iNGoDbMaJrnGqC4d6RJt2QQczYIWfNLSw7ZKPy_8LFM5zMgLEJZaToxj-webmHx">humanity is experimenting</a> with the dosage of AI.</p><p>With private businesses, the labor cost is competing with (<a href="https://hbr.org/2026/01/companies-are-laying-off-workers-because-of-ais-potential-not-its-performance">the potential</a>) compute cost. In government work, on the other hand, the challenge is <strong>not simply</strong> <strong>productivity</strong>.</p><blockquote>Even if public institutions can adopt AI as fast as private industries, the real change AI introduces to public institutions is a shift in decision authority. Work will still be done by people, but increasingly through systems whose reasoning they did not personally perform. The central question becomes how people <strong>remain accountable for outcomes produced by machines</strong>.</blockquote><p>We can imagine the worst outcomes from science fictions like iRobot and the Foundation (one of my favorite books), both authored by Issac Asimov written <strong>in the 50s</strong>. In the Foundation novel, humanity intentionally pushed AI out of society after long conflicts and fear. In the early era, it worked. Intelligent robots were governed by the Three Laws. Despite having the “perfect AI”, it led to a stagnant society and robot wars. I won’t spoil the plot twist for you. Similarly in the popular Dune series, humanity <em>chose to overthrow</em> the “thinking machines” that governed them.</p><p>I’m hopeful for a better future though, at the risk of being errored on the side of optimism. One can dream, right?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/682/1*uswMDPR58cG7zacuT9nuFQ.png" /></figure><p>The truth to where we’re headed — I think — is going to depend on the <strong>intent and how fast we achieve that intent to other unintended consequences.</strong></p><p>Two <strong>intents</strong> stand out at the surface:</p><ol><li>AI can <strong>trim the growth of public servants or government spending. </strong>We’re already seeing layoffs happening in techs. <strong>Costs shift</strong> from payroll to vendors, compute infrastructure, and the industries that support them. In fact, this was much of DOGE’s mission to reduce the national debt through trimming government spending.</li><li>AI can <strong>help public servants to do more work, with less</strong> — delivering more public goods with less resource, reducing the burden of work, and solving problems from previous bottlenecks. Agencies may not become smaller — they may become <strong>denser in responsibility</strong>.</li></ol><p>I’m not entirely sure how I feel about the first and especially the second one. There’s definitely a preference for reducing the burden of work on a personal level. The other effects seem also very real.</p><p>Nevertheless, I believe in nobler goals like these will win:</p><ul><li>Increasing accessibility for the public</li><li>Enable agencies to handle more cases</li><li>Enhancing consistency to produce similar outcomes for similar cases</li><li>Prioritization of targeted goals under limited resources</li><li>Provide the institutional memory minimizing turnover delays</li></ul><p>At first, I want to argue that focusing on AI’s influence on workforce size is the wrong focal point. I still think it is to some degrees, but I know it’s the first thing that come to mind for some, if not most people.</p><p>It makes sense because that’s all in the headlines these days. The <a href="https://hbr.org/2026/01/companies-are-laying-off-workers-because-of-ais-potential-not-its-performance">threat to white collar jobs</a> looms over the smartest of us. It’s less about technical skillsets because that can be bought, fairly cheaply (<a href="https://fortune.com/2026/01/30/openai-ipo-ai-bubble-investor-tolerance-for-cash-burn-unprofitable/">at least for now</a>).</p><p>The cost shift is already happening.</p><p>Here are the changes to the federal workforce in the last 10 years.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iaDhUkfz8874uNiAgmAIww.png" /><figcaption>Source: <a href="https://data.opm.gov/explore-data/analytics/workforce-changes">OPM Federal Workforce Data</a>, snapshot on 2/24/26</figcaption></figure><p>Here is <a href="https://www.gao.gov/products/gao-14-215">the data</a> for prior to 2015. In summary, since 9/11, the federal workforce has been generally trending upward except for what just happened in 2025 with multiple rounds of <a href="https://www.npr.org/2025/03/15/nx-s1-5328721/reduction-in-force-rif-federal-workers-job-cuts-musk-doge-layoffs">voluntary reductions</a>. For Californian state public employee data, see <a href="https://publicpay.ca.gov/Reports/ReportBuilders/Trending.aspx?ChartType=1&amp;DataSet=1&amp;year=2024&amp;EntityTypeID=1,2,3,4,5,6,9,8,7,10,11">here</a>. And just maybe, it’s a coincidence that large government <a href="https://apnews.com/article/trump-ai-openai-oracle-softbank-son-altman-ellison-be261f8a8ee07a0623d4170397348c41">investment into AI</a> is happening at the same time because <a href="https://www.whitehouse.gov/articles/2025/07/white-house-unveils-americas-ai-action-plan/">it’s becoming a national security risk measure</a>.</p><p>In California, Governor Newsome provided an <a href="https://www.gov.ca.gov/wp-content/uploads/2023/09/AI-EO-No.12-_-GGN-Signed.pdf">executive order to propagate the use of responsible Gen AI</a>. The legislatures also want <a href="https://ai-analytics.wharton.upenn.edu/wharton-accountable-ai-lab/sb-53-what-californias-new-ai-safety-law-means-for-developers/">safety and responsible AI framework </a>moving forward. I already see good work from the City of San Jose — <a href="https://www.sanjoseca.gov/your-government/departments-offices/information-technology/ai-reviews-algorithm-register/govai-coalition">organizing with local governments</a> across the nation to help each other through implementation and doing so collaboratively.</p><p>Whether or not you agree with the intentions, these signals mean AI coming to the public space sooner or later. The next questions to ask are: when and how?</p><p>Much of the government employees&#39; tasks are eligibility determinations, risk scoring, inspection prioritization, and interpretation of the law. Decisions are being made that impact people’s quality of life from protecting our nation, producing clean water, to approving an EBT voucher.</p><p>Many tasks — assembling documentation, formatting analysis, preliminary review — can now occur before a human meaningfully intervenes. The employee becomes less the originator of the reasoning and more the validator of a conclusion. When that happens, <strong>responsibility no longer matches comprehension</strong>.</p><p>Machines may recommend, but people act. That is the starting assumption, but in reality, we will all experiment with the tolerance for machines to hold authority somewhere in our process.</p><p><strong>What conversations should we be having?</strong></p><ol><li>What vision do we have for AI potential future to work beside us?</li><li>How should we organize information in a way that help us leverage this tool?</li><li>How can we structurally protect what is most sensitive or critical for human-control versus machine-control?</li><li>What boundaries are going to save us in the long run when <a href="https://arstechnica.com/information-technology/2026/01/ai-agents-now-have-their-own-reddit-style-social-network-and-its-getting-weird-fast/">changes</a> exceed our capacity to process them?</li></ol><p>If you don’t answer them now, then you might forget how your answers might change a year from now. This is the danger to corroding decision authority. It happens quietly. In little details and micro decisions. Like the trees that quietly fell in the forest.</p><p>As an analyst these days, this feels strange. Sometimes AI feels like a frenemy I’m training to potentially replace me, but I know it pushes me to think more critically and more understanding how results affect people and processes. Problems that used to be analytical start becoming coordination problems. Skills I once associated with project management — communication, judgment, and timing — matter as much (if not even more so) as computation.</p><p>Even if we become more sophisticated analytically, we will still battle assumptions over the truths and special interests competing with public interests.</p><p>Let me end my train of thoughts here. <strong>What must a typical public servant actually do differently now?</strong></p><ul><li><strong>Organize the chaos </strong>— The gut feeling has a pattern, so being organized pays. Your inputs are your treasures.</li><li><strong>Create traceable work product you can understand</strong>— If the product outpaces your ability to verify it, slow down. Restructure information based on data principles and translate-ability.</li><li><strong>Document decisions like a decision engineer</strong>— Document your inputs and outputs. Note on why a conclusion was acceptable to act on. A better blog post that explains it <a href="https://www.thdpth.com/p/whats-wrong-with-data-teams?utm_campaign=post&amp;utm_medium=web">here</a>.</li><li><strong>Find likeminded colleagues</strong> — In public work, we don’t reduce burden of work through secretly harboring usage of AI. We do that meaningfully when we collaborate as a team. Your strengths, my weakness — all the same as before. We just have a new player entered the game.</li><li><strong>Remember your purpose —</strong> I’ve seen organizations with a <strong>strong mission</strong> and <strong>disciplined</strong> practices, as cringey or cliche it may sound, strive for greater accomplishments because the employees understand the expectation and where they fit in to influence the process.</li><li><strong>Be the Advocate and the Critic — </strong>It feels like hypocrisy if you’re both an advocate and critic, but after much time using AI, that is my conclusion on where I land. AI can be right and still be unsafe. I have to remind myself that I have to critically think along with AI, not let it does all the critical thinking for me. I have to force myself to do the uncomfortable thing and not become an autopilot. I must be cautious even if I set up the framework to succeed with the new tools. I need to grow, so I need to put in the work. With growing pain, there’s a reward at the end. You’d have to be patient enough to see it.</li></ul><p>— — —</p><p>Speaking of growing pain, my personal AI writing coach told me to keep publishing more instead of striving for polished essays. I’m thinking about writing a series — “Working with Machines — Public Servant Edition”, focusing on <em>one situation, one tension, and one insight </em>at a time.</p><p>If you’d like to see that, clap, comment, and/or subscribe!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a327dffeb858" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How can Building Code Compliance be “Antifragile”]]></title>
            <link>https://medium.com/@yungcodes/how-can-building-code-compliance-be-antifragile-dce8cecfdb61?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/dce8cecfdb61</guid>
            <category><![CDATA[building-code-compliance]]></category>
            <category><![CDATA[building-code]]></category>
            <category><![CDATA[energy-code-compliance]]></category>
            <category><![CDATA[antifragile]]></category>
            <category><![CDATA[california]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Tue, 06 Jan 2026 00:36:25 GMT</pubDate>
            <atom:updated>2026-01-06T01:13:56.269Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Fair warning: this post is part rant, part thinking out loud, and part trying to make sense of how messy systems I spend way too much time thinking about can work.</em></p><p>Let’s be honest.</p><p>Building code compliance is not sexy.</p><p>It’s paperwork. Inspections. Missing forms. Someone forgot to put the CF1R on the drawings <em>again</em>. And somehow… this is how we’re supposed to protect public safety and energy goals.</p><p>So let’s talk about a simple idea called <strong>antifragile</strong> — and why it actually fits building code implementation <em>really</em> well.</p><p>And yes, this post is inspired by the book <strong>Antifragile</strong> by Nassim Nicholas Taleb on my Libby’s library shelf. The book introduces the idea that some systems don’t just survive stress and problems — they actually get better because of them. And if you’ve worked in construction before as a designer, contractor, or even as a homeowner or renter, you know that there’s plenty of problems and stress with construction to share all around.</p><p>The book isn’t about building codes or energy compliance, but it got me thinking about how this idea could apply to real-world energy codes, construction, and the way implementation systems can actually function.</p><h3>First: What Does “Antifragile” Even Mean?</h3><p>Most things fall into three buckets:</p><ul><li><strong>Fragile</strong> — breaks when stressed<br> (Glass. Eggs. Overly complicated compliance systems.)</li><li><strong>Robust</strong> — survives stress but stays the same<br> (A rock. A binder full of rules.)</li><li><strong>Antifragile</strong> — gets <em>better</em> because of stress<br> (Muscles. Immune systems. Good feedback loops.)</li></ul><p>Most code systems try to be robust.</p><p>Antifragile systems say, “<em>Stuff will go wrong. Let’s learn from it.”</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*VK3DX0mHKLQfQoxUs2vE0A.jpeg" /></figure><h3>The Problem: We Treat Problems Like Embarrassments</h3><p>In building code compliance, we often act like:</p><ul><li>Noncompliance = failure</li><li>Data gaps = mistakes</li><li>Different enforcement styles = inconsistency to eliminate</li></ul><p>So we try to:</p><ul><li>Average things out</li><li>Hide the messy parts</li><li>Pretend the system is cleaner than it really is</li></ul><p>That’s fragile thinking.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*fOEAJsOWBNzWc1gWJAHWyw.jpeg" /></figure><h3>Antifragile Thinking Flipped The Switch</h3><p>Instead of asking: <em>“Why isn’t everyone complying the same way?”</em></p><p>Ask: <em>“What is this mess trying to teach us?”</em></p><h3>What This Looks Like in Real Life</h3><p>An antifragile building code system:</p><ul><li>Expects mistakes</li><li>Tracks patterns, not just averages</li><li>Uses failures to guide training and policy</li><li>Focuses effort where things are hardest</li><li>Gets better every few years</li></ul><p>Not perfect. Not clean. But <em>stronger</em> over time. This sounds kind of fitting to building code implementation systems.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*NkDcHJ0x-IoKCc5yYeBLBQ.jpeg" /></figure><h3>How Antifragile Thinking Fits Energy Code Compliance in California</h3><p>California’s energy code system is big, complicated, and always changing.</p><p>We have:</p><ul><li>New code cycles every 3ish years</li><li><strong>Different</strong> rules for <strong>different</strong> buildings under <strong>different</strong> scopes</li><li>Over 500 local jurisdictions (I only learned about this growing number at the CEC)</li><li>Many people involved (Owner, leaser, lawyers, realtors, inspectors, plan checkers, designers, contractors, different trades, and energy consultants, etc.)</li><li>And a lot of paperwork moving between all of them</li></ul><p>Because of all these wonderful and natural combination, things <em>will</em> go wrong. That’s not a surprise. That’s normal.</p><p>Under Antifragile thinking, the simple idea is: Instead of trying to stop every problem, use problems to make the system better.</p><p>So I’ve been racking my brain on how this concept can apply to specific examples — selfishly — things that are applicable to what I’m privileged to work on every day currently. This is what I came up with.</p><h3>Example 1: Using Noncompliance as a Signal, Not Just a Mistake</h3><p>When a project doesn’t comply with the energy code, the usual response is simple: point out the mistake, fix it, and move on. The decision on how to “fix it” is mostly left to the local building department.</p><p>So what do building officials and inspectors naturally do over time?<br>If they keep seeing the same mistake show up again and again, they realize it’s probably not random. Once that pattern is clear, they start watching for those same issues in similar projects in the future.</p><p>Now imagine this.</p><p>Every correction — every correction slip — is recorded digitally, not just people’s brain and experience. Not as scanned handwritten notes or word docs with their own decorations, but as clean data in a table. Let’s go one step further and imagine that all corrections are captured digitally, even the ones inspectors talk through with contractors but don’t formally issue as correction slips.</p><p>At that point, you can clearly see:</p><ul><li>Which mistakes show up over and over</li><li>Which ones are rare or unusual</li><li>Where problems cluster by measure, building type, or project scope</li></ul><p>If you have, say, 70 or more best samples within the same scope, you can start to reach a reasonable level of confidence about what actually needs closer attention.</p><p>So what can you do with that information?</p><p>A lot.</p><ul><li><strong>At the state level</strong>, this feedback loop can highlight energy code requirements that are unclear or frequently misunderstood.</li><li><strong>For program implementers</strong>, repeated documentation issues can signal where training isn’t reaching the right people.</li><li><strong>For vendors and software developers</strong>, patterns may show where tools don’t match what actually happens in the field.</li><li><strong>For local jurisdictions</strong>, sharing this information openly with <strong>contractors</strong> can raise the bar, improve consistency, and make it easier for the workforce to succeed.</li><li><strong>For contractors,</strong> if they know the inspectors are going to look for something, they’ll try to get it right the first time.</li></ul><p>This approach turns noncompliance into useful information — about what’s happening, where it’s happening, and why.</p><p>And honestly, this isn’t some futuristic idea.</p><p>My doctor already records our conversations using AI tools, and that information likely ends up as rows in a database that helps improve care over time.</p><p>So the real question is: Why can’t building department staff— the doctors of our built environment — get the same support?</p><h3><strong>Example 2: Different Cities Enforce the Code Differently — and That’s OK</strong></h3><p>In California, energy code enforcement looks different everywhere. It took me some time to learn this and really experience it.</p><p>Some jurisdictions:</p><ul><li>Have dedicated energy specialists</li><li>Have strong plan review but limited inspections</li><li>Are short-staffed and stretched thin</li></ul><p>This difference is often treated as a problem to erase.</p><p>A fragile response is to force uniformity and tries to make every city enforce the code in the exact same way. The problem will never go away because real life doesn’t work that way. There are inherent inefficiencies in government (e.g. hiring, guardrails, blah blah bleh) that will not make that solution successful. When you force uniform rules onto very different local jurisdictions one or two things happen:</p><ul><li>The rules get watered down so every can “meet” them, or</li><li>Cities quietly work around them to survive</li></ul><p>Either way, you lose information about what actually works.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/575/1*vsvsbUzk6OeUyu0hyWHeJA.jpeg" /></figure><p>Antifragile thinking says:</p><ul><li>These differences between cities are <strong>real-world testing. </strong>Some cities enforce strictly and catch issues early. Some rely more on documentation. Some focus on inspections an riskiest measures.</li><li>Then we can ask: Which approach hold up when staff is limited? Which practices catches the most problems with the least effort? Where does it breakdown?</li></ul><p>Each city becomes a real-world experiment — without anyone having to design one on purpose.</p><p>Instead of forcing everyone into the same mold, the system can:</p><ul><li>Learn from what works in practice</li><li>Share effective approaches</li><li>Focus support where it’s needed most</li></ul><p>Variation becomes a <strong>learning tool</strong>, not a weakness. Antifragile thinking doesn’t pretend cities are the same. It uses their differences to make the system smarter.</p><p>Now, I know what you’re thinking. If different cities enforce the code differently, doesn’t that send the message that the rules don’t really matter — or that people can ignore them if they think the rules don’t apply to them?</p><p>Short answer: <strong>No.</strong></p><p>Antifragile thinking does <strong>not</strong> mean different rules for different cities.<br> It means different <em>paths</em> to the same rules. What changes is how enforcement works under real-world conditions.</p><p>What stays the same:</p><ul><li>The requirements don’t change by city</li><li>The performance targets don’t change</li><li>Noncompliance is still noncompliance</li></ul><p>What changes is <strong>how enforcement works under real-world conditions</strong>.</p><p>Some cities may:</p><ul><li>Focus inspections on the riskiest measures to the region</li><li>Rely more on documentation where staffing is limited</li><li>Catch issues at different points in the process</li></ul><p>It can feel unfair if enforcement doesn’t look the same everywhere. Antifragile thinking asks questions like if the same types of buildings are meeting the same standards over time, what are the highest risk issues being caught and corrected, and is compliance improving not just paperwork. Fairness comes from <strong>consistent expectations and accountability.</strong> We can’t pretend every city has the same resources. The pro of having a single central authority can speed up larger scale development. The pro of having decentralized authorities provide flexibility and support local workforce.</p><h3>Example 3: Paperwork Problems Show Where the System Is Fragile</h3><p>Energy code compliance in California relies heavily on forms, files, and handoffs.</p><p>When things go wrong, it’s often because:</p><ul><li>A form is confusing</li><li>A step is easy to miss</li><li>Responsibility isn’t clear</li><li>Too many people touch the same document or nobody touch the document they’re supposed to</li></ul><p>If lots of projects mess up the same step, that’s a sign the system is fragile there.</p><p>An antifragile approach asks: Can we simplify this? Can we remove steps? Can we make errors harder to make? Can we design an information system that self-identify mistakes?</p><p>It’s <strong>fixing the process</strong>, not blaming the people.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/544/1*X17SRK8D-er54dWKAjp1hQ.jpeg" /></figure><h3>4. New Code Cycles Are Stress Tests (and That’s a Good Thing)</h3><p>Every new energy code cycle puts stress on the system:</p><ul><li>New requirements</li><li>New technologies</li><li>New learning curves</li></ul><p>Each cycle becomes a chance to:</p><ul><li>See what requirements are hardest to implement</li><li>Identify where training falls short</li><li>Improve enforcement tools and guidance</li></ul><p>I’m still learning about the code process and to think in what way this stress test can be applied to make the next cycle better, but maybe I’ll defer it to the experts.</p><h3>Final Thoughts</h3><p>At the end of the day, antifragile thinking is really about one thing:<br> <strong>using stress to make the system better instead of pretending stress shouldn’t exist.</strong></p><p>Energy code compliance will never be perfect. There will always be mistakes, gaps, and messy handoffs between people and systems. That’s not a failure — that’s the reality of how construction and enforcement work in the real world.</p><p>It keeps the rules firm, but treats problems as signals instead of embarrassments. Over time, that leads to clearer requirements, better training, smarter enforcement, and a system that actually improves with each code cycle versus starting over with a new implementation strategy every few years as people find problems (which will never cease to exist).</p><p>On a more personal note, this way of thinking also hits close to home for me. It lines up a little <em>too</em> well with my high-functioning anxiety — always scanning for problems, always trying to stay ahead. Antifragile at least gives that instinct a better name than “just a coping mechanism,” and maybe even points it in a healthier direction.</p><p>So I’ll end with a question:</p><p>What would energy code compliance look like if we designed it to learn on purpose?</p><p>And can you think of other places — inside or outside building codes — where antifragile thinking could help systems get stronger instead of just more rigid?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dce8cecfdb61" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The lessons in maturing in my careers]]></title>
            <link>https://medium.com/@yungcodes/the-lessons-in-maturing-in-my-careers-f52bf53bf3b5?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/f52bf53bf3b5</guid>
            <category><![CDATA[career-advice]]></category>
            <category><![CDATA[professional-development]]></category>
            <category><![CDATA[spirituality]]></category>
            <category><![CDATA[careers]]></category>
            <category><![CDATA[life-lessons]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Wed, 08 Oct 2025 03:23:33 GMT</pubDate>
            <atom:updated>2025-10-08T03:33:16.781Z</atom:updated>
            <content:encoded><![CDATA[<p>The thing about writer’s block is that you feel so many words in you and you can utter none of it, feel proud in none of it, and hope for less of this silence.</p><p>To break this writer’s block in me and the hundreds of incomplete drafts, I’m trying something new. Write something. Anything.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*6F1KbpdYMaAmaCxgyHoXOg.jpeg" /><figcaption>I am going to start writing… starting now.</figcaption></figure><p>I’m close to <strong>13 years </strong>now since the inception of starting my career in energy efficiency, wait-no, maybe <em>engineering </em>— eh… well, close enough to math, reasonings, and practical related profession. October 31st, 2012. I still remember my first day at the very first internship I’ve prayed meticulously for.</p><p>Here are 13 hard lessons I learned that I’d have told myself at 22.</p><ol><li><strong>Perception and feelings dictate decisions.</strong> Reasons and evidence are just ways to justify a feeling — assurance. After many years as an analyst and chronic overthinker, I was always troubled by the art of decision-making and influence on an intellectual level. One day, I realized that most of us already carry a heavy mental load, especially after becoming parents, achieving important positions, or becoming significant to someone else. Making decisions is easier when you trust your gut rather than trying to navigate the noise. That’s also why people will believe what they want to believe. Don’t just learn logic and reasons; better understand persuasion.</li><li><strong>Fake it until you make it. Fine print: within a moral boundary. </strong>Speaking of being consultants, that’s one thing I learned I needed to do — learn how to become the expert quickly and appear that way. Treat it like a sport. You win some, you lose some. Some people take this too far, becoming fraudulent and fake. Don’t do that. Don’t become a shallow shell. That’s not a good example. The best value from this lesson is when it lulls your insecurity to sleep and helps you gain the confidence needed to deliver what was always there — the jewels you’ve been mining that truly matter and make way.</li><li><strong>Train yourself to be different versions of who you want to be</strong>. Early in your career, you’re likely trained to be detail-oriented and to excel at what you do. You might still feel mediocre, and that’s okay. That version isn’t meant to last forever. If you get promoted or find yourself in a higher position or a different role, you’ll likely acquire a new set of skills. Sometimes, these new skills can seem counterintuitive to those you’ve been honing. It’s helpful to train your mind to grow and adapt to changes. It’s uncomfortable, but you can train your mind. No one stays the same forever. Someone’s misjudgment of who you are means nothing more than a false impression. Don’t let anyone define who you can be, including yourself.</li><li><strong>Integrity is built over time and never leave you.</strong> Remember all the bridges you might want to burn in ten years. Is it worth it? Most of the time, no. Something about getting older makes meeting old acquaintances more frequent than you thought it would be. Aim to build relationships, no matter how small they might seem at the time or how annoying they feel right now because you’re rubbing elbows, rather than just being right. Your integrity and reputation are your aura and your shadow. With that said, don’t let any bully overstep your boundaries.</li><li><strong>You catch more flies with honey than vinegar</strong>. These are famous last words when you’re caught in the web of bureaucracy. Being adept at navigating bureaucracy means knowing when to use sweet words and when to cut through the nonsense. There are fewer tools to cut through the nonsense due to the numerous guardrails against power abuse and the layers of processes that protect a cog in the machine. This leaves well-intentioned, high-performing public servants with the task of giving themselves permission to let go of what they can’t control — other people’s actions and personalities. You can only do so much. If they don’t want the honey, well, vinegar is all that’s left. In these situations, my natural tendency is to add some salt to make my favorite flavor of chips: salt and vinegar. Don’t let them mistake your niceness for being a doormat.</li><li><strong>The squeaky wheels get greased</strong>. The loudest and those who speak most often are likely to get promoted. The most articulate person tends to be chosen for middle management roles because they seem knowledgeable. Again, it’s just statistics. Come to terms with that if you don’t like public speaking. Practice communication because it’s more efficient than spinning your wheels and wasting your precious time following those who can’t.</li><li><strong>Don’t make your job your whole personality</strong>. A plague that affected me earlier in my career was that too much of my identity was tied to my job. It harms your mental health in the long run more than anything. Mentally exhausting. Protect your energy. Do the things that make you feel like you by choice and find freedom, not because you’re supposed to.</li><li><strong>Knowing when to care — and when to let go. </strong>It’s not easy to talk about this openly, but over time I’ve learned that adjusting how much energy I invest in different situations is essential for both effectiveness and well-being. I’ve noticed — often through observing colleagues and even my own spouse — that some people approach work strategically: they focus on doing just enough to achieve the desired outcome, rather than striving for perfection in every task. It’s still very hard to control for me because the working ethics from my single mother who was killing herself working multiple jobs to 3AM throughout my childhood had a strong imprint on me and my sister. Deliver excellence when it truly matters or when it’s properly valued and compensated requires strong awareness. But don’t overextend yourself for things that have little appreciation. I’ve realized that caring too much can sometimes backfire — it can strain relationships, create unnecessary tension, or even trigger envy with your closest allies. Sometimes, the greatest gift you can offer — to yourself or others — is let them/you the opportunity to not give a damn and still get paid.</li><li><strong>Should you stay or should you go?</strong> The question that you’ll face many times in your career. Whichever path you pick, don’t look back.</li><li><strong>This too shall pass. </strong>Ah, the phase that makes a happy man sad, and sad man happy. Whenever something feels like it’s on fire, I think about the worst time in my life and remember that whatever it is — it’s likely pale in comparison. It’s funny when you see professionals give each other the middle fingers while being politically correct. Oh, you think you can mock me and talked about my appearance? Sorry, my Asian upbringing already prepared me for that. Oh, you think your little passive aggressive tactics are going to give me heartburn? Please, get in line.</li><li><strong>Serve yourself. Serve the people closest to you. Serve the mass. </strong>There are three types of people I met along the way. Most people work to serve themselves. Some people work to serve the people closest to them — their family, their team, and the people who are going to be in their future. And then, there are seldom a few who work to serve the masses, truly, not coming from some kind of ego-boosting bull crap kind of place. Strive to serve those around you when you’re starting out. When you meet the rare one that truly serve the masses beyond themselves, give me a call.</li><li><strong>You are your worst enemy, your worst critic, and — most importantly, remember — a child of God</strong>. And as a child of God, you should be treated with dignity and respect. Say sorry less. Say thank you for your patience instead. We are all learning to make mistakes and striving for more.</li><li><strong>One true act of kindness will spark another.</strong> In<strong> </strong>the web of complex relationships and politics, if you’re like me, you didn’t pick jobs that ends with “Engineer” or “Analyst” because you like the drama between actual people. They happen though. Looking back, sometimes I’m the righteous asshole because I also got an ego too. It grows and it shrinks, depend on the day. We all play a role in this giant play. Especially in the world of work, there isn’t a day where there isn’t a problem. It doesn’t matter how well you plan. Murphy Law of what can go wrong will go wrong is strong. It’s just statistics and probability when you deal with a lot of people and a lot of things to be done sequentially in a certain way, in a setting that hasn’t exactly been done the same before, there’s going to be issues as long as there are egos. That’s something I learned during my time in the Navy and at the California Energy Commission. As public servants, sometimes being a punching bag comes with the territory — especially in roles with higher visibility. It’s often a signal that something needs to be improved, and you’re there to help make it better, even if it means taking a few hits along the way. The light on the other side often hides within that mess — one true act of kindness, a few words of encouragement, can go a long way. There are days when you’ll feel jaded and defeated. Can you truly appreciate the light without first knowing the dark? Always remember to be that spark — the candlelight that rekindles kindness for someone who may be feeling bleak or lost in the dark. That small act can become a rope of hope that lifts you both up and out. Even years later, you’ll remember that spark fondly. I know I will have.</li></ol><p>Alright, writer’s block, are you still there?</p><p>Good. You’re getting sleepy?</p><p>Good.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f52bf53bf3b5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Expectation vs. Reality — Where careers are born to die]]></title>
            <link>https://medium.com/@yungcodes/expectation-vs-reality-where-careers-are-born-to-die-c048c21fdbce?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/c048c21fdbce</guid>
            <category><![CDATA[careers]]></category>
            <category><![CDATA[project-manager]]></category>
            <category><![CDATA[engineer]]></category>
            <category><![CDATA[career-development]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Mon, 03 Mar 2025 18:31:03 GMT</pubDate>
            <atom:updated>2025-03-03T18:52:14.725Z</atom:updated>
            <content:encoded><![CDATA[<h3>Expectation vs. Reality — Where careers are born to die</h3><p>Life update …</p><p>It has been several years since I took a pause mid-way into OMSA (Online Master of Science, Analytics) at Georgia Tech, a program ahead of its time in being the most affordable option of the most reputable universities you can find in the U.S.</p><p>Why — you asked? Was the program too hard? Did I get a stroke or an ailment so bad? Did I get weeded out?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/498/1*F5Sq-UtQs0Bye61Tiv68nA.gif" /></figure><p>No, bro. My 4.0 game was strong, alright. Some of the classes were hard, but most universities still grade on a curve, so really, it’s a game of efforts and how much you stood out among your peers who look a lot different than when you just graduated high school.</p><p>What is it then?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/220/1*ZcUY0XuEz0C9jWMPR08siA.gif" /><figcaption>A fork in the road… ? Too soon?</figcaption></figure><p>Simply a change in direction.</p><p>I pursued OMSA because I wanted to go beyond the little niche I was in. So, a specialist to a generalist with higher magical abilities. That was the path, but the wind blew elsewhere.</p><p>The wind was part environment, part will. I could’ve stayed the path. I took up a job as a Senior Project Manager in construction in a foreign country. And I became a kid in the sandbox. Then I join the motherhood sandbox. Then back to America where its vastness affords me the pause to think about where I want to go from here.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/902/1*YDGopEunpKPRyqeXWJEZkw.png" /><figcaption>The detours are part of the journey.</figcaption></figure><p>When I look at my peers who were in the same classes at the time, they’ve grown to become data scientists and active contributors to pretty neat projects on Github. While my activities were… behind secured walls for the job’s requirements and frankly — not data science-y.</p><p>It is not all bad though. This detour felt like 4-years away-from-home college I badly wanted and have gotten. I took the things I learned from OMSA and applied it straight away. So, if you’re taking “Business in Analytics” and dreading how is this going to be useful at all for the $1,000 something it costed — trust me, it all won’t be lost.</p><p>Here’s to better things… and unleash the reflections I held back in the last 5 years. More to come.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/400/1*1k-nOh-RSAlOhRKcbJ-_aw.gif" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c048c21fdbce" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Practice Markov Chain in Two Ways: Excel & Python]]></title>
            <link>https://medium.com/@yungcodes/practice-markov-chain-in-two-ways-excel-python-3edcef8b3c5e?source=rss-3c36767d8f45------2</link>
            <guid isPermaLink="false">https://medium.com/p/3edcef8b3c5e</guid>
            <category><![CDATA[excel]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[probability]]></category>
            <category><![CDATA[markov-chains]]></category>
            <dc:creator><![CDATA[Yung Codes]]></dc:creator>
            <pubDate>Sun, 05 Jul 2020 06:54:53 GMT</pubDate>
            <atom:updated>2020-07-08T23:43:46.936Z</atom:updated>
            <content:encoded><![CDATA[<h4>Simulate Markov Chain in Excel &amp; Python, using probability theories. Which way is easier and why?</h4><p>In this flash-card on Markov Chain, I will summarize a basic understanding of Markov Chain and show how to implement this method in the context of a simple problem using Excel and Python.</p><h4>Background: What is Markov Chain?</h4><ul><li>A stochastic (random) model that simulates the sequence of possible events where the next event is determined by the state of the previous event.</li><li>Note: “Memoryless” property of Markov Chains means a process for which predictions of the future outcome is solely based on its present state. So the past and future is independent of each other. However, the future is conditional upon the present state.</li><li>Discrete (finite state or space): Examples of discrete Markov Chain are random walks based on an integer.</li><li>Continuous (over time): Examples of continuous Markov Chain are Wiener process (Brownian motion) and Poisson process.</li></ul><h4>Motivation: Why should I care?</h4><ul><li>An important and fundamental concept that has many applications.</li><li>It sounds fancy and has historical significance. For example, Einstein&#39;s Nobel Prize can be inferred to relate to this probability method (via Brownian motion, Wiener process). You also use it every day whenever you google anything.</li></ul><h4>Application: When to use Markov Chain?</h4><ul><li><strong>Famous applications: </strong>Google uses Markov Chain to rank webpages. Heavily used in forecasting such as certain weather pattern and the stock market trading algorithm (Hidden Markov Model (HMM): assumes that there is another process Y whose behavior depends on X, an unobservable state), and more. Chances are your life’s savings that you threw into investment funds are utilizing automated trading using Markov Chain principles.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*gTr35hmf4CKgLl0fSryfDg.gif" /><figcaption>(Image by Gfycat) It’s a box of worms the more I learn about MC.</figcaption></figure><ul><li><strong>Built Environment applications:</strong> Transportation (traffic) forecasting, solar irradiance variability</li><li><strong>Building Energy:</strong> Used in occupancy simulation and predict energy usage — usually associate occupancy state (<a href="https://www.sciencedirect.com/science/article/pii/S0378778815002054">1</a>, <a href="https://link.springer.com/article/10.1007/s12273-011-0044-5">2</a>, <a href="https://www.diva-portal.org/smash/get/diva2:608393/fulltext01.pdf%7C">3</a>, <a href="https://www.mdpi.com/1996-1073/11/2/358/pdf">4</a>).</li></ul><h4>Practice: How can I apply the Markov Chain method in practice?</h4><p><strong>Simple Problem:</strong> Suppose the weather is divided into two states: “Sunny” and “Rain”. We will denote “<em>R”</em> as “Rain” and <em>“S” </em>as “Sunny”. We are given the probability matrix of the transition states:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/429/1*_2LFUF66wFgupQ6GmO0fLw.png" /><figcaption>(Image by Author)</figcaption></figure><p>Suppose we initialize the weather state on Monday or X_0 as “Rain”.</p><p><strong>Question:</strong></p><ul><li>What is the probability that it will rain on Wednesday, given that it is raining on Monday?</li><li>In other words, what is Pr(X_2 = ‘R’ | X_0 = ‘R’)?</li></ul><h4>Practice: Excel</h4><p>Logic used:</p><pre>P(R|Xi-1)== If(Previous Day R/S = &quot;R&quot;, 0.8 or R-&gt; R, 0.4 or S -&gt; R)<br>Ui == RAND()<br>Ui &lt; P0 == If (Ui &lt; P(R|Xi-1), “Yes”, “No”)<br>R/S == If (Ui&lt;P0 == “Yes”, “R”, “S”) <br>Count == COUNTIF(Range of R/S, “R”)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/887/1*QkrVXxC9DsWHOMhyjlGNQg.png" /><figcaption>(Image by Author) Implementing Markov Chain in Excel</figcaption></figure><ul><li><strong>What I learned: </strong>The trickiest part about implementing this in Excel was the sheer amount of columns for the X_n desired event (X_2 in our case).</li><li><strong>Pro:</strong> Excel provides a good drawing board for simple tests and helps visually map out the logic. Good for beginners.</li><li><strong>Con:</strong> If n is large for X_n, Excel computation time may be slower and cumbersome.</li></ul><h4><strong>Practice: Python</strong></h4><p>First, import numpy and input the 2 x 2 events.</p><pre>import numpy as np <br>import random as rm</pre><pre>#define the states<br>states = np.array([&quot;R&quot;, &quot;S&quot;])</pre><pre>#Position sequence of events<br>transition_name = np.array([[&quot;RR&quot;, &quot;RS&quot;],[&quot;SR&quot;, &quot;SS&quot;]])</pre><pre>#Probability matrix of transition states<br>P = np.array([[0.3, 0.7], [0.1, 0.9]])</pre><pre>try:<br>  sum(P[0]) + sum(P[1]) == 2<br>  print(&quot;Probability Matrix input looks good to go!&quot;)<br>except:<br>  print(&quot;Error: Check your Probability Matrix Input!&quot;)</pre><p>The try/except gatekeeper is kind of excessive, but it’s good to flex debug muscles and also points out that <strong>each ROW of the transition matrix needs to add up to 1</strong>. On the other hand, it doesn’t matter if the columns add up to 1.</p><p>Next, I built a function to evaluate the next state based on the current state and the state in question (what we care about).</p><pre>def get_next_state(current_state, question_state): <br>#Return next state</pre><pre>  #Generate random variable<br>  U_i = rm.uniform(0,1)</pre><pre>  #Get P(R|X_i-1)<br>  next_state = current_state + question_state<br>  P_index = np.where(transition_name == next_state)<br>  Pr_next_state = P[P_index[0], P_index[1]]</pre><pre>  #Condition U_i &lt; P(R|X_i-1)<br>  if U_i &lt; Pr_next_state:<br>     next_state = question_state<br>  else:<br>     next_state = states[states != question_state].item(0)</pre><pre>  return next_state</pre><p>Next, it’s time to simulate this many times.</p><pre>def get_prob_next_state(days, n_sample, initial_state, question_state):</pre><pre>  assert initial_state == &quot;R&quot; or initial_state == &quot;S&quot; , &quot;Need to     assert R or S for the initial state&quot;</pre><pre>  assert question_state == &quot;R&quot; or question_state == &quot;S&quot; , &quot;Need to assert R or S for the question state&quot;</pre><pre>  state_list = []<br>  i = 0<br>  j = 0</pre><pre>  for j in range(n_sample):<br>    current_state = initial_state<br>    for i in range(days):  <br>      next_state = get_next_state(current_state,question_state)<br>      current_state = next_state  <br>    state_list.append(current_state)</pre><pre>prob = state_list.count(question_state)/(len(state_list) + 1)</pre><pre>print(&quot;The probability of &quot; , question_state , &quot;on &quot; , days , &quot;day(s) since the initial day with condition &quot; , initial_state , &quot;, for &quot; , n_sample , &quot; samples, is: &quot; , prob)</pre><p>Now… let’s test it out with this:</p><pre>get_prob_next_state(2, 1000, “R”, “R”)</pre><p>The result:</p><pre>The probability of  R on  2 day(s) since the initial day with condition  R , for  1000  samples, is:  0.17882117882117882</pre><p>This shows a similar result from the Excel version. Here is the <a href="https://colab.research.google.com/drive/1Mz4RP2CfjZwAaM50c-uO4qpBSpN4X8cA?usp=sharing">Google Colab notebook</a>.</p><ul><li><strong>What I learned: </strong>Numpy array can be faster than Python’s <em>list</em>, especially easier to maneuver than “list of list”.</li><li><strong>Pro: </strong>Scaleable. Flexible to answer more than the original question.</li><li><strong>Con: </strong>Take more time to develop compared to the Excel tool initially. Easier to make mistakes, thus need to conduct tests to make sure the code performs as intended.</li></ul><h4>Verify with Probabilities</h4><p>Alternatively, you can also do a probability calculation in this problem. There are two possibilities:</p><ul><li>Mon (R) -&gt; Tues (R) -&gt; Wed (R) : (0.3) * (0.3) = 0.09</li><li>Mon (R) -&gt; Tues (S) -&gt; Wed (R) : (0.7) * (0.1) = 0.07</li></ul><p>Therefore, the total possibilities that it can rain on Wednesday is the sum (0.09+0.07 = 0.16). That is very close to 0.17 we got from our simulations.</p><h4>How to Level Up?</h4><ul><li>Try to solve more complicated examples (<a href="https://stephens999.github.io/fiveMinuteStats/simulating_discrete_chains_1.html#simulation_1:_3x3_example">1</a>, <a href="https://www.johnwittenauer.net/markov-chains-from-scratch/">2</a>, <a href="https://www.kaggle.com/tacyonlord/youtube-title-generator-markov-chains">3</a>).</li><li>Recreate interesting papers using Markov Chain (<a href="https://ieeexplore.ieee.org/abstract/document/6298069">1</a>, <a href="https://www.sciencedirect.com/science/article/abs/pii/S0306261917304737">2</a>, <a href="https://www.sciencedirect.com/science/article/abs/pii/S0378778811002556">3</a>).</li></ul><p>In this series, I’m going to write a flash-card version of interesting topics I learned while being a student at Georgia Tech OMSA. I find that writing these summaries to do “actively recall” (aka studying). Last update: July 2020.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3edcef8b3c5e" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>