<?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 trcull on Medium]]></title>
        <description><![CDATA[Stories by trcull on Medium]]></description>
        <link>https://medium.com/@trcull?source=rss-89c24437500a------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*BDNg6YxGqvn7uLMO.png</url>
            <title>Stories by trcull on Medium</title>
            <link>https://medium.com/@trcull?source=rss-89c24437500a------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 08:01:01 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@trcull/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[Dark matter as evidence of a floppy universe]]></title>
            <link>https://trcull.medium.com/dark-matter-as-evidence-of-a-floppy-universe-1b27111911c?source=rss-89c24437500a------2</link>
            <guid isPermaLink="false">https://medium.com/p/1b27111911c</guid>
            <category><![CDATA[dark-matter]]></category>
            <category><![CDATA[astrophysics]]></category>
            <category><![CDATA[space]]></category>
            <dc:creator><![CDATA[trcull]]></dc:creator>
            <pubDate>Sun, 14 Mar 2021 22:17:14 GMT</pubDate>
            <atom:updated>2021-03-14T22:17:14.741Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uRSxWXAqxK9HJbiYcE1cMQ@2x.jpeg" /></figure><p>I’m not an astronomer, by any stretch. But there’s this thought about the universe I can’t get out of my head, so I’m sharing it with you. And it has to do with gravity, the fabric of space-time, and dark matter.</p><p>So, first, imagine you put a bowling ball on your bed. No matter how tightly you tuck your covers, if you put a bowling ball in the middle of your bed, it’s going to create a divot. And if you roll the bowling ball around in your bed, that divot is going to travel with it…and it’s also going to leave a trail of slightly messed up, looser covers behind it. The reason it leaves a trail is because your covers aren’t perfectly elastic: when they bend due to applying force (in the form of setting a bowling ball on them) then they snap back partially, but not completely, to the shape they were before.</p><p>Now imagine that you set the bowling ball in the middle of a steel beam that’s spanning a gap. You can’t really see it with your naked eye, but the steel beam deforms juuuuuuuust a liiiiiitle bit from the weight of the bowling ball. And, when you take the bowling ball back off the beam, it snaps back to its original shape. Well, pretty much. It, also, isn’t 100.00000….% elastic, so it snaps back to its original shape, minus a miniscule, tiny, teeny, itty-bitty bit of deformity. If you take the bowling ball on and off the steel beam, like, a literal billion times, then you might deform it enough to see with your naked eye. Or, maybe, measure with sensitive instruments. Basically no material in the universe is perfectly elastic: if you deform it a little with force, then remove that force, it doesn’t snap back to 100.0000000…% of its original shape. Maybe the most elastic material in the universe snaps back to 99.9999…..% of its original shape. Close enough to be imperceptibly different from 100% elastic, but not exactly 100% elastic.</p><p>Ok.</p><p>So, we have this metaphor of space-time being a “fabric”, like your bed sheets. And we say that gravity is a force created by matter deforming that fabric, much like a bowling ball deforms your bedsheets. Well, matter moves around in space, just like that bowling ball moved around on your bedsheets. There is one critical difference between space-time and your sheets, though. When we talk about the fabric of space-time, we talk about it as if, when matter moves from one place to another, then the “divot” that it had created in its previous position moves with it, and the fabric of space-time that it leaves behind snaps back to a state of perfect zero gravity. As if it’s perfectly elastic and retains not a single trace of its previous deformity.</p><p>But, what makes whatever that “fabric” of space-time is made of so special? What if it isn’t perfectly elastic? When you disturb anything else in the universe, it does not snap back to exactly its original position. It retains some echo of its deformity. Why would whatever space-time is made of be different than everything else in the universe?</p><p>So, imagine space-time is at least a little less than perfectly elastic, like everything else in the universe. Let’s say, it’s 1 minus 10 to the negative eleventy-gazillion-trillionth away from being perfectly elastic. Well, after 13 billion years of the universe existing, with stars and planets and black holes moving around in it, and stars converting matter into energy in their nuclear furnaces, what if all that stretching and snapping back of space-time over 13 billion years left its fabric….floppier? 13 billion years is a very long time for even the tiniest deviation from zero to accumulate into something more interesting.</p><p>Well, if that happened, then you’d see something that looks like the effect of gravity without seeing…matter. What you’d be seeing is the “tracks” left behind in space-time from matter moving around and/or being converted into energy. Wouldn’t you? And isn’t that exactly what we call “dark matter”? Dark matter is simply seeing the evidence of there being pockets of gravity without those pockets interacting with matter in any other way. And those pockets are mostly, but not completely, evenly distributed across the universe, just like matter was, especially in the beginning of the universe.</p><p>This doesn’t feel as crazy as it sounds. I mean, we just recently actually measured gravity waves. What are gravity waves other than a more dynamic version of what I’m talking about? Something huge like two neutron starts collided, they stretched space-time, it tried to snap back, it overshot zero in the opposite direction, tried to snap back again, overshot zero again by just the tiniest bit less, etc, until the resulting “wave” hit Earth. Well, if gravity can exist in wave form, without there being matter directly attached to it, then why can’t it also exist in “divot” form? Or waves that are harmonically locked somewhere? Or everything in between?</p><p>Ok, I got that off my chest. What do you think? Any actual astronomers out there?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1b27111911c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The new doomsday climate models: an explainer]]></title>
            <link>https://trcull.medium.com/the-new-doomsday-climate-models-an-explainer-ff17a4aa2497?source=rss-89c24437500a------2</link>
            <guid isPermaLink="false">https://medium.com/p/ff17a4aa2497</guid>
            <category><![CDATA[climate]]></category>
            <category><![CDATA[climate-change]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[cmip6]]></category>
            <category><![CDATA[cmip]]></category>
            <dc:creator><![CDATA[trcull]]></dc:creator>
            <pubDate>Thu, 14 Jan 2021 17:50:54 GMT</pubDate>
            <atom:updated>2021-01-14T17:50:54.698Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SPITdwT-6KXbcPLJsdmtdw@2x.jpeg" /></figure><p>You might have noticed a bunch of news coming out recently about revised climate scenarios and the world being more screwed than we originally thought it was. Why is that all coming out now? Probably a lot of reasons, but one of the biggest is something called CMIP6.</p><h3>WTF is CMIP6, You Say?</h3><p>Several decades ago, in 1995, the climate science community realized it needed to coordinate its efforts around climate modeling so that experiments done in one laboratory (“what will happen to the Greenland ice sheet”) can be cross referenced with experiments done in another laboratory (“how does deforestation in the Amazon contribute to the growth of atmospheric carbon”) without having to do a bunch of converting and reinterpreting and messing with data to make the connections. Out of that desire was born something called the Coupled Model Intercomparison Project (<a href="https://www.wcrp-climate.org/wgcm-cmip">CMIP</a>) run by an international organization called the World Climate Research Programme (WCRP). CMIP has had several iterations over the years, but the three you’ll probably encounter the most are the no-longer-used <a href="https://www.wcrp-climate.org/wgcm/references/CMIP3_BAMS_2007.pdf">CMIP3</a> in ~2007 and the currently-used <a href="https://pcmdi.llnl.gov/mips/cmip5/">CMIP5</a> in ~2012 and the soon-to-be-used <a href="https://www.wcrp-climate.org/wgcm-cmip/wgcm-cmip6">CMIP6</a> which is just starting to come out in 2021 and will probably finish in the next year or so.</p><p>As parts of CMIP6 come out, some features of the data are already starting to make headlines, most notably the finding that global temperatures are probably going to <a href="https://www.carbonbrief.org/cmip6-the-next-generation-of-climate-models-explained">go up more than we originally appreciated</a> (called <a href="https://www.carbonbrief.org/explainer-how-scientists-estimate-climate-sensitivity">climate sensitivity</a>)</p><p>Anyway, in starting off my <a href="https://trcull.medium.com/day-zero-is-today-9ce8b97298b2">new venture</a>, I’ve been doing a lot of learning about CMIP6. There’s an incredible amount of data being exposed by thousands of talented researchers. And, in a business-thesis-validating revelation: it’s completely incomprehensible to normal people, even technically savvy normal people. So, I’m about to drop some knowledge to make it less confusing, right here, right now.</p><h3>CMIP Isn’t One Thing, It’s Lots of Things</h3><p>The first thing to know is that CMIP isn’t really one organization of researchers working on one thing. What it actually is is more than a dozen different laboratories around the world working independently on what they probably would have worked on anyway, but, in doing so, making some relatively standard base assumptions and reporting back in relatively standard formats.</p><h3>Climate Researchers Love Acronyms</h3><p>Scientists, as a species, love acronyms. International governing bodies love them even more. What do you get when you gather a bunch of scientists together into an international governing body? Mind-blowing acronyms.</p><p>Part of what makes understanding CMIP difficult is every last freaking thing is an incomprehensible acronym. Take the official “<a href="https://pcmdi.llnl.gov/CMIP6/ArchiveStatistics/esgf_data_holdings/">summary table of currently available data</a>” page. You might, rightly, assume you’d see something like “if you want to see projected rainfall through 2100, it’s here” kind of stuff, but, instead, what you get is this totally incomprehensible (to outsiders) vomit of acronyms:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_GXKS2kOdZdQrhsTyvZkAw.png" /></figure><p>Trying to click around and figure out what those acronyms mean immediately gets you mired in a vocabulary of innocent-sounding terms like “model” and “experiment” and “variable” and “activity” that you feel like you should know, but quickly realize you have no idea what they mean, actually. Pretty soon, in trying to answer your seemingly-simple question “what is the projected rainfall through 2100” you’re still…lost. So, I’ll try to save you the three solid days I spent orienting myself by breaking it down right here, starting with what “variable” means.</p><h3>The Thing You Actually Care About is A Variable</h3><p>You’re probably coming to this field with a particular question in mind. The example question I’ll use is “what is the projected rainfall through 2100”. In that question, the “variable”, in CMIP6-speak, is “rainfall”. Or, more specifically, the super-descriptively-named variable “pr” for “precipitation”. Not all data sets contain all variables. In fact, probably zero data sets contain all the variables. For that matter, any given data set will only contain a pretty small subset of all the possible variables. What are all the possible variables, you ask? Well, they’re in a super-easy-to-understand <a href="http://clipc-services.ceda.ac.uk/dreq/mipVars.html">official list of variables</a>, of course.</p><p>I’m kidding, they’re not easy to understand at all, unless you’re a climate scientist.</p><p>There are hundreds of variables. Most of them only make sense to climate scientists (e.g. “cfadLidarsr532” for “CALIPSO Scattering Ratio”). A few of them, you, as an amateur, actually do care about. Here are some of those more amateur-relevant ones you probably actually care about, to give you a head start:</p><p>What’s the temperature?…</p><ul><li>tas: near surface air temperature and</li><li>ta: air temperature and</li><li>tsl: temperature of soil</li></ul><p>Will it snow and can I ski?…</p><ul><li>snd: snow depth and</li><li>snc: snow area percentage</li></ul><p>How much carbon are we talking about anyway?…</p><ul><li>co2s: atmosphere CO2 and</li><li>co2mass: total atmospheric mass of co2 and</li><li>co2: mole fraction of co2</li></ul><p>Will my tomatoes have any water?…</p><ul><li>pr: precipitation and</li><li>wtd: water table depth and</li><li>dgw: change in groundwater and</li><li>mrtws: terrestrial water storage and</li><li>mrsos: moisture in upper portion of soil column</li></ul><p>Will California still be on fire?…</p><ul><li>fFireAll: how much carbon is going into the atmosphere due to fire from all sources</li></ul><p>Will my Alaskan hunting shack sink into the permafrost?…</p><ul><li>tpf: permafrost layer thickness</li></ul><p>…and even cool, unexpected stuff like</p><ul><li>flashrate: lightning flash rate and</li><li>fLulccAtmLut: how much carbon is going into the atmosphere due to land-use changes</li></ul><p>The values of variables are calculated from a particular run of an “experiment” through a “model”. So, let’s get into what a “model” is, first.</p><h3>A Model Is A Set Of Assumptions A Particular Lab Made When Running Simulations</h3><p>Like I’d mentioned before, dozens of different labs across the world are contributing data. The scientific community pretty much agrees, in broad strokes, that certain things are important in simulations. They all care about cloud cover, or changes in land use from forest to agriculture, or how much carbon the ocean can absorb. Where they differ, though, is in the details. How much, exactly, should we assume the average cloud reflects the sun? Should I also try to model leaves breaking down and releasing carbon into nearby water bodies? How many different layers should I divide the ocean and the atmosphere into? All of these differences are captured in different models.</p><p>I’m still learning about the different models, but here are some of the important-seeming ones if you’re hoping to narrow your focus a bit:</p><ul><li><em>The Australians:</em> They’ve put together a model called <a href="https://www.publish.csiro.au/ES/pdf/ES19035">ACCESS-ESM1.5</a> that incorporates the land, atmosphere, ocean and ice and all the cycles among them, including the interactive carbon cycle</li><li><em>The Do-It-Yourselfers:</em> Sponsored by the National Science Foundation in the United States, you can run climate models yourself if you have enough Linux machines and patience, using the <a href="https://escomp.github.io/CESM/versions/cesm2.2/html/introduction.html">CESM2</a> model.</li><li><em>The Europeans:</em> If you’re a Europhile at heart, you can use the <a href="http://www.ec-earth.org/about/">EC-Earth3</a> model from a consortium of European countries.</li><li><em>Good, Old NASA:</em> I have a soft spot in my heart for NASA, and if you do, too, then you want the <a href="https://data.giss.nasa.gov/modelE/cmip6/">GISS-E2.1-G</a> model from the good folks in the Goddard Institute for Space Studies.</li><li><em>Japan:</em> If you want to branch out of the Western world, which, frankly, feels overrepresented so far, then check out the <a href="https://www.jstage.jst.go.jp/article/jmsj/advpub/0/advpub_2019-051/_article/-char/en">MRI-ESM2.0</a> model out of Japan for a different perspective.</li></ul><p>Anyway, the main point to keep in mind as a relative amateur is that models are kind of like a software program: you put some data in and you get some data out. The data you get out is “variables”. The data you put in, loosely-speaking, is an “experiment”.</p><h3>An Experiment Is Like a Description of an Alternate World</h3><p>Climate researchers are constantly asking things like “what if this happened” or “what if that happened”. As in, “what if, just for lulz, we assumed that carbon in the atmosphere quadrupled instantly, overnight?” which is a real experiment referred to by the super-descriptive name “abrupt-4xCO2”. Want to know what all the experiments are? No problem! Check out the easy-to-read <a href="https://github.com/WCRP-CMIP/CMIP6_CVs/blob/master/CMIP6_experiment_id.json">official Github repo</a>. Too hard to read? The same thing was helpfully rendered into an easy-to-read <a href="https://wcrp-cmip.github.io/CMIP6_CVs/docs/CMIP6_experiment_id.html">HTML form</a>, too.</p><p>By now, you’ve probably figured out I’m always being facetious when I say “super easy to read”.</p><p>Non-climate-scientists mostly don’t care about most of these experiments. They’re conducted in order to help scientists find better insights, but most of them are not that helpful for “tell me what’s likely to happen in the real world that I actually live in.” Well, dear reader, I’ve extracted some of those more-useful-to-us experiments here to give you a head start:</p><ul><li><em>historical and historical-ext:</em> these are the “normal” experiments that most of the articles you read in the media are probably referring to whenever they’re talking about “this is what the normal temperature was like each year between 1850 and now-ish”. The only thing not in here is historical CO2 concentrations</li><li><em>esm-hist and esm-hist-ext:</em> these are the “normal” experiments that try to spell out how much CO2 was in the air between 1850 and now-ish</li><li><em>ssp119, ssp126, ssp245, ssp370, ssp434, ssp460, ssp585:</em> these are the experiments that try to predict the future under various scenarios like “we’re globally-coordinated eco-superheroes (ssp119)” or “Big Oil forces us to burn fossil fuel indefinitely like the crack addicts we are (ssp585)”. These are the experiments being referenced in the “this is how screwed we are” articles you’re probably reading on Vox</li></ul><p>For that last set of experiments, you probably noticed a naming convention of “ssp” followed by some numbers. That’s not an accident.</p><h3>CMIP6 Uses Standardized Societal and CO2 Assumptions for Predictions</h3><p>This is probably the coolest innovation to come out of CMIP: standardized future scenarios. So, instead of each lab making predictions based on slightly different values of “how does humanity respond to climate change” we’ve now standardized a universal set of possible scenarios. In CMIP5, things were easy and there was only one axis of assumptions called <a href="https://en.wikipedia.org/wiki/Representative_Concentration_Pathway">Representative Concentration Pathways</a> (RCP) which were represented with names like RCP2.6 and RCP8.5. With RCPs, lower numbers generally (but not exactly) meant “humanity pumps less CO2 into the atmosphere” and higher numbers generally meant “humanity pumps more CO2 into the atmosphere”. The media tended to describe RCP8.5 as “business as usual, worst case scenario” and tended to describe RCP4.5 as “we get our shit together relatively soon”.</p><p>Well, it turns out it’s not that simple. So, in CMIP6, the scientific community decided to create a two-dimensional grid of “what’s happening in the atmosphere”, which was always represented by RCP, and crossed it with “how is humanity accomplishing that”, now represented by the new <a href="https://en.wikipedia.org/wiki/Shared_Socioeconomic_Pathways">Shared Socioeconomic Pathways</a> (SSP) classifications.</p><p>SSP1 roughly means “we all hold hands and do the right thing together”. Frustratingly, unlike with RCP, you can’t assume a lower number (i.e. “1”) is “worse” or “better” than a higher number. There are five SSPs, roughly described as:</p><ul><li><em>SSP1, Sustainability:</em> we all hold hands and do the right thing together at a sustainable pace while living in yurts</li><li><em>SSP2, Middle of the road:</em> we pretty much kind of lurch toward the right thing, incrementally, and at varying paces across countries</li><li><em>SSP3, Regional rivalry:</em> this is kind of the worst case scenario, where nationalism takes hold and everyone does the selfish, short term thing in isolation. Kind of a <em>Mad Max Fury Road</em> scenario.</li><li><em>SSP4, Inequality:</em> the rich countries do the right thing, but the poor countries are screwed. Kind of the <em>Hunger Games</em> scenario.</li><li><em>SSP5, Fossil-Fueled Development:</em> we all buy the fantasy that we can innovate and carbon capture our way out of this mess without giving up our fossil fuels. Global capitalism rules the day. Kind of the <em>Minority Report</em> scenario.</li></ul><p>So, under CMIP6, if you want to talk about the future you have to talk about an intersection of one SSP with one RCP. There are 5 SSPs and 7 RCPs, for a total of 5 x 7 = 35 different futures to talk about. However, some of those intersections obviously don’t make sense. For example, it’s very unlikely that SSP1 x RCP8.5 will happen, so it’s not worth talking about it. Likewise, only in the Koch brothers’ wildest fever dreams is SSP5 x RCP1.9 ever going to happen. So, CMIP spells out <a href="https://www.researchgate.net/figure/SSP-RCP-scenario-matrix-illustrating-ScenarioMIP-simulations-Each-cell-in-the-matrix_fig1_301594661">four “tier 1” combinations</a> of the two that they encourage people to use:</p><ul><li>SSP5 / RCP 8.5: we’re so fucked it’s actually quite unrealistic, but is still a useful upper bound on fuckedness</li><li>SSP3 / RCP 7.0: the more realistic “we’re fucked” scenario that best represents “business as usual”</li><li>SSP2 / RCP 4.5: the more realistic “we get our shit together, pretty much, but not quickly” scenario</li><li>SSP1 / RCP 2.6: the barely-realistic, optimistic scenario, pretty much tracking to the now-seemingly-unachievable Paris Climate Agreement</li></ul><p>So, there you go. A guided tour of what’s actually a very complicated subject. You must feel like you have all the tools you need to understand climate change’s impact on your organization, right? Yeah…probably not. But, fear not, here at <a href="https://pollen.io/">pollen.io</a> we’re working on an even more comprehensive guide (plus some helpful tools) to make this all even easier for non-climate-scientists to use.</p><p>See you soon!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ff17a4aa2497" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Day Zero Is Today]]></title>
            <link>https://trcull.medium.com/day-zero-is-today-9ce8b97298b2?source=rss-89c24437500a------2</link>
            <guid isPermaLink="false">https://medium.com/p/9ce8b97298b2</guid>
            <category><![CDATA[climate-change]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[career-paths]]></category>
            <dc:creator><![CDATA[trcull]]></dc:creator>
            <pubDate>Tue, 05 Jan 2021 23:12:06 GMT</pubDate>
            <atom:updated>2021-01-05T23:14:23.380Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QLfQr0wnjIlGyunp0SjUMA@2x.jpeg" /></figure><p>Life took an unexpected turn yesterday when I lost a really good job opportunity that I thought had been a sure thing.</p><p>It’s good, it’s bad, it’s good-and-bad. But, in the end, I think the universe was nudging me to the place where I really needed to be, and it’s all going to be good. Much more good than bad.</p><h3>The Backstory</h3><p>Almost two years ago now, I left a really solid, very well-paying job as an engineering director at Adobe. It was kind of a big leap for me and my family, and I gave up 70% of my income in order to join the fight against climate change.</p><p>At the time, I knew I had two possible paths: 1) directly contribute to battling climate change itself in some technical sense, or 2) fight the GOP’s active obstruction of applying what we already knew about fighting climate change. On both paths, I’d always assumed I’d be starting my own company, like I’d already done once before, with a fair bit of success, back in 2009.</p><p>I decided on path #2: fighting the GOP. It didn’t make sense to build technical capability when the deployment of that capability was being so actively hindered in the policy space. And, as luck would have it, I stumbled directly into running engineering at <a href="https://swingleft.org/">Swing Left</a>. My plan was, I’d fight the good fight there until the election, and then, after our sure victory and the removal of political barriers, I’d then start my own climate startup in November 2020, and dedicate myself to that for the remainder of my career.</p><p>The plan worked reasonably well.</p><p>Well, at least we won, <a href="https://trcull.medium.com/our-uncomfortable-look-at-our-own-privilege-e6d9391e8e45">barely</a>, and we’ve removed Trump from office. The path might not be as clear as I’d like in the policy space, but it’s a heck of a lot clearer than it was before. And I feel like I’ve done my tour of duty there for now, in any case.</p><p>So, after the election, it was time to execute Part 2 of the plan: start my own climate startup. I had a great idea, and the expertise to pull it off, and I was ready to go.</p><p>Then, a friend/former client of mine called me and asked if I was game to run engineering at his startup. And that caused a bit of a crisis that only resolved itself yesterday.</p><h3>That Startup</h3><p>You see, this friend/former client of mine is someone I liked working with A LOT. And another friend/former client of mine was also working at that same startup and I liked working with that person, A LOT. And the company was in a sweet spot and well funded and had hit product-market fit. And what they needed was exactly what I’m very good at. Basically, it was an absolutely, positively, once-in-a-career confluence of all that’s good about my profession. It was extremely tempting. This company was literally the only non-climate company in the world I’d consider pausing my own startup ambitions to join. And they called me of their own volition, out of the blue. The big problem? It had nothing to do with climate.</p><p>I lost a lot of sleep over this decision. It was extremely difficult. In the end, for reasons I might describe in another post at some point, I decided to go with the friend’s startup and ditch my own.</p><h3>Oops, Wrong Choice</h3><p>My first problem was I thought this was a done deal. We’d even talked numbers and they’d called my references, back in early November. And my friend seemed to have thought this was a done deal, too. But…it wasn’t. I make a religion, in business, of never assuming a deal is done until the ink is dry. But this one felt so done that I forsook that usual tendency and relaxed as if it were, in fact, a done deal.</p><h3>Narrator: It Wasn’t a Done Deal</h3><p>There was always one more thing, one more phone call, one more minor hurdle, until it became pretty clear around Christmas that I was irrecoverably screwed. And my poor friend had to call me yesterday to give me the final bad news: they were going to pass on me.</p><p>It’s not important why this fell apart, specifically, except that you, dear reader, know that it was mostly them and mostly not me :)</p><h3>This Is a Blessing</h3><p>And now, three months later than I’d originally intended, I get to go back to my own idea. It has to do with leveraging what I know about big data from running the big data engineering team for a product at Adobe and what I know about easy-to-use data tools from both Adobe and from Swing Left, and deploying that fairly specialized knowledge to make climate data usable for everyone.</p><p>I. Am. So. Excited.</p><p>And scared.</p><p>And embarrassed.</p><p>And that’s ok. In fact, it’s better than ok. The universe gave me a little nudge to say: “You were trying to take that smoother-looking path. I understand, but it’s wrong for you. Your destiny lies here and it’s going to be amazing.”</p><p>And the universe is right. Thank you, universe. And stay tuned.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9ce8b97298b2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Our Uncomfortable Look at Our Own Privilege]]></title>
            <link>https://trcull.medium.com/our-uncomfortable-look-at-our-own-privilege-e6d9391e8e45?source=rss-89c24437500a------2</link>
            <guid isPermaLink="false">https://medium.com/p/e6d9391e8e45</guid>
            <category><![CDATA[election-2020]]></category>
            <category><![CDATA[democrats]]></category>
            <category><![CDATA[privilege]]></category>
            <dc:creator><![CDATA[trcull]]></dc:creator>
            <pubDate>Thu, 05 Nov 2020 00:35:43 GMT</pubDate>
            <atom:updated>2020-11-05T00:37:24.304Z</atom:updated>
            <content:encoded><![CDATA[<p>None of the BIPOC people I know are shocked today, but every, single white person of privilege I know (including me) is shocked today.</p><p>What we’re seeing is a contrast of expectations.</p><p>The privileged (mostly) white people who form my social and professional orbit wanted (NEEDED, actually) this election to be a sweeping repudiation of Trumpism, as if Trumpism were a phenomena limited only to Trump himself. We wanted it to be a cleansing bath that proved, clearly, decisively, that he (Trump) and it (Trumpism) were a fluke that popped into existence in 2016, like magic, and could pop out of existence just as easily, like the anomaly we wanted it to be. We really needed it to not be our fault. Poof, see! We defeated Trump and his enablers in the Senate! It was a resounding landslide and now none of the sins of the last four years are our fault!</p><p>The problem is, many of those sins actually are our fault.</p><p>By the definition of us being privileged, we’ve benefited from the very system that created Trump and Trump’s adherents. That means, logically, in order to dismantle Trump, we have to also dismantle our own privilege, too. What’s making today so uncomfortable is that it’s one of those very few days where it’s impossible to overcome the cognitive dissonance created by these two, currently-uncomfortably-visible facts:</p><ol><li>Slightly under a majority of America has watched Trump be Trump for four years and has re-affirmed, with its eyes open, that they’re perfectly ok with it, simply because he promises to dismantle privilege.</li><li>Our current neoliberal economic and social system laid down, initially, by Clinton in the 90’s (and strengthened during a decades-long tech boom) has benefitted us (privileged, mostly-white people) at the expense of both BIPOC and most people in #1</li></ol><p>Why is there cognitive dissonance there? Because we hate #1 (for good reason) and the only way to fix #1 is to dismantle #2. But we benefit from #2, so we don’t want to dismantle it. And, boom, you have basically the textbook definition of cognitive dissonance. And it’s <em>extremely</em> uncomfortable.</p><p>BIPOC, on the other had, have been living actual reality for basically forever. The neoliberal system set up in the 90s never benefitted them. In fact, it harmed them. Income inequality has grown in the last 30 years, not shrunk. Racial inequality has grown in the last 30 years, not shrunk. Police brutality today feels no different than it did in the 90s. The war on drugs has imprisoned record numbers of BIPOC for crimes their white counterparts openly flout in the street. The system knocks on their door once every four years asking for their vote, and then the system goes right back to grinding them down until the next four years cycle back again. And so did the system before it. And the system before that.</p><p>In fact, if it weren’t for the naked white supremacy in the Trumpist camp, BIPOC would actually have far more in common with the lives of anti-establishment Trumpists than with the neoliberal, privileged, actual-establishment that claims to represent them. And they know it. This is a major miss on the part of Trumpists, actually. And one Trumpists are slowly starting to rectify, as is evidenced in Miami-Dade county.</p><p>In short, even though they generally hate Trump as much as I do, BIPOC aren’t nearly as shocked as I am by the outcome of this election because they saw the system for what it was. And they’ve been living this struggle for generations. And they were not harboring illusions about the system dismantling itself for their benefit, so they have very low expectations. Their expectations were fairly simple: they were throwing in their lot with those who at least didn’t oppress them loudly and on purpose vs those do who.</p><p>I, on the other hand, only just now realized that what just happened to us is that the best, kindest, most well-meaning and honest personification of the establishment that has benefited me for decades came up to the plate, took a big swing, and….only juuuuuust barely connected for a base hit. And, not only that, he was batting against the absolute worst, fourth-string, off-the-injured-reserve, frankly-not-very-smart pitcher that the other team had. And it’s really hard to watch. And it’s impossible to ignore, any longer, what it means.</p><p>What it means is that we have a system of three groups, with one group in charge that is playing the other two groups against each other. And those groups are:</p><p>Group E: The (mostly white) still-privileged</p><p>Group R: The (mostly white) recently de-privileged</p><p>Group D: The (mostly not white) never-privileged</p><p>The truth we’re being forced to see right now is that Group E runs both major parties and, depending on what party we’re talking about, tries to preserve its privilege at the expense of either Group R or Group D.</p><p>I am a member of Group E.</p><p>Every one of those three groups, of course, has decent people in it and horrible people in it. I genuinely think I’m one of the decent people. There’s nothing, however, about personality or morality that defines their membership. Literally the only two things that define their membership are: a) if they have privilege now and b) if they used to also have privilege before.</p><p>The true power struggle is and always has been happening inside Group E, alone, because that is where the power resides, by definition. Factions develop inside it (Democrats vs Republicans) and then each faction tries to generate advantage and create the illusion of progress by taking from Group R/D and giving to the opposite Group R/D, never, of course, having ever taken from itself.</p><p>Sometimes the mixture of basically-good people and basically-evil people inside those factions gets out of balance. I happen to think that right now is one of those times: Republicans (in power) are basically, nakedly, evil and Democrats in power are basically, blithely, not-evil. But they all, at their core, came from the same group: the (mostly white), still-privileged Group E.</p><p>The true tragedy is that the members of Group R and Group D don’t realize they’re being played by a common enemy. And that, if they joined forces, they’d actually be far, far stronger than that enemy. And, for that matter, most members of Group E don’t even realize they are that enemy because realizing that’s so would cause too much cognitive dissonance. So we all keep generating the collective fiction that the struggle is between Democrats and Republicans, when the real struggle is actually between Group R/D and Group E.</p><p>Except on days like today. On days like today, when we see people like Latinos in Miami siding with Trumpism. When we see black people staying home rather than vote. When we see decent, gosh-darn midwesterners siding with Trumpism. When we see people who should know “their own self interest” voting against it, then the matrix starts to break down. And we see the truth underneath. And it’s shocking. And our minds want to shut down rather than deal with the dissonance of it.</p><p>But reality, being what it is, stubbornly refuses to go away. And, the reality is, until we in Group E start genuinely sharing our own bounty with others, then this system is unstable. And, like all unstable systems, it’s doomed to fail. And, eventually, Group R and Group D are going to realize they have common needs and a common barrier to fulfilling them. And, when that time comes, it’s anyone’s guess where that goes. But I’m sure it’s not pretty. And in my very most honest moments, usually alone, when I’m able to see my distress for that it is, then that is what I fear: the great, inevitable, and very-not-pretty reckoning of that imbalance, at my expense.</p><p>Unless we find another way. Soon.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e6d9391e8e45" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[You are worthy of being loved.]]></title>
            <link>https://trcull.medium.com/you-are-worthy-of-being-loved-222cf1a2f8e8?source=rss-89c24437500a------2</link>
            <guid isPermaLink="false">https://medium.com/p/222cf1a2f8e8</guid>
            <category><![CDATA[christianity]]></category>
            <dc:creator><![CDATA[trcull]]></dc:creator>
            <pubDate>Sun, 23 Jun 2019 02:17:12 GMT</pubDate>
            <atom:updated>2019-06-23T02:19:57.894Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*S1thLNe2bXmNK3yRuZkgsA.jpeg" /></figure><p>I had a challenging conversation with my 16-year-old son last night.</p><p>It’s all Denzel Washington’s fault. Or, more specifically, Denzel’s Christian-propoganda-disguised-as-post-apocalypse-thriller (but, nonetheless quite good) movie <em>The Book of Eli</em>’s fault.</p><p>The fact that I had to say “but, nonetheless quite good” tells you most of what you need to know about the conversation. But I’ll describe it anyway.</p><p>The movie is cinematically gorgeous. The action is measured and exciting. The dialog is sparse and poetic. The plot is inspiring, the costuming is beautifully convincing and the scenery stark to the point of hypnosis. My son came into the room in the middle of the movie and saw…none of that. Instead, he keyed into the religious undertones immediately (and they weren’t that heavy handed, really) and dismissed the entire thing with “ugh, I hate religion. It’s awful, am I right?”</p><p>Ehhhhh….no? Well, yes. But, I wish, no?</p><p>Western Christian religion as expressed by (most notably) fundamentalist Christians on the Religious Right? Horrific. Hateful. Everything-phobic. Vaguely racist. Full of exclusion, privilege, money-worship and ignorance. Morally cherry-picked. Piousness of convenience. Selfish and self-serving. Bankrupt. Essentially, the worst PR job Jesus has had in two thousand years, including the era of the early Romans who crucified him.</p><p>Christian philosophy as expressed by Jesus Christ himself? That’s a much better story. And one that isn’t getting the airtime we all probably need.</p><p>Our household, for as long as our children have been alive, is unapologetically and firmly atheist. Not agnostic, nor “spiritual but not religious” nor any of those other non-committal pseudo-religions that modern Americans are fond of making up. Athiest. Nothing-but-the-void-when-you-die Athiest. And it’s a badge that my son, especially, wears proudly. But our household wasn’t always that way.</p><p>I grew up actively, observantly, Catholic. I was an altar boy. I played guitar in the church choir <em>for years</em>. Sunday school. Youth group. First Communion and Confirmation. Spring carnivals and casino night in the sanctuary. Soup kitchens and spaghetti feeds and collecting cans for the poor. For a moment there, I even considered the seminary (thank God I knew, even then, there was no way I could live a life of celibacy). And it was wonderful. I loved my church and it loved me back. We went through three or four pastors in the twenty years or so I went there and all of them had the same, consistent message: God is love, serve your fellow man, seek goodness and practice forgiveness.</p><p>That’s it.</p><p>They didn’t talk about The Gays or about politics or about abortion or about literally any of the holy hand grenade issues the Religious Right has co-opted to sow division and shore up its own power. They talked about Love. With a capital “L”. And all of its manifestations. Just like Jesus did.</p><p>So, suffice it to say, I didn’t lose my religion because I found the Church hateful. In fact, quite the opposite. In the Church I felt the perfect love of a benevolent, perfect, and omnipresent parent figure.</p><p>Realizing that parent figure didn’t really exist remains the second most traumatic experience of my life. In fact, I might even articulate it another way: murdering that parent figure with my own mind for no other reason than, well, reason, remains the second most traumatic experience of my life.</p><p>So it made me deeply sad and angry that the jackasses currently representing Christ the loudest have ruined him for my son. I wanted him to know the Christ I knew, even if neither of us is buying that it’s actually possible to rise from the dead.</p><p>So here, son. Here’s what the Jesus of my youth taught me. The real Jesus. The one described by John when he said “let us love one another, for love comes from God. Everyone who loves has been born of God and knows God. Whoever does not love does not know God, because God is love.” Here’s what religion is theoretically capable of, if not kidnapped by the cynical and powerful for their own gain. Here’s what I wish you could feel with the certainty I once felt. Please know that Christian philosophy, as actually expressed by Jesus, didn’t give a shit about the gays or about patriotic purity tests, or about which political party you belong to. Please don’t throw away what Christianity has to say just because its loudest, most contemporary adherents are so hateful and ignorant. And because it literally believes in ghosts.</p><p>You know what Jesus did have to say?</p><p>He said you are worthy of being loved. You. And every last person who ever lived on this Earth or ever will live on this Earth. Every. One. All of them. Any exceptions? Nope. None.</p><p>He said that giving is a virtue. In fact, it’s expected. You have to do it. You have to give until it hurts and then you have to give some more. And giving is, itself, a gift. Jesus required every one of his apostles to give away every possession they owned (some of them were quite wealthy before that). Do you think that was a coincidence? They just didn’t have room in the rowboat? Nope. He wanted it that way. He thought that possessions ruined people and that giving was an expression of God’s love. Do today’s bombastic, mega-church Christians gloss over that little, inconvenient fact? You bet your ass they do. But they’re wrong.</p><p>Jesus charged us with being caretakers of God’s creation. Caretakers. Not masters, not exploiters, not subjugators, but caretakers. It’s a position of service, not a position of mastery. Jesus was the original tree-hugging environmentalist before hippies made it fashionable. We are utterly failing at this task. And he would be very disappointed, starting with those who invoke his name to justify their immoral harvests.</p><p>And, guess what? He said <em>we</em> are all part of God’s creation. So, by extension, he also charged us with being caretakers of each other. Again, not masters. Caretakers. Lovers. Let-he-who-is-without-sin-cast-the-first-stone advocates for the persecuted, misunderstood and maligned. Jesus <a href="https://www.joe.ie/news/broadcaster-eli-bosnick-with-a-far-more-humane-skittles-analogy-561050">would stuff himself with Skittles</a>. If Jesus were alive today and in America, he’d be at the border springing babies from cages and then BURNING THOSE CAGES DOWN SO NOBODY IS IMPRISONED IN THEM AGAIN. Jesus hung out with lepers, adulterers, foreigners and debtors. He protected them and lifted them up and saved them from harm. You know the only people he ever really had harsh words for? Bankers. And even them he loved anyway.</p><p>Basically, he taught us that we are expected to do good things. We are accountable.</p><p>And, in exchange, we got to feel, like, deeply, that we will never be alone. Ever. He taught us that God is always with us. We got to know, to the bottom of our heart, that there will be “someone” who is there for us, even in our darkest hour. That, just like Denzel’s character quoted from the Bible in the movie last night, “yea, though I walk through the valley of the shadow of death, I will fear no evil: for thou art with me.” I wish you knew what that felt like. I’m actually getting choked up just from writing this paragraph, that’s how powerful that feeling is. I knew what it felt like, once. I can’t imagine navigating adolescence without that knowledge. It must be hell. I wish I still knew what that protection felt like, myself, even as a grown-ass man. It’s now-absence is a hole in my spirit that cannot be filled. I hope you’re finding an adequate substitute to keep the darkness away. I hope I can help in some small way.</p><p>So, yeah. I wish I knew a way. I wish I knew how to teach the real lessons of Christian philosophy without having to evoke burning bushes and literal ghosts and omnipresent deities that somehow, as if in a cruel joke, only showed up in our consciousness in the last 0.0001% of the time that humanity has been on Earth (and only to Western people’s consciousnesses). I wish I knew how to give you, son, that peace. And I wish so many people in the news today weren’t ruining it utterly for you.</p><p>Also, I really do recommend <em>The Book of Eli</em>. It’s quite good. Please give it a second chance. Also, if you’re going to read the Bible, start with Matthew or John. Not Samuel or Genesis.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=222cf1a2f8e8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[This is exactly how to interview with me]]></title>
            <link>https://trcull.medium.com/this-is-exactly-how-to-interview-with-me-713462ea6bf2?source=rss-89c24437500a------2</link>
            <guid isPermaLink="false">https://medium.com/p/713462ea6bf2</guid>
            <category><![CDATA[hiring]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[interview]]></category>
            <category><![CDATA[jobs]]></category>
            <dc:creator><![CDATA[trcull]]></dc:creator>
            <pubDate>Wed, 20 Jun 2018 06:24:24 GMT</pubDate>
            <atom:updated>2018-06-20T06:24:24.221Z</atom:updated>
            <content:encoded><![CDATA[<p>I’ve been asking the exact same interview question for 15 years. This gives me an absolutely enormous (say, 500–1000) sample size of answers to compare to and a decent idea of where the applicant pool has been going over time. I still see novel solutions to the problem, even today, but, I’ve got to say, the quality of answers to this question has taken a steep dive over the last five years and I don’t know why.</p><p>Here’s the question: “Pretend I own a chain of hotels and I’m hiring you to write a hotel reservation system. Imagine there are no databases; everything the system does has to be done as the interactions of classes and data structures in memory. And you have infinite memory. What would the design or class diagram of this system look like?”</p><p>That is literally, verbatim, what I ask. I’ve asked it hundreds of times. I’ve seen some truly brilliant designs come out of this question. But, much more often, I’ve mostly seen horrendous-to-mediocre solutions come out of it. Interestingly, seniority seems to have essentially zero correlation one way or the other with the elegance of the candidate’s design.</p><p>The first part of this interview session is the only “trick question” part of the interview. I fully recognize that this question is somewhat ambiguous, exactly the way that business requirements from stakeholders are ambiguous. So I expect a very strong candidate to ask clarifying questions before just diving into a design. The very best always do. But most people don’t.</p><p>The very best kinds of questions are simple things, like:</p><ul><li>How many hotels do you have? How many rooms do they have?</li><li>What kinds of rooms are in these hotels?</li><li>Do I need to support different rates for different rooms or time periods?</li><li>Do I need to support overbooking the hotel?</li><li>Do rooms have things like optional amenities? Different bed sizes?</li><li>Are guests allowed to reserve specific rooms, or do they need to just reserve a type of room and we’ll figure out which one, exactly, when they arrive? (This last question is very, very important. We’ll get to that later.)</li></ul><p>Everyone has stayed in a hotel at least once. Hell, some of these candidates are staying in a hotel we put them up in to get to this interview. So, the concept of a hotel reservation system should not be some kind of special domain knowledge. If you don’t ask any clarifying questions then I let you flounder around for about five minutes and then I bail you out by saying:</p><p>“Let’s focus on one use case. Say someone calls up and asks if I have one, queen-sized, non-smoking bed for July 1 through July 4. How do the classes and data structures in the system interact with each other in order to answer this question?”</p><p>This is the end of the trick questions. Absolutely everything I say (including this question) from this point forward is an attempt to help you perform better. No, really. I think trick questions are lame.</p><p>From here on out, all I really care about is seeing you reason through all the tradeoffs inherent in this problem. Will you keep track of rooms that are booked, or rooms that are <em>not</em> booked? Why did you choose one over the other? Did you store reservations in a data structure that’s optimized for writing or optimized for searching? Why? Will you even bother to keep track of reserved rooms at all, or will you keep track of reserved room <em>types</em> and figure out the exact room later? Do you seem to understand that, for a computer, iterating through a list of only 500 items in memory is basically the same thing as iterating through a list of 2, performance-wise?</p><p>Historically, I used to get into really interesting discussions about these tradeoffs. These days, many of the candidates I’m interviewing don’t even understand that these tradeoffs are there and don’t have a clue what I mean when I ask about them. For that matter, literally 50% of the candidates I interview these days leave the concept of dates out of their answer entirely on the first go. Seriously. They design a reservation system that only understands what’s happening <em>right now</em>, not in the future or in the past. Like a big, dopey yellow lab catching a frisbee, their software lives perpetually in the moment, blissfully unaware of who’s going to arrive next, or when.</p><p>Anyway.</p><p>A really good candidate will tend to quickly create a solution that’s functionally correct and semi-efficient-ish, but has some room for optimization. That’s ok! In fact, I get excited if there’s room for optimization because that means we get to talk through the problem together and I get to see how you think.</p><p>Let’s take this straw-man solution, for example. Say we have one array containing one Room object for each room in the hotel, kinda like this:</p><pre>rooms = [{roomNum: 1, bed: “queen”, reservations: {…}},<br>{roomNum: 2, bed: “queen”, reservations: {…}},<br>{roomNum: 3, bed: “king”, reservations: {…}},…..]</pre><p>Say each “reservations” property is itself a map where the key is a date like “2018–07–01” and the value is true or false depending on if the room is reserved for that day. Not super efficient, but not horrific either and it can be reliably navigated to answer the question. The pseudo code for this might be like this (imagine “criteria” is a structure that holds things like “queen size non-smoking”):</p><pre>def isAvailable?(startDate, endDate, criteria)<br>  foreach room in rooms<br>    if room.matchesCriteria(criteria)<br>      booked = false<br>      foreach date between startDate and endDate<br>        if room.reservations[date]<br>          booked = true<br>        end<br>      end<br>      return true if not booked<br>    end<br>   end<br>   return false<br>end</pre><p>As you can see from my mishmash of Python/Ruby/Java/Totally Fictitious syntax, I don’t care what language you know or if you can create perfect syntax for it. I care about your algorithm. If you’d been performing really solidly up to this point, I won’t bother digging into how “matchesCriteria” works or what, exactly “criteria” looks like. If you’ve been a little shakey, then I will dig into them.</p><p>Some candidates never really get an answer that actually works correctly, regardless of how efficient or inefficient the solution is. In fact, it’s the people who try to prematurely optimize too dramatically that most often get incorrectly working answers. As a rule of thumb, if you’re trying to do things with bit fields and whatnot right from the beginning, you’re probably going to screw it up irrecoverably (and you’re hyper-optimizing something that won’t have a perceptible impact on performance, anyway). These people obviously don’t get hired, so let’s focus on people who do get correctly working answers.</p><p>After you’ve produced something that works, then I’ll ask “ok, that’s awesome. It’s going to work for our problem and get the right answer. Now, if you wanted to optimize it to go faster, what would you do?” Don’t be embarrassed by this question. Absolutely every first try at a problem has some inefficiency in it. The very best candidates own that and are willing to revisit their solution. Horrible candidates stick to their answer like a drowning person at sea clinging to a piece of driftwood and refuse to admit there are ways to improve it.</p><p>There are lots of things you could do. For example, you could keep the rooms themselves in a map instead of a list where the key to the map is the room type. That’s fine, but if you do that then I want to see that you’ve thought about partial matches, for example if someone definitely wants non-smoking, but they don’t care about what bed type they get. Maybe you’ll decide to flip the problem on its head and keep a map where the date is the key and the value is a list of available rooms. I’ve seen some really interesting answers that make use of set operations to come up with the verdict. Anyway, the main point is I’ll take whatever your answer is and probe you on all the places where you traded something off against something else and see 1) if you made that tradeoff with your eyes open, and 2) how you decided where to draw the line.</p><p>We work through the various ways to optimize the solution until we run out of them, or (more likely) I get a good enough idea of how you’re doing. That’s when I throw this at you:</p><p>“Ok, I want to throw a change in requirements at you. Not everybody knows this, but it’s common practice in the hotel industry to take more reservations in a day than you actually have rooms. This is because some percentage of people don’t actually show up, and we want to keep the hotel as full as possible. Let’s say I’ve done some data crunching and I figure I can take an educated risk and overbook my hotel by up to 10% on any given night without leaving anyone short of a room, most of the time. What would you have to change about your design, if anything, in order to accommodate this change in requirements?”</p><p>You always have to change something. Trust me. Only a half-dozen times in 15 years has someone not had to change anything.</p><p>At this point, people who designed the system to reserve specific rooms, as opposed to room types, are really hating life. That’s ok! You’re not in the hotel industry so I didn’t expect you to know that! Ok, once I actually did give this question to someone who had worked in the hotel industry. But only once. Boy was that weird.</p><p>Anyway, I fully expect you to have to refactor your design. The very best candidates are willing to revisit everything, if necessary. Mediocre candidates will only tweak their design around the edges the bare minimum necessary to get the job done. Horrible candidates will bludgeon their exact design into submission, forcing it to do truly unnatural things to avoid changing anything and, yet, somehow still get the right answer.</p><p>If you started your design with something that reserves specific rooms and you change it to something that reserves room types instead, you will blow my mind away. Very few people are willing to revisit their design to that extent mostly just for the benefit of getting the most elegant answer, beyond simply getting a correct answer. To a certain extent, if you’ve gotten this far, I don’t even care about what your solution is. What I really care about is how well you handle changing requirements and what you’re willing to revisit in your previous assumptions.</p><p>At this point, a good solution might look like this, where we have a map of maps like this:</p><p>room type =&gt; date =&gt; integer</p><p>Where the integer is the count of the number of rooms of that type that have been reserved for that day and the system checks that number vs the total number of rooms of that type that exist and returns true if it’s less than 110%.</p><p>If things are going really well, I throw a few more complicated variants of this changing requirement at you and see if you can accommodate them, too, and where you made tradeoffs.</p><p>That’s it. That’s the end of the interview. The whole process takes about 40 minutes and it very accurately surfaces the three things I truly care about when hiring someone:</p><ol><li>Can you deal well with ambiguity?</li><li>Do you make tradeoffs with your eyes open and how do you weigh them?</li><li>Can you deal with change and revisit your assumptions or do you dig in your feet?</li><li>Ok, fine, the fourth thing is do you have a shred of software design sense? See, I revisited my design there by adding a fourth thing.</li></ol><p>Those qualities, more than almost anything else (including if you’ve ever even worked in the language I’m hiring for) are the best predictors of your success on my team. And, so, I screen for them like a man on a mission.</p><p>Why am I giving away my one, prized interview question? Because you can’t study for it. It’s a conversation, not a single, correct answer. And I want better conversation. Please.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=713462ea6bf2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Delivering on Something That Doesn’t Exist]]></title>
            <link>https://trcull.medium.com/delivering-on-something-that-doesnt-exist-bfa2e84edbec?source=rss-89c24437500a------2</link>
            <guid isPermaLink="false">https://medium.com/p/bfa2e84edbec</guid>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[trcull]]></dc:creator>
            <pubDate>Fri, 20 Apr 2018 06:15:20 GMT</pubDate>
            <atom:updated>2018-04-20T06:15:20.880Z</atom:updated>
            <content:encoded><![CDATA[<p>Yesterday, in a fit of frustration, I tweeted <a href="https://twitter.com/trcull/status/986741190576848896">this</a>:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uIjIj7DnaKpBTFy5xQX57A.png" /></figure><p>I was just venting, and quickly moved on with my day. But then, a few hours later, I was shocked to go back to Twitter and see the thing had been liked and retweeted a dozen times. I’m a pretty smalltime Twitter personality, so that’s a significant number. It was enough to capture my ongoing attention and now, the next day at about 11pm, the tweet has been re-tweeted 2,300 times. I don’t know how to emphasize this enough: <em>I’m not a big Twitter personality</em>. If I get one retweet, the dopamine pathways in my brain light up like Christmas trees. So, clearly, 2,300 meant I’d struck a nerve and this subject deserved a more nuanced reflection.</p><p>So, here’s that more nuanced reflection.</p><h3>My Employer</h3><p>First, let me just say for the record that my original tweet wasn’t meant to be read as pertaining to anything specific happening at my current employer. It’s not hard to figure out where I work by Googling me, and once you’ve done that exercise you’ll discover that I’m a low-ish-medium-high-ish level engineering director at a big, recognizable, reputable software company. And, yes, they are reputable. And, no, they don’t fraudulently sell vaporware to noobs and then abscond with the money stashed in offshore bank accounts. The company is full of good, hardworking people who want to do the right thing by their teammates and their customers. In fact, I’d consider my current company to be above average in this department. No, really. So, while reading this, please keep two things in mind. One is that I said “my 20 year career” quite deliberately. That means, I’m speaking from the cumulative experience and scars I have working at (or with) literally dozens of companies. Second is I may have been being hyperbolic when I said “doesn’t exist.” As any grownups in the room would be quick to point out to me, there’s a lot of nuance around what exists and what doesn’t exist in the world of software. So let’s go there now.</p><h3>Defining “Doesn’t Exist”</h3><p>The most jaded in the crowd were quick to say that sales guys will cynically sell complete vaporware that they know damn well doesn’t exist <em>at all, even a little bit</em> just to make a sale. I wish I could say this never happens, but, sadly, it does. It does quite a bit. But, when I’ve seen this happen in the wild, it’s almost always the CEO or the founder of the company doing it and it’s almost always an early stage company whose survival depends on the next sale. I don’t consider it an ethical practice and I don’t consider it even an effective business practice (I’ve seen more than one company go completely out of business for this exact reason) but I do acknowledge that, as the head of the company, it’s that person’s call if they want to bet the farm like that or not. In any case, I think the “complete vaporware” case is actually the least common case. Let’s explore more common cases, instead.</p><p>I’ll dub the most common case the “overly optimistic hallway conversation” case. In this case, a sales person is <em>this close</em> to closing a deal and they know if they could just claim to solve one or two more pain points for their client they’ll be securely over the line. So they look at what the client has talked about during the sales cycle, and they look at the capabilities of the product they sell and they start to see similarities there, if only you hold your head at juuuuust the right angle and apply juuuust a little creativity and thinking outside the box. All of a sudden, in their mind, Feature A can solve Problem B for the client. So they rush to their favorite developer or product person or the like and ask them “hey, could someone solve Problem B with Feature A?” And the answer they get back was probably some variety of “hmm, yeah I suppose you could if you blah blah blah blah blah blah expensive and time consuming modification blah blah blah if the schedule allows blah blah blah talk to Joe to confirm that part,” but, basically, they stopped listening after “I suppose you could” and failed to retain all the rest.</p><p>I’ll dub the second most common case the “didn’t say it couldn’t do it” case. This is where the client themselves starts to see functionality that isn’t there, kind of similar to the “overly optimistic hallway conversation” case, but from the client’s point of view. I think any moderately-intelligent sales person knows damn well the client is doing this, but, to avoid upsetting the deal, they just go right on ahead and let the client think the product does that, assuming it will all work out in the end after the contract settles.</p><p>I’ll dub the last case the “easy, but only if we’re not working on anything else” case. In this case, the product could be modified to do what the client wants without too much difficulty. Not a problem, right? Well, in the world where the engineering organization is sitting around just waiting for things to do, then it really isn’t a problem. But, in reality, getting engineering time is a zero-sum game. So doing even this small modification means bumping off something else, often at the last minute and certainly at the expense of some other client/sales person/product person.</p><p>One can imagine many more incarnations of “doesn’t exist” like “poorly trained sales force” or “too junior to know better” or “was mislead by product or marketing teams” or “ambitious, detached, and/or optimistic senior engineering manager ok’d it” or “absolute expert in product could configure it, but mere mortals don’t stand a chance to get it right” or even “works fine at a certain scale, but totally falls apart at one order of magnitude larger” or “could be done, but creates an unacceptable ongoing operational cost”. So, for the sake of simplicity, please read everything I described above in place of anywhere I say “doesn’t exist.” All of this is to say that most sales people aren’t completely evil, they’re just humans with aggressive goals to meet and families to feed like anyone else and they’re just doing what they’re paid to do: solve problems creatively for the company’s clients. In fact, let’s talk for a moment about sales people…</p><h3>Sales People</h3><p>You can’t really find two tribes of people who are more different from each other than developers vs sales people. So, I suppose I shouldn’t have been surprised by the vitriol I saw directed at sales people as a whole. But, really, I was surprised. And I don’t think most of it is merited. I love sales people. One of my best friends is a software sales person. I’ve found most people I run into in sales to be intelligent, engaging, energetic, optimistic, gets-things-done kinds of people. So I’m not knocking sales people, as a class. What I am knocking is two things: 1) the system of incentives that creates the moral hazard that attracts people to sell things that don’t exist, and 2) the handful of individuals (and it really only takes one) who push this system too far. This is what needs to be held accountable. Not just for the sake of engineers, but also for the preponderance of good sales people who suffer because they won’t cheat in this prisoner’s dilemma problem to make their quota. Because sales people really do have to work their ass off to move the company forward, just as much as engineers do. Which brings me to…</p><h3>Making Revenue and the Contribution of the Sales Team</h3><p>I got several tweets like “without sales you wouldn’t have paychecks” or “you’ve never been in the hot seat so you don’t know what it takes to sell.” Let me address those one at a time.</p><p>First, the easy one: “without sales, you wouldn’t have paychecks”. Ok, well, without engineering, you wouldn’t have a product to sell. And without marketing, nobody would know who the company was when you cold called (because without marketing, cold calling is all you’d be doing) and the prospect would just tell you to fuck off. And without customer service, many of the customers you sign wouldn’t stick around long enough to pay back their ROI. A company is a team, plain and simple. Just because sales is the only team that gets paid a commission doesn’t mean they’re the ones solely, or even largely, responsible for the revenue.</p><p>Next, the one that really pisses me off: “you’ve never been in the hot seat so you don’t know what it takes to sell”. Let me start with me, personally. For five years, I ran a pretty successful company that I bootstrapped from nothing and grew to employing seven people with zero outside funding. No, that’s not Facebook, but it’s definitely more than enough to call bullshit on anyone who’s addressing that statement to me, personally. If you’re addressing that to engineers, as a class, then, sure, most of them probably haven’t had to sell. Also, you’ve never been dragged away from your daughter’s birthday or from Thanksgiving dinner or woken up at 3am to fix a flaming pile of software you’re on call for. Yes, all of those things have happened to me. And they’ve also happened to most of the engineers I know. And do you think those engineers cut corners on fixes or ignore their phones just because it’s too hard? I can tell you, definitively, they do not. At least not the ones who stay employed, especially if they work for me. So, let’s just say the “you don’t know what it’s likes” should all cancel each other out on both sides.</p><p>Which then leaves the question: is sales so hard that it’s ok to do unethical things and screw your customers just to avoid missing your quota? When I express it that way, I dare you to say “yes”. You wouldn’t. Especially not in front of your CEO. But, even a moral compass isn’t necessary to see the flaw in this argument. Let’s just take it on its surface and ask ourselves the Machiavellian question: is the expected lifetime value of customers you lie to and then inevitably disappoint higher than that of customers you never disappoint but occasionally lose a sale to? No, it is not. I’ve lost sales because I wouldn’t promise the impossible and my competition would. Guess what usually happens? Those customers come back, six months later, after my competition has completely let them down. Sometimes after the internal sponsor at that company has been fired.</p><h3>So Why Do We End Up Here?</h3><p>Clearly, from the tweet storm I created, this happens a lot in spite of it being a Bad Thing. So, why? I think it’s a mixture of a testosterone-driven culture, misaligned incentives, and lack of visibility.</p><p>To gratuitously throw around some stereotypes, sales culture tends to be a set-the-bar-high-and-you’ll-get-results, sports-metaphor-inspired, go-big-or-go-home kind of culture. Good sales people are competitive to their core. They <em>hate</em> losing. And they are doggedly persistent. And we want these things because they are all actually critical to success in sales, and being successful in sales is, pretty obviously, one of several prerequisites for staying alive as a company and paying salaries. Engineers tend to find this truth mystifying because many of those same qualities in an engineer are a <em>complete and total disaster</em>. This testosterone culture can be harnessed to do some beautiful things. Where this testosterone culture goes wrong, though, is when it’s directed by the wrong incentives.</p><p>And this part is where actual sales people can correct me if I’m getting things wrong. In companies where this is the worst, sales people are compensated based on the contracts they get signed, primarily, and the actual revenue that comes off those contracts over the long term secondarily, if at all. This creates an incentive to get things signed at all costs. And if fulfilling those contracts requires engineering investment, the value of that investment isn’t taken off the top of the commission. So there’s all the incentive in the world to promise things and no incentive not to promise things, at least at the individual sales person level.</p><p>VPs and Directors of sales teams, however, have a more holistic incentive. Their compensation tends to be tied a bit more to long term company outcomes which are, quite obviously, tied to long term customer happiness and overall company effectiveness. They, if anyone, should be in the best position to make this situation better. The problem is, how do they tell the difference between someone selling something that isn’t there and engineering simply screwing up or being slow? Yes, engineering screws up at the same rate sales people screw up, which is the same rate that humans in general screw anything up. Basically, these executives can’t tell the difference, because they don’t have the visibility. And they don’t have the visibility partially because we, in engineering, aren’t creating that visibility in a way they can digest. They don’t know if a Bad Thing happened or if they’re just the victims of bad engineering. And, in the absence of better information, why wouldn’t they believe their own team, the people they’re in the trenches with every day, over people they hardly see?</p><h3>What’s the Cost of Being Here?</h3><p>Too often, we focus on the client we’re going to lose if we don’t deliver the “doesn’t exist” feature. But what about the other clients and, it should be called out, <em>the other sales people</em> who the company made commitments to in an orderly fashion and put on a roadmap to deliver features to? The sales person who promised a world that doesn’t exist is not just screwing the engineering team who has to deliver the feature, he or she is also screwing their comrades who sold what was actually on the roadmap. And they’re screwing their comrade’s customers. They’re probably even screwing themselves, because some of those screwed customers are going to be their own accounts.</p><p>There’s also a cost to developer attrition. Especially in this tight labor market, why would the most talented developers stick around for this kind of treatment? They wouldn’t, that’s the answer. Pretty soon you have a negative self-reinforcement cycle where bad developers stay and good developers leave and, eventually, no company at all to sell for. Think that’s extreme? I’ve seen it, several times, with my own eyes.</p><p>Lastly, just like in the classic prisoner’s dilemma, once you’ve rewarded cheating and all the players in the game see it, then every rational actor in the equation should also start cheating. Or, let’s say “swinging big” instead of “cheating” if that’s easier to digest. And once you have everyone promising things that don’t exist, it’s only a matter of time until that same negative reinforcement cycle kicks in and the company circles the drain.</p><h3>Ok, Tough Guy, How Would You Fix This?</h3><p>I’ve got two easily-digestible fixes and one hard-to-swallow pill.</p><p>The first, easily-digestible fix is to simply stop celebrating this behavior. Did fulfilling a sale require midnight heroics and lots of heat and light to pull over the line? Don’t celebrate it as a victory then. Don’t make it part of your company lore. Don’t stand up in company all hands and sing kudos to the sales people who made it happen. Treat it like the narrowly-missed disaster that it was and just say nothing about it. Instead, celebrate the top sales people who actually sold what your company has. Don’t promote the people who engage in this behavior. And then see how much more of the right kinds of sales you get.</p><p>Second, create visibility. This rests squarely in the hands of engineering leadership. We’re not a terribly confrontational bunch. We tend not to like to have uncomfortable conversations. But we’ve got to do it anyway. I’m speaking to myself here, as well as all my peers. Create relationships with your counterparts in product and sales and make sure that, each time this happens, everyone knows exactly what was traded off in order to make the sale happen. Don’t be a dick about it, just convey facts. If you can, trace it all the way to other sales that didn’t happen because the product wasn’t built that would have enabled them. Articulate the cost in terms that make sense to sales people. Don’t just assume your gripes will be understood in a salad of engineering speak. Don’t just complain once. Complain repeatedly and constructively and to the right people. Say it in words that a team player would say, not a disgruntled do-nothing. But do, in fact, say it. And don’t be afraid to make people uncomfortable, including yourself.</p><p>Last, and here’s the bitter pill, create a financial incentive to discourage this behavior. If completing a sale requires engineering work to get things done, then take 10% of the sales commission and put it in a bonus pool for engineering that gets paid out at the end of the year. Think that’s unfair? Well, would the sale have happened without the engineering work? No? Then engineering should get a cut because they were basically (and involuntarily) part of your sales team. I call that fair. And, besides, if it really was that valuable of a sale compared to sales you could have made with the products you already had, then you’ll come out ahead still, anyway. And if it wasn’t valuable enough to overcome that 10%, then the sale shouldn’t have been made. Because, at the end of the day, it really did cost the company to make that sale. Maybe it didn’t cost <em>you</em> to make that sale, but it did cost the company. So it’s not good for the company as a whole unless it was a really, really good sale. And, anyway, engineers aren’t monks. Just see how much more cooperative and vigorous those engineers are if they’ve got skin in the game, too.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bfa2e84edbec" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>