<?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 Emmanuel Gautier on Medium]]></title>
        <description><![CDATA[Stories by Emmanuel Gautier on Medium]]></description>
        <link>https://medium.com/@emmanuel.gautier?source=rss-fbe4cf777dd1------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*GbZqz9BxuKXygD49xxDTuQ.jpeg</url>
            <title>Stories by Emmanuel Gautier on Medium</title>
            <link>https://medium.com/@emmanuel.gautier?source=rss-fbe4cf777dd1------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 13:20:00 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@emmanuel.gautier/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[Frank teller]]></title>
            <link>https://medium.com/@emmanuel.gautier/frank-teller-a1198409104?source=rss-fbe4cf777dd1------2</link>
            <guid isPermaLink="false">https://medium.com/p/a1198409104</guid>
            <category><![CDATA[magical-realism]]></category>
            <category><![CDATA[short-story]]></category>
            <category><![CDATA[short-fiction]]></category>
            <category><![CDATA[fiction]]></category>
            <dc:creator><![CDATA[Emmanuel Gautier]]></dc:creator>
            <pubDate>Thu, 29 Sep 2022 15:58:24 GMT</pubDate>
            <atom:updated>2022-09-29T15:58:24.123Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oA2azOO86iJe6RugeIvKDQ.png" /></figure><p>“Oh, my little girl.”</p><p>“What is it, dad?”</p><p>“Here we are, waiting for this old body to give in. You’ve been here hours. This is a splendid afternoon, you could be spending it in much better places than a hospital room on the banks of River Styx with me.”</p><p>“Don’t you fucking say that. I’m not leaving you.”</p><p>“Come, don’t sob now. I hate you putting everything on hold for me, is all. Regardless of what the doctors are saying, it could be a matter of weeks, months even, for all I know. What a puzzling state of things. I don’t even feel that sick — frankly, I’ve <em>worked</em> on far worse days! But ah. If this is indeed it, my last hurrah, and if you insist on staying, then I might as well tell you the story.”</p><p>“What story? What are you talking about?”</p><p>“The one story I’ve never told you — or anyone, for that matter. A story that runs through unobserved corners of my life, moments I was on my own. I was planning on keeping it to myself.”</p><p>“Why?”</p><p>“Because, darling, it doesn’t make any kind of sense. No matter how hard I’ve wanted it to. I’ve tried to lead a sensible life, one with purpose and value. But I could never quite shake the knowledge that throughout, I was carrying this burning nugget of madness with me, which invalidated everything else. Now, facing the prospect of bringing it with me, this truth feels far too crucial to bury. I must tell you that <em>there is something wrong with the world</em>.”</p><p>“Dad. I don’t understand.”</p><p>“When I was 19, I had a dream. The kind that leaves a mark on you. The kind you wake up from in a <em>malaise</em>, covered in sweat that feels like someone else’s.</p><p>“In the dream, there was a man, his back to me. He wore an immaculate black suit and a matching fedora hat. He stood completely still — impossibly so — and gazed over a city from a large glass window. I couldn’t get close to him. Everything was calm for a while, until he started… speaking.</p><p>“Now, ‘speaking’ here is something of a stretch, as this wasn’t in any tongue known to man. I would describe it as something between whispering and whistling. I somehow <em>knew</em> this was a language, even though I did not understand a word of it. In the ebb and flow of these peculiar sounds, I could sense sentences, tone and structure. There was a distinct quality to the man’s speech, one of calm command. As he finished stating his instructions, something was set in motion.</p><p>“A different set of sounds began, still quite soft but more machine-like, from a separate source. A mix of whirring and erratic scratching. It went on for a handful of seconds, and when it stopped, a couple of words appeared to me, superimposed in the air over the man’s silhouette, in large letters and a shade of blue slightly lighter than the evening sky. They read, ‘rich judge’.”</p><p>“Rich judge?”</p><p>“Those very words. That’s when I woke up, feeling bent out of shape, foreign. I had never seen written words in a dream before. Thinking about what I had just experienced was uncomfortable. I pushed away these thoughts and went on with my day.</p><p>“Nothing happened until, some three weeks later, I had the same dream. The man, the office room, the sounds. It wasn’t exactly a repeat, though. The sky seemed to belong to a more advanced part of the night, and I was viewing the scene from a slightly shorter distance. And most importantly, a different pair of words appeared at the end. This time, it was ‘ocean storm’.</p><p>“Once awake, I lingered in my bed and pondered the bizarre re-occurrence. People can have the same dream several times, but so specifically? And what of the words? Little did I know…</p><p>“That very day, Hurricane Bhola formed in the Indian Ocean. It went on to lash the coasts of a place we called East Pakistan at the time, and which is now parts of India and Bangladesh. It killed hundreds of thousands and was the deadliest storm of the century. I watched the disaster unfold evening after evening on the TV news, in shock.</p><p>“I could not believe what was happening to me. It was massive and it was simple and the sheer impossibility of it was trounced by it… being. I could not even bring myself to put ‘it’ into words. I was amazed, terrified, and disgustingly, excited. I went to bed each night dreading and hoping for another instance of the dream, one that would confirm the unthinkable.</p><p>“Oh, I never waited for something else so impatiently. And the man kept me waiting for a maddeningly long time, as though I had to earn this third appearance. So much so that when it finally came, six months later, I was frantic.</p><p>“The words that night were ‘art price’. They sent me on a hunt the entire day, combing through every art magazine I could put my hands on, seeking anything related to a particularly expensive work. This was wholly unnecessary, as what I was looking for made the evening news even in mainstream outlets: a Van Gogh had been sold for twenty-three million dollars, in a record-breaking transaction.</p><p>“There it was. These dreams <em>meant something</em>. Can you imagine how mind-shattering this revelation was to me? I was enthralled. My whole life became background noise, as I spent every waking hour either waiting for another encounter with the man, or investigating the implications of a past one.</p><p>“I never found a clear pattern as to when I would be visited. Sometimes I would wait for months, like I did after the second dream. Others, I would have three visions in the space of two weeks.</p><p>“Every dream followed the same format. Man, whispers, machine, words. The time of day varied mildly, though only between late evening and moments before dawn — I eventually noticed sunlight was sometimes coming from different sides, marking the distinction between sunset and sunrise, though I had no idea which was which.</p><p>“And it was always two words. Never more, as if to make the guessing more challenging. After the seventh or eighth dream, I started writing them down to track the meaning of each one. That first clue, ‘rich judge’, I eventually traced back to a bribery scandal which cost a magistrate his seat. The scandal had broken a few days after I had the dream.</p><p>“All the enigmas I solved always had to do with some world event. Anything that made the news: death, war, sports, discoveries, records… Oftentimes, I wouldn’t find out my dream’s meaning until weeks later, when I saw something related in the papers. But when I traced back the date of the dream, it was usually the day the event actually happened.</p><p>“One day in the 90’s, I drew ‘page turner’, and what do you know? By the end of the year, the entire planet was reading Harry Potter. I looked it up, and it turned out I had had the dream the night before the book was first published.</p><p>“A few years ago, it was ‘gene cook’. I had to wait a lot longer, but it turned out, it was about that Chinese scientist who had edited the DNA of a pair of human fetuses, in spite of every single bioethics rule.</p><p>“Some clues, on the other hand, were too vague for me to ever figure out what they were referring to: ‘miles long’, ‘dawn hunt’, ‘amber hall’… Regardless of the outcome, I had tremendous fun going after each of these.”</p><p>“I’ve only just realised… that’s why you were always so interested in the news. You would read the paper front to back every single day, and insist on putting on the evening news on TV.”</p><p>“Indeed. Though this habit kind of fell off with nonstop news and the Internet. Nowadays I mostly check push notifications on my phone.”</p><p>“Did you get a clue for… every event? Did you dream about 9/11?”</p><p>“Not to my knowledge, no — I did get ‘paris burns’ the night before the Notre Dame fire, though. Exactly which events I get warned about is also part of the mystery, and perhaps solving all the clues would help in cracking this particular question.</p><p>“At one point in the early years of the dream, when I was the most feverish about the whole situation, I started wondering whether some of the clues could be directed to me specifically. I had two dreams in the spring of ’72, which said ‘june love’ and ‘summer darling’. I was absolutely convinced I was about to meet the love of my life that summer. This failed to happen, in a most humbling turn of events. These two clues remain unsolved to this day. I did, however, meet your mother some five years later — in November!</p><p>“Now, this may come as a surprise, but I have yet to get to the oddest part of my story. Because of course, even solving every riddle would not have quenched my curiosity. I wanted to solve The riddle, the how and why I was suddenly bestowed the improbable gift of vague prophecy. A dangerous idea bubbled in my mind: what if the place where I saw the words at night was real?</p><p>“I started paying attention in the dreams — not just to the words, but to details. I wrestled my subconscious (it’s hard to focus when you’re asleep!) for every shape I could make out on the skyline.</p><p>“Remember, this was the 70’s. No Google Maps, no Wikipedia, no reverse image search. So I went and bought a cartful of atlases and city photographers’ coffee table books. I had ‘New York in pictures’, ‘San Diego sights’, ‘Portraits of Austin’, you name it. I even expanded my research to a few Canadian and Australian cities. For hours I would look at the photos and try to cross-reference them with my dream notes.</p><p>“It took months, it took sweat and it took tears. I was constantly following three to five serious leads, regularly declaring one not viable when too many inconsistencies had piled up. All of this, for me to candidly stroll into a showing of Rocky II on a scorching saturday afternoon, and simply recognize the city I had obsessively been after.</p><p>“It was the famous scene where Sylvester Stallone jogs through Philadelphia that sealed it. I recognized a bridge, then a building block, then a skyscraper, then the whole skyline. It was electrifying. As soon as I got home, I took my copy of ‘Cities of Pennsylvania’ and pinpointed the locations of the items I had recognized on a map of Philly. Placing them on my line of sight restricted my building to a roughly five-by-five block area, and I was already on the phone booking a ticket.</p><p>“Two days later, I was staring at a thirty-five-story giant, on the corner of 5th and Walnut. Believe me, it was staring back. I’m sure any passerby would have described it as a terribly banal tower. I couldn’t help singling it out, a monstrous oddity, offensively passing as a mundane stack of offices. An antenna to the realm of the impossible, hiding in plain sight.</p><p>“My visions likely took place on the building’s west side, between the twentieth and twenty-fifth floor — slightly above a neighboring tower’s roof. I had put on a suit and was prepared to say I had an appointment with the law offices of Garrison &amp; Graves, which were on the twentieth floor. No one asked any questions, as it turned out, and my elevator ride was uneventful, if a bit sweaty.</p><p>“The twentieth wasn’t right. Clearly too low, and the carpet didn’t match the one from my dreams. I tried my luck with the twenty-fourth, maybe a bit too high. I took the stairs down to the twenty-third, and the moment I stepped into the deserted floor, I knew I was there.</p><p>“It felt like no one was aware this floor existed. Office cubicles were neatly furnished, ready for use — except that everything was covered in a thin layer of dust. I didn’t see any clocks or calendars. I approached a desk and pulled open the first drawer. It contained a paper tray with company letterhead inside. It read, ‘Verity Hall’s Luxury Clothing Ltd.’</p><p>“I reached the end of the bullpen, turned a corner, and my heart stopped. Right in front of me, behind a glass wall with initials VH printed onto the door in calligraphy, was the man. Striking the exact same pose as in my dreams, fedora on his head.</p><p>“I almost made a run for it. Breathless, I reached for the door handle, and as I was entering the room, it hit me. Mannequin. Of course. I slowly paced around the silhouette, until I was next to him, facing the window. The featureless, glossy polymer face was scanning the horizon, undisturbed.</p><p>“The room appeared to be intended for storage of tailor-made suits. Against one of the walls, a rack was doing its best to hold a heavy array of sleeved suits of various fits and colors. Next to it was a table. On it, a Xerox printer of a make I’d never seen, set up so as to spit out listing paper directly into a tray for inspection.</p><p>“Then the sound came. The whisper, or whistle, I’d heard so many times. Except this time, I was awake, and could easily place its origin. The air duct embedded in the suspended ceiling was very close to the outside wall, and a flag’s movement on a neighboring building confirmed my suspicion: it was the wind.</p><p>“I circled back to the Xerox machine. It was on, though at rest, and more importantly, the paper stack in its out tray was the only dustless item on the whole floor. I pulled at the first page from the listing, and the rest followed in a zigzag. Each page was an order receipt for a suit with VH.</p><p>“And something arrested me. A pair of words, written in the light blue typography I knew by heart. All lowercase. ‘jewel cruise’. These, darling, were the two words I had seen in my latest dream, two weeks earlier.</p><p>“I ran through the other pages. Summer Darling. Rose Thorn. Dawn Hunt. Art Price. Rich Judge. They were all there. My little girl, they weren’t nouns. They were… customer’s names.</p><p>“My mind went spiralling. I could not think straight. I dropped the stack of paper, and ran out of the office, took the stairs all the way down to the ground floor, out the building, and I just puked in the gutter, right there on the street. I left, and never came back.”</p><p>“What? You <em>never</em> came back? But what about those people? Were they real? And that company? Didn’t you try to find out more?”</p><p>“I pondered it a good deal. The decision, in the end, wasn’t a rational one. The way I see it, curiosity works a bit like appetite. And this episode, on that day, was more to ‘eat’ than a lifetime can even bear. It left me content to keep receiving my little clues every now and then, and try and tie them to world events. I did not want anything more to do with the absurdity of that room outside of my dreams.”</p><p>“But dad, everything you said, everything you’ve been through… like you said in the beginning, it feels so… important! Why would you leave it be like this? Don’t you want to know what all of this means?”</p><p>“Hah, I figured you would respond this way. One of the many reasons I love you so dearly. This fight has brought me joy and adventure, but it is no longer mine to wage. You have everything you need, should you choose to investigate this matter. The office room is still where it was, as far as I know. I wish for you to you find your answers, Hope. Few do.”</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a1198409104" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to build a smart multi-step form in React]]></title>
            <link>https://medium.com/doctolib/how-to-build-a-smart-multi-step-form-in-react-359469c32bbe?source=rss-fbe4cf777dd1------2</link>
            <guid isPermaLink="false">https://medium.com/p/359469c32bbe</guid>
            <category><![CDATA[react]]></category>
            <category><![CDATA[software-architecture]]></category>
            <category><![CDATA[multistep-form]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[refactoring]]></category>
            <dc:creator><![CDATA[Emmanuel Gautier]]></dc:creator>
            <pubDate>Thu, 15 Apr 2021 13:19:24 GMT</pubDate>
            <atom:updated>2021-04-21T13:07:23.588Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PVujFfvsf-mLldUN3G9Hrg.png" /><figcaption>Illustration by Garance Bigo</figcaption></figure><p>A housing architect friend of mine once told me that every building has exactly two sides: the visible, and the invisible. It’s something you’ve probably thought about if you’ve ever had the eerie experience of visiting multiple cookie-cutter condos in the same building: the floor plan, structure, and foundations are identical, but the tiling, paint, and decoration have diverged over the years. So I asked my friend, ‘What about the opposite? If you built visually identical houses, with completely different foundations for instance? Which one would be the better house?’ He laughed and said, ‘The one that collapses last!’</p><p>I recently thought about my friend’s joke in terms of software and, well, he was spot on.</p><h3>The mission</h3><p>Back in september 2020, my team was tasked with adding a feature to our medical software: a multi-step form (also known as a <em>wizard</em>) for patient identity verification. Basically, at the beginning of a consultation, the doctor will need to scan their patient’s social security card, link it to the right patient file, fill in some crucial info manually if missing (names, birth date, sex), and — why not — merge duplicates of the patient file if our software finds any.</p><p>As you may have guessed, this was not exactly a top feature request — nobody likes an extra administrative step before getting on with work — but rather, a micro-chore to keep the doctor’s medical data reliable and up-to-date. Why does it matter? Imagine prescribing medication to the wrong person!</p><p>With that in mind, our main goal was to build a painless user experience: make the wizard simple, forgiving, and capable of automatically skipping steps that aren’t relevant for a given patient.</p><p>‘Cool’, I told my product manager, ‘ambitious but cool. Let’s build something nice!’</p><p>‘Oh’, he said, ‘by the way we need a first version of this for the beta, which we’re launching in three weeks.’</p><p>‘WHAT?’</p><p><em>(dialogue exaggerated for dramatic effect)</em></p><p>Well, this was going to be <a href="https://medium.com/doctolib/the-two-kinds-of-tech-scoping-ab8d1915ec5a">more duct-tape than beaver dam</a>, I thought to myself. Of course, the two of us were happy to trim off some non-essential parts of the project, but at the end of the day, there was still a large chunk of features to ship in a tight three weeks. We weren’t going to have time for satisfying abstractions.</p><h3>Version 1: A bare-bones sequence of modals</h3><p>Once we had built the components for each step’s modal, we needed to plug everything together. To help us in our quest, we have an in-house modal presentation API that can be called from anywhere in the app:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/100f15fa9d60ee2e81c79c224f464c2e/href">https://medium.com/media/100f15fa9d60ee2e81c79c224f464c2e/href</a></iframe><p>We had a pretty big constraint: after finishing the wizard, we needed to initiate a medical consultation. The callback that triggered this was generated in a very context-specific way, so we were forced to generate it from the beginning of the wizard and pass it around the modals throughout.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/a99622c950c21cc741a61ed6b9ceb546/href">https://medium.com/media/a99622c950c21cc741a61ed6b9ceb546/href</a></iframe><h4><strong>Verdict?</strong></h4><p>Well, it worked. We shipped on time. Our ‘house’ looked (and behaved) the way we wanted it to, pretty much. This was satisfying enough for a beta version, but looking just one step ahead, we could already see why the woodwork wouldn’t hold up:</p><ul><li><strong>The code was very hard to read and to maintain.</strong> Someone looking at the code for any of the modals could barely tell that it was part of a wizard. Even if you were aware of the wizard’s existence, you would need to go on a quest across several files just to figure out its step progression, never mind understanding the subtle logic of skipping this step or that one.</li><li><strong>Our wizard was not flexible. </strong>Want to insert a new step between two others? You’ve gotta unplug and replug everything, basically. Want to have two versions of the flow with slightly different behaviors? Get ready to bloat your payloads even more with ternaries. Want to be able to switch the patient file you started with midway if you realize it wasn’t the right one? No way, the callbacks were computed from the start of the wizard. As a matter of fact, the wizard <em>was</em> supposed to do all of those things eventually, so, yikes.</li><li><strong>There were a few unwanted behaviors</strong> due to the fact that our wizard was stateless, the most annoying one being, no going back to the previous step (again, feasible but painstaking to code, so it didn’t make it in version 1).</li></ul><p>Going forward, we simply couldn’t afford this architecture, and given that we had met our deadline, we now had time for a rewrite of our wizard.</p><h3>Version 2: Like a state machine 🎶</h3><p>We set two main objectives for our second iteration on the wizard:</p><ul><li><strong>Centralize the code</strong>, so that understanding and maintaining its logic is as easy as possible.</li><li><strong>Add stateful logic</strong>, so that we can observe, control, and modify its behavior.</li></ul><p>At the end of the day, the goal was that we able to control the wizard from anywhere, like so:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/50bcfb2718f269dc9895a198c5837264/href">https://medium.com/media/50bcfb2718f269dc9895a198c5837264/href</a></iframe><p>With the wizard’s entire logic black-boxed away in a single file, the wizard just handles itself and saves you the trouble of wondering what the next step in the process should be. <strong>So how did we achieve this?</strong></p><p>We start our wizard component with an array of steps, each represented by an object containing two callbacks, <em>effect</em> and <em>shouldSkip</em>. Here’s an example:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/83de54a3a7f714cec76ba03c25106718/href">https://medium.com/media/83de54a3a7f714cec76ba03c25106718/href</a></iframe><p>And we introduce a state variable <em>currentStepIndex</em>. See where this is going? When asked to trigger ‘the next step’, we move forward in the steps list, until we find one that shouldn’t be skipped, and we trigger that one.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/bcfe9029892f9402ca0ca12f5efcd106/href">https://medium.com/media/bcfe9029892f9402ca0ca12f5efcd106/href</a></iframe><p>The <em>initiateIdentityWizard</em> callback is even more straightforward, it initializes the state with the data it’s received and kickstarts the wizard. Typically, the initial patient is stored in the state and can be easily switched later on:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/3281d701d808b408b6315d7944b5b4d8/href">https://medium.com/media/3281d701d808b408b6315d7944b5b4d8/href</a></iframe><p>And that’s pretty much it.</p><p>Note that our wizard component doesn’t render any visual components itself. It uses our openModal API to present a different modal on each step. The only JSX we render is a React context provider:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/3130c48e384ec9cc3a0de1b3b6ca03a1/href">https://medium.com/media/3130c48e384ec9cc3a0de1b3b6ca03a1/href</a></iframe><p>Now that our component exposes its member functions via a React context, anywhere in the tree below the provider, we can consume this context and call these functions:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/5f87d4dc2897a7664b0231282ffc3837/href">https://medium.com/media/5f87d4dc2897a7664b0231282ffc3837/href</a></iframe><p>Just like we wanted! And now, editing the steps in the wizard is as easy as adding, removing, swapping, or modifying items in the STEPS array.</p><h3>Takeaways</h3><p>As any engineer knows, there are a million ways to code a given user interface, with a given behavior. The one you want, ideally, is the one that doesn’t collapse on the slightest change, especially for user-critical wizards with complex rules.</p><p>This project in particular was a great way for us to see the difference between a ‘naive’, bare-bones architecture, and a well thought-out one. Adding the right layer of abstraction made our code noticeably more robust, and more maintainable.</p><p><em>If you want more technical news, follow our journey through our </em><a href="https://doctolib.engineering/engineering-news-ruby-rails-react/"><em>docto-tech-life newsletter</em></a><em>.</em></p><p><em>And if you want to join us in scaling a high traffic website and transforming the healthcare system, we are hiring talented developers to grow our tech and product team in France and Germany, feel free to have a look at the </em><a href="https://about.doctolib.com/jobs?department=Engineering"><em>open positions</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=359469c32bbe" width="1" height="1" alt=""><hr><p><a href="https://medium.com/doctolib/how-to-build-a-smart-multi-step-form-in-react-359469c32bbe">How to build a smart multi-step form in React</a> was originally published in <a href="https://medium.com/doctolib">Doctolib</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The two kinds of tech scoping]]></title>
            <link>https://medium.com/doctolib/the-two-kinds-of-tech-scoping-ab8d1915ec5a?source=rss-fbe4cf777dd1------2</link>
            <guid isPermaLink="false">https://medium.com/p/ab8d1915ec5a</guid>
            <category><![CDATA[project-management]]></category>
            <category><![CDATA[software-architecture]]></category>
            <category><![CDATA[software-engineering]]></category>
            <dc:creator><![CDATA[Emmanuel Gautier]]></dc:creator>
            <pubDate>Tue, 05 May 2020 11:13:03 GMT</pubDate>
            <atom:updated>2020-05-05T11:13:03.617Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dJtXMtfOdRjl5mtyAnlahw.png" /><figcaption>Illustration by Gaëlle Malenfant</figcaption></figure><p><em>“In a day?!”</em>, I asked incredulously.</p><p>I had been telling my manager, Alex, about my first experience <a href="https://medium.com/doctolib/the-tech-holder-the-key-to-empowering-engineers-about-the-build-bc2c1be67871">as a tech holder</a>. The exercise had proved super fun and challenging to me, but I was slightly concerned about the roughly two weeks I had spent <em>scoping</em> the project before we kicked it off with the entire team. Not that I had delayed coding for the team — I’d got started well in advance while we were finishing up our previous tasks — but I felt I could have been more efficient.</p><p>“There’s really nothing to worry about”, Alex told me. “It’s your first, plus that project was pretty chunky, so I’m glad you took the time to get it right.”</p><p>And that’s when he dropped that bomb on me. “But, you know, depending on who you ask around Doctolib, <em>some people will tell you that they can sit and scope a project pretty much in a day</em>.”</p><p>Cue my reaction.</p><p>It made me reflect on what I thought tech scoping was. On the most basic level, a project’s scoping is supposed to answer these questions:</p><ul><li><strong>What features are we going to add, change or remove in the app? What’s the intended user impact? </strong>That part is essentially provided by the product manager, but it can be challenged and iterated over together with the tech holder.</li><li><strong>How are we going to build those features? </strong>That is, the technical architecture of the project. What parts of the codebase are we modifying to get the desired features?</li><li><strong>How are we going to build those features, <em>Part two</em>? </strong>This time, meaning, who’s going to work on them, what’s the timeline for shipping each feature?</li><li><strong>What are the risks?</strong> What’s the plan if they materialize?</li></ul><p>Now, here’s how I went about it the first time ‘round.</p><h4>The Beaver-Dam Scoping strategy: thorough and robust</h4><p>First, I’m in <strong>Architect mode</strong>, making structural choices for the project.</p><p><em>“For this feature, we want to add a signup form to this page, what does that entail? We’re not displaying it to all our users, so we need to compute which users are eligible.”</em></p><p><em>“The business rule is pretty complicated, and we’ll use it multiple times, so a helper on that model seems relevant, and we’ll pass that information to the front-end.”</em></p><p><em>“Interface-wise, we already have a signup form component, with one field that we’re not going to use. We can probably re-use that component and pass it a property to indicate whether or not it should include that field.”</em></p><p><em>“Since we’ll be displaying it in a high-traffic funnel, we might want to wrap the feature in a feature-switch, so that we can easily disable it in production in case anything goes wrong.”</em></p><p>If the feature is simple enough that it can be coded (automated tests included) by a developer in less than two, maybe three full days, then that’s it, that’s one story. If it’s more complex than that, I’ll split it up in smaller stories, that can be independently coded, tested and shipped.</p><p>Then comes <strong>Scout</strong> <strong>mode</strong>. I validate my macro choices by digging a bit further in the codebase, hunting for overlooked complexities. This is also a great occasion to assess risks and design their mitigations.</p><p><em>“So, which controller is used to pass information to that page? Okay, I can see that there’s some property-engineering done in the view, we should be careful with that.”</em></p><p><em>“Ah, there’s a helper method on our model that almost computes what we need, but not quite. Let’s see where else it’s used, and decide whether it makes sense to add our case in it.”</em></p><p><em>“This database query is more complex than I thought. We might need to consider doing this lookup elsewhere to preserve the app’s performance.”</em></p><p>In addition to handy implementation notes for my team in the stories, the product of this mode is something we do for almost every project at Doctolib: a scoping document, which lists the planned features, along with the controllers, models, views, components, jobs, etc. that we’re planning to touch. It serves two purposes:</p><ul><li><strong>Information sharing with other feature teams.</strong> Without reading my colleagues’ scoping documents, I would have little idea of what most of them are working on. More pointedly, this has sometimes prevented two projects colliding!</li><li><strong>A sanity check on tech choices.</strong> Every single scoping document I’ve ever seen has received at least one or two important questions from other developers, and some of those have led to major project rethinking, which is pretty healthy.</li></ul><p>Finally, <strong>Planner mode</strong>. Typically, the project’s stories depend on each other, and can’t be coded in any order. I need to chart the dependencies between stories. There:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*yqRRftddJGGPDAMtr-75lA.png" /><figcaption>The actual chart I made for my first project scoping (except with slightly more vague feature names)</figcaption></figure><p>That way, in a glance, we know where to pick up coding to start a feature, and where the bottlenecks are.</p><p>Using that, I can build an awesome timeline for the project, that accounts for:</p><ul><li>ballpark estimates of the duration of each stories,</li><li>The availabilities of each dev,</li><li>Planned bug days (pink rows in the picture below),</li></ul><p>and compute a (very) rough delivery date.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/914/1*LYPEAx49OBu1IoB7LBFF2g.png" /><figcaption>If you look closely at this timeline, you’ll notice that I’m a spreadsheet nerd.</figcaption></figure><p>Now, this isn’t really rocket science, but the great thing about this is that I can adjust this timeline as we go through the project, take weekly snapshots, and know when we’re falling behind our estimates. Case in point: the picture above is the timeline <em>at the end </em>of the project, and makes it obvious how 245 and 257 were long tasks that maybe could have been split.</p><p>It almost goes without saying (but <em>almost</em> is key here): this is a linear account of a process that was cyclical.</p><p>The project turned out really well! We’d anticipated most pitfalls, and everything was working as intended when we shipped. I got some positive feedback from my teammates for my scouting work on the stories, and felt like we’d tackled a complex project without much hassle. The only mildly disappointing outcome was the long time spent on two tickets, but as we weren’t under a hard deadline for delivery, this was far from a catastrophe. Plus, reflecting on them helped me consider how to better split stories in the future.</p><p>But in a later project, I got to try a <em>very </em>different approach.</p><h4>The Duct-Tape Scoping strategy: frugal and agile</h4><p>For the whole of Doctolib’s tech division, the COVID-19 crisis has meant major priority adjustments. Our product is a critical tool used on the frontlines of the response to the pandemic, and health care practitioners need us to ship many features as quickly as possible to help them organize their activity.</p><p>So when it was my turn to scope a COVID-related project, on a topic out of our team’s usual scope, with a *gasp* two-week deadline, scoping included, I thought, ‘Well, there’s an opportunity to test the one-day scoping, if I ever saw one.’ So, how did I proceed this time?</p><ul><li><strong>Talk To Someone Who Knows™</strong>. This one’s a no-brainer. Since you’re picking up a topic from another team, even though they’re obviously busy, a 30-minute bootstrapping meeting with them is well worth it. It saves you hours of fumbling through their code while still being unsure you’ve covered everything relevant.</li><li><strong>Don’t overdo the Scout mode.</strong> Instead of turning every stone in the codebase, perform a quick due diligence exercise, to detect only what’s really out of the ordinary. Document only what either makes you change significant architecture choices, or what will save the developer who codes the story over 15 minutes. Let Convention Over Configuration pay off, it will take care of the rest. Better yet: if you think there are pieces of code that might be tricky, but clearly don’t threaten project architecture and aren’t top-priority, don’t hold off your team’s first commit to spend more time scoping them — let the coding begin, and look into them down the line. As a tech holder, you will likely be much more useful as a central repository for the project’s design than as one more person taking up stories.</li><li><strong>Optimize for “time to Minimum Viable Product (MVP)”.</strong> Most of the small details, edge cases and other improvements will probably be ten times quicker to detect once you ship your core features. In typical lean fashion, ship a minimum viable feature as soon as you can, hidden behind a feature-switch to avoid upsetting users, and crash-test the hell out of it. Magically, you’ve saved precious time by not building complex hypotheticals, and you can now use it to clean up any quirks and polish your feature.</li></ul><p>Paradoxically, this kind of scoping demanded much more discipline from me than the other one. It’s actually very hard to let go of working out everything in detail. But it became clear by the project’s end that not only did this approach work, it was pretty much the only viable approach in our situation.</p><p>For the record, I didn’t exactly scope the project in a day, but rather, over two to three days, plus all the time spent throughout the actual coding. This is still massively different from the two weeks I spent on my first project, which was comparable in complexity, and given the short timeframe we had for delivery, it made all the difference in the world.</p><h4>Lessons learned</h4><p>If you had told me about those two types of scoping in theory before I had done any, I would have said that the beaver-dam version is the better one. The last thing a project needs is unpredictability. The risks are many: developers playing treasure hunt for unknown lengths of time, shipping a broken or incomplete feature, complete project rewrite after discovering a hidden problem…</p><p>But then,</p><ol><li>Does extensive scoping really eliminate those risks?</li><li>Even if it does, is it worth two full weeks of developer time?</li></ol><p>Now, after my two experiences, it might be tempting to answer both questions with ‘no’, and declare the duct-tape scoping better. After all, it delivered comparable success, with much less time spent scoping.</p><p>You know where this is going, so let’s at least wrap it in a neat metaphor: between a flathead and a Phillips-head screwdriver, neither one is better than the other. What’s better is having a well-furnished toolbox, and knowing when and how to use the tools in it.</p><p>Some projects warrant investing in agility. Others in predictability. Some are worth building the huge charts, documenting the precise tech choices for. Others are better approached with the leanest mindset. All projects need a dedicated team, with a tech holder ready to own its tech choices.</p><h3>What’s next?</h3><p>If you want to learn more about our tech team we write a weekly newsletter, you can sign up for <a href="https://doctolib.us14.list-manage.com/subscribe?u=6f3cf1e1e4554097575191f5d&amp;id=a22645bb6c">here</a>. And if you want to join us, <a href="https://about.doctolib.com/jobs">we are hiring</a>!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ab8d1915ec5a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/doctolib/the-two-kinds-of-tech-scoping-ab8d1915ec5a">The two kinds of tech scoping</a> was originally published in <a href="https://medium.com/doctolib">Doctolib</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Soufflé, the King’s Pawn, and other stories]]></title>
            <link>https://medium.com/@emmanuel.gautier/the-souffl%C3%A9-the-kings-pawn-and-other-stories-631cfd804c4b?source=rss-fbe4cf777dd1------2</link>
            <guid isPermaLink="false">https://medium.com/p/631cfd804c4b</guid>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[future-of-work]]></category>
            <category><![CDATA[future]]></category>
            <dc:creator><![CDATA[Emmanuel Gautier]]></dc:creator>
            <pubDate>Thu, 20 Feb 2020 15:46:42 GMT</pubDate>
            <atom:updated>2020-02-21T13:30:58.971Z</atom:updated>
            <content:encoded><![CDATA[<h4>What tech in the West may look like in the 2020&#39;s.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HiQ2winZ9hYxznadkfinSg.jpeg" /><figcaption>Photo by <a href="https://unsplash.com/@guilhermestecanella">https://unsplash.com/@guilhermestecanella</a></figcaption></figure><p>It’s the 31st of December, 2029. Nine billion people are preparing to celebrate the passage to a brand new decade. As is now customary of new-year’s, by virtually any metric, more data will be shared in the upcoming twenty-four-hour slice than on other day in human history. Sure, the ritually expansive emoji-laced, filter-coated tide of messages, selfies, and stories is more or less inevitable by now. But this time, reflecting on the 2020’s, the tech world can use a milestone like this one.</p><p>Some analysts predicted a new Internet Bubble at the end of the 2010’s. What we got instead was an <strong>Internet Soufflé</strong>. It wasn’t the spectacular crash that rattled Silicon Valley for the first twenty months of the millenium. Rather, after a string of disappointing IPOs from promising unicorns, investors carefully allowed their outpour of money to ebb over the years, while diversifying their portfolios for smaller deals.</p><p>The most interesting side effect is that the Valley became a tiny bit less crazy. With steamroller startups less financially able to drive wages and rents way up, the place was finally able to breathe, and slightly bridge its inhabitability gap with non-tech people. This was further amplified by the other great trend in the tech industry…</p><h4>🏡 The R<strong>emote Boom.</strong></h4><p>At the end of the 2010’s, a moderate but growing fraction of the global office workforce worked remotely. By the mid-2020’s, most office workers were working away from the office at least once a week. 2028 saw the threshold of half of all working hours performed remotely be pulverized. At the same time, in tech, this number was sky-rocketting to a dizzying 78%. Whatever drawbacks were pointed out about remote work were mostly swept aside by the overwhelming satisfaction it was bringing employers and employees alike, to the relief of most transportation infrastructure.</p><p>The one issue that <em>did</em> arise, was that people were becoming even more sedentary. Although the thriving gyms market did compensate for the shift with some, morbid obesity is now looking like the runner-up for great health catastrophe of the 2030’s, second only to the un-ending environmental disasters and their related conflicts.</p><p>But that didn’t stop an entire chunk of tech from throwing its weight behind the remote el-dorado. Everything, from cloud collaboration tools to video-conferencing hardware, was revisited. It even kickstarted an apparent Google renaissance; the firm finally found a way out of its monolithic revenue structure, by charging an affordable $5/user/month for the premium version of its remote bundle. Its smashing success was hardly left to chance, given Google’s investments in smart video — that is, AI-buoyed media compression and enhancement technology. What could be more attractive than a videochat app where you’re certain to get pristine, glitch-less audio and video, even on a pre-fiber Internet access?</p><p>And of course, remote tools aren’t the only war in which smart video tech was weaponized, but we’ll get back to that.</p><h4>👥 Facebook is dead, long live Facebook.</h4><p>Back in 2020, if you asked a journalist, ‘What do you think will cause the downfall of Facebook, eventually?’, they would probably wave at an array of its most glaring issues: the privacy scandals, the predatory ad-based model, the fostering of all kinds of extremism… The truth is, most social media is dysfunctional, and their audiences hardly care. Anyone born after 2000 knows the actual reason. Facebook died just as MySpace, MSN, and AOL died: it wasn’t cool anymore.</p><p>Teenagers who used it for general laughs, selfies and gossip were already deserting <em>en masse </em>to Instagram, Snap and TikTok by the late 2010’s, and by 2023, there wasn’t a soul under 25 there. The twenty-somethings who used it for vaguely staying in touch with relatives, pop culture discourse and activism left for a variety of fragmented, small-community-based platforms that thrived in Reddit’s wake. But when looking for a particular culprit for Facebook’s end, the one with the biggest smoking gun is Gather, an app that came out mid-decade, bent on destroying the one reason millenials were still using Facebook: events planning.</p><p>So, #FaceBye made many people happy, but chief among them? Mark Zuckerberg. Finally unburdened of a flailing venture, he was able to concentrate on platforms that worked better (Instagram, Messenger, and a few newcomers). It’s no secret that this was the best thing to happen to Facebook, Inc. in many, many years.</p><h4>🗞 Old spreadsheets die hard.</h4><p>Away from friends getting together, and in the noticeably different social circle that is work, it’s the traditional recipes that thrived. In 2030, the plain old email will enter an impressive fourth decade of marketing supremacy. That’s true of both commercial emails and personally-maintained, niche newsletters. In the end, all the web platforms did for business influencers was draw in more inboxes, ever the places of choice for value delivery.</p><p>Fittingly, the technology that shared the success of this kind of insights-sharing business is about as old as email itself: spreadsheets. Remember 2012, when a Google Search for the best notes apps, CRM suite, or Photoshop tutorials would take you to obscure, unhelpfully curated, ad-riddled tech websites? That’s a thing of the past now. The business insight currency of the 2020’s has been the public spreadsheet, minutely curated by passionate people.</p><p>Covering everything from public datasets to links aggregation, often enriched with precious comparative details, and sometimes maddeningly interlinked, Google Sheets has been dubbed ‘the Web within the Web’, and some people seriously consider it a social network ten times LinkedIn’s size.</p><p>For people who weren’t around when the Web began, it’s the closest we’re getting to the excitement of the early days, and the best part is, it feels as though this is just the tip of the iceberg of potential for this brand new, infinitely celled world.</p><h4>🎥 The First Streaming War.</h4><p>If you think the streaming revolution of the 2010’s was crazy, think again. It’s only once the four main players — the DANA, we call them — were on the pitch, that the confrontation really began.</p><p>The first shot was drawn when Disney+ surprise-dropped an MCU movie about Thanos’ origins on Christmas Day, 2020. It became the most-streamed feature-length movie of the platform in just a week, drew heavy praise from Marvel fans, and reportedly attracted one million new customers to the service over the next month.</p><p>The next year, came Amazon’s blockbuster <em>Lord of the Rings</em> show, which, while massively successful, fell just short of achieving its ostentatious goal of becoming the <em>Game of Thrones</em> of the 20’s. That crown was snatched by Netflix, who achieved planetary supremacy with its all-Chinese, futuristic, low-fi, ensemble-led epic, titled <em>Windwalkers</em>, a loose adaptation of a French sci-fi novel.</p><p>Competitors left no stunt un-pulled — the <em>Friends</em> reunion feature-length movie! Interactive <em>Dora the Explorer</em>! Randomized storylines! Seamless, one-episode, ten-hour seasons! Amazon buys HBO! This vertiginous content war left one player wounded: Apple TV+. Somehow stuck in its old-fashioned cool, premium brand, unable to bring anything game-changing to the table, the service retreated to a niche of prestige, traditionally-made content. It would probably have closed down, if it weren’t for the one trick Apple had up its sleeve: smart video.</p><p>What Google did for remote work, Apple pledged to do for entertainment streaming. The company found a new revenue stream in becoming the sole provider of smart video tech for its three competitors, keeping its own (small) catalog alive as a showcase of its sanity-defying advances in AI-powered, inexpensive image quality.</p><p>All in all, the first streaming war ended at a major standstill between its four main players, and it’s easy to see why: it’s very hard to starve one competitor out of the cake, when the cake is growing faster than anyone can control. The actual story behind these developments? The ritual killing of antiquated forms of film consumption; cable, cinemas, physical home video are, if not on their deathbeds, migrating to a more modest state of existence.</p><h4>🦠 #TheKingsPWN</h4><p>Everyone with an Internet access remembers where they were on June 7, 2026. What only few remember, is that it begins with Instagram and Twitter publishing a joint blog post, disclosing a security breach they had uncovered.</p><p>Some time in 2017, someone published a small utility package for Node.js. It appeared to provide string management functions for developers. Deep down, below a few layers of encryption and bad-faith code minification, was a deathly payload, filter.js. It would run silently and do one simple thing: duplicate any network requests with an image file, and send them to a specific server. Diabolically tailored to disable itself whenever running inside debugging tools, or test environments, filter.js was virtually undetectable.</p><p>The package’s author pull requested his way to adding it as a dependency to dozens of repos, and by 2019, his code was running on the world’s largest apps and websites.</p><p>And that’s pretty much all we know about the guy, save for one thing: he was waiting to be discovered. Not three hours after the Twitter/Insta blog post, detailing that they had alerted most industry players and asked NPM to kill the malicious package, he released everything he had. And so came the day that eight years of pretty much everyone’s private photos were published online.</p><p>First, the bots. For every Twitter account @user, they would create an evil twin named @user_tkp_bot tasked with tweeting <em>en masse</em> that user’s DMed pics. The bots’ profile descriptions would advertise the website tkp-twitter.com, hosted far from the reach of official Twitter moderation, where any lay-online-person could find the full, easily searchable database. There was a tkp-reddit.com, a tkp-linkedin.com, a tkp-slack.com, you name it, all under the mysterious banner of The King’s PWN.</p><p>After the planet was done dealing with the maelstrom of difficult political, professional and personal conversations triggered by the trillion-photo hack, a number of things happened:</p><ul><li>Spectacularly, people didn’t change their social media habits. The absence of one impacted app to be deserted overwhelmed any attempt to perform a rational switch.</li><li>End-to-end encryption did however become a thing, after a few years. A reasonable portion of people outside of tech became aware of what it was and why it mattered. Less acknowledged by the public, Content Security Policies became a widely adopted standard, too.</li><li>The main tech casualty turned out to be open-source-software. As developers at large companies stopped popping dependencies like candy, an entire ecosystem suffered a massive slowdown, followed by concentration around larger, more structured and more reliable projects. Small packages, already not viable for their maintainers as it was, were killed in troves by declining usage. Attempts at security certification were made, and remained just that: attempts.</li></ul><p>To this day, the motives of TKP’s author remain an enigma. Some say he was trying to teach us a lesson about privacy. Others point to a state actor trying to disrupt… the world? But the one thing that stayed with me is when I heard someone online describe it as a new kind of art form.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=631cfd804c4b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Wrapping my head around the monolith: my first months at Doctolib]]></title>
            <link>https://medium.com/doctolib/wrapping-my-head-around-the-monolith-my-first-months-at-doctolib-a8c48a5f88d0?source=rss-fbe4cf777dd1------2</link>
            <guid isPermaLink="false">https://medium.com/p/a8c48a5f88d0</guid>
            <category><![CDATA[ruby-on-rails-development]]></category>
            <category><![CDATA[onboarding]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[test-driven-development]]></category>
            <dc:creator><![CDATA[Emmanuel Gautier]]></dc:creator>
            <pubDate>Fri, 24 Jan 2020 16:12:53 GMT</pubDate>
            <atom:updated>2020-02-18T09:19:48.718Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3sQ9qF5EEnO4jhV8Ve-95A.png" /><figcaption>Illustration by Garance Bigo</figcaption></figure><p>After two months at Doctolib, I wrote a short piece to share my new joiner perspective on the company. It was the end of December so it was pretty calm in the office and I wanted to share some feedback with the team. I was contacted by the team that runs the blog and they encouraged me to share my feedback with the world, so here it is. Here are my thoughts on what it is like as a new developer joining Doctolib in 2020.</p><h3>Organization</h3><h3>#PeopleOverProcesses</h3><p>This isn’t easy to uphold in a thousand-people company, but the spirit really is here. I get help whenever I need it, and the rest of the time, I enjoy autonomy, engagement, and a freedom to take initiatives. Safeguards feel barely there (even though they are), which is encouraging.</p><p>As a corollary, however, I can tell that this balance is fragile. It’s easy to dip into either bureaucracy or chaos.</p><p>In this regard, the private StackOverflow we have is a good example of something that works well. It’s not an attempt to hoard every single piece of information in the company, and yet it comes in very handy for specific questions. I will try to use it more systematically.</p><h3>#Onboarding</h3><p>I thought the DA (<em>editor’s note, DA is what we call our Doctolib Academy that all our team do in their first few weeks</em>) and the DA tech were an excellent onboarding experience. Again, knowledge management is critical in any company, and it just felt right to begin my job with an intense crash course on everything at Doctolib.</p><p>If I have a minor quibble, though, I wish the tech modules of the DA had been packed right behind the general DA, instead of stretched over weeks, because:</p><ol><li>It got in the way of a lot of things when I was taking up my first tasks in my team,</li><li>By coming in later, it inevitably covered some topics which I had found out about on the job.</li></ol><p>I can only imagine the logistics gauntlet it is to make the DA happen, but I feel like it could have an even larger impact if it came earlier.</p><p>Starsky&amp;Hutch is also a fantastic thing, and a bit like the DA tech, I only wished it had come even sooner. To those of you who may not know this, Starsky&amp;Hutch is the technical mentoring system at Doctolib, where you go through Rails and React 101 and are bootstrapped by a mentor to learn the stack. I really enjoyed it!</p><p>As a very minor last remark, I noticed that new joiners (myself included) have had to make quite a bunch of requests, even though they are the same for everyone. For straightforward things like access to the GitHub repo and particular mailing lists, I’m sure there is a fix for this.</p><h3>Code</h3><p>I had never worked on a codebase this large, by far, so I was surprised by a lot of things.</p><h3>#Rails</h3><p>Everyone told me, but I had to experience it to believe it: learning Rails is difficult, frustrating even, for about a month, and then it starts paying off very, very quickly. I don’t know why I couldn’t quite come around to trusting “you’ll see, give it a month and it gets much better” — likely my imposter syndrome at work but it turned out I was dead wrong.</p><p>I think there are two main reasons for this steep learning curve. One is obviously “Convention over Configuration”. The other is a concrete manifestation of it: there simply is a ton of implicit code in Rails. The first time you bump into, say, a <em>&lt;controller_action&gt;_path</em> or <em>&lt;attribute&gt;_was </em>-type helper, you just can’t find an explicit definition for it, and Google will likely prove unhelpful. Yet, in retrospect, what these methods do is both painfully obvious and devilishly helpful. The first times, this can lead to difficult code exploration sessions, and even yield some negative feelings (“Was I supposed to just <em>guess</em> this?”).</p><p>But that’s where perseverance and, above all, mentoring by teammates help get through. Again, on the other side of the tunnel, the payoff is massive.</p><h3>#TDD</h3><p>I thought that TDD (that’s Test-Driven Development), which I was also new too, would be one those good practices which, though I fully support in the abstract, are too demanding to observe in practice, like journaling, or giving up saucisson.</p><p>In fact, I found that TDD is stunningly easy to take up, and astonishingly satisfying. It actually makes a lot of sense: TDD comes with its built-in reward mechanism. I write a test, implement, pass the test, and iterate. It’s like a build-your-own-hero’s-journey. In addition to this being riotously motivating, the comprehensive test suite we have makes this applicable to almost any use case.</p><p>Even for early code fiddling, where I used to practice a savage form of console-driven-development or Chrome-dev-tools-driven-development, I can now say that in 90% of cases, I find myself using tests.</p><p>Two caveats:</p><ul><li>The only part of TDD that requires an effort on my part is writing tests <em>before</em> jumping to an implementation. I’m getting better at this though.</li><li>For front-end stuff, I wish Capybara were set-up to spin Chromes with the React dev tools. It would be so helpful when I need to pick apart the components on a page and figure out which does what. (<em>Editor’s note, we have now updated our React dev tools to include this</em>)</li></ul><h3>#Monolith</h3><p>The monolith is impressively robust. You can actually go ahead and make changes without knowing 100% of how everything works, and with little risk of breaking everything…</p><p>…and yet, it feels kind of inevitable that you get flakies on a regular basis. I got a lot of them on my first PRs (by chance I think), and they feel pretty destabilizing when you don’t know what’s normal and what’s not. They get much, much less annoying after a while.</p><p><strong>In conclusion,</strong></p><p>The first few months have been really challenging, lots of learning, lots of new people... A moment of information overload which had me frustrated at the time. However, with hindsight, it was incredibly rewarding. I have learned so much and the team here has done a great job of helping me get settled in and I now really feel part of the team.</p><p>Now I feel like a true Doctoliber and I can’t wait to change healthcare for good.</p><p><a href="http://bit.ly/doctotechlife">Sign up for our newsletter</a> to get more news about our team and all our favourite Ruby on Rails and JavaScript articles.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a8c48a5f88d0" width="1" height="1" alt=""><hr><p><a href="https://medium.com/doctolib/wrapping-my-head-around-the-monolith-my-first-months-at-doctolib-a8c48a5f88d0">Wrapping my head around the monolith: my first months at Doctolib</a> was originally published in <a href="https://medium.com/doctolib">Doctolib</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>