<?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[door2door Engineering - Medium]]></title>
        <description><![CDATA[door2door&#39;s space where our developers share insights into their daily work and the amazing products they build. - Medium]]></description>
        <link>https://medium.com/tech-blog-door2door?source=rss----b94e3e744719---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>door2door Engineering - Medium</title>
            <link>https://medium.com/tech-blog-door2door?source=rss----b94e3e744719---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 02:38:43 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/tech-blog-door2door" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Introducing our Female Software Developers: Leoni]]></title>
            <link>https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-leoni-43f7b29d0e00?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/43f7b29d0e00</guid>
            <category><![CDATA[mobility]]></category>
            <category><![CDATA[berlin-start-ups]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[women-in-engineering]]></category>
            <category><![CDATA[tech-jobs]]></category>
            <dc:creator><![CDATA[door2door]]></dc:creator>
            <pubDate>Mon, 06 Dec 2021 08:49:08 GMT</pubDate>
            <atom:updated>2021-12-06T08:49:08.851Z</atom:updated>
            <content:encoded><![CDATA[<p>Episode 6: Leoni</p><p><em>We would like to take a moment and shine the light on our female engineering colleagues, in a series of short interviews. We hope that this will inspire more women to join the tech world and make a difference! This week we are introducing #{name}</em></p><pre>name = Leoni</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*roKQ4t84qdCM81CNIVUDEg.jpeg" /><figcaption>Leoni at coding bootcamp Le Wagon</figcaption></figure><p>Leoni joined door2door in March 2017 and her original position was Assistant to the CEO. It was here that she was first introduced to software development. In January 2020 she joined a coding bootcamp at Le Wagon in Berlin and in April she returned as a Junior Software Developer for the Bookings Team and is now on her way to becoming a Fullstack Developer in the Driver and Dispatch team.</p><p><strong>When did you first start writing code?</strong></p><p>If you exclude several attempts here and there, I seriously started learning how to code in November 2019.</p><p><strong>What was the first language you started in?</strong></p><p>I started with a mix of Ruby (vanilla &amp; Rails), HTML, CSS and some Javascript basics.</p><p><strong>Do you have a preference between frontend or backend development?</strong></p><p>I find both equally interesting, actually! I really love the instant feedback you get from frontend but also like everything you can do with the backend.</p><p><strong>What was your biggest struggle when starting out?</strong></p><p>My biggest struggle was to not compare myself to others. Sometimes there were a few concepts that just wouldn’t get into my head. It was hard for me to not get disappointed about the progress my peers seemed to be having. What really helped me was to start comparing myself with myself and all the progress I have done. I also created a list to write down my biggest achievements and learnings. It’s really satisfying to revisit the list and see how it gets longer and longer.</p><p><strong>Do you have any learning tips?</strong></p><p>What has worked well for me is to have focus times, think of the pomodoro technique. Another thing that has helped me is to have my own “documentation” of things that I found helpful or bugs that I have encountered and what I did to solve them. I don’t always go back to read it but documenting things like that helps me solidify the knowledge.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Y2z1LaIS4KzwhGcwTg1_0Q.jpeg" /><figcaption>Leoni and Nadine at a Wooga game coding workshop</figcaption></figure><p><strong>What made you want to become a developer?</strong></p><p>The idea of being able to build software that can be easy to use by as many people as possible. I also really like the flexibility it provides — you can basically work from anywhere!</p><blockquote>It was hard for me to not get disappointed about the progress my peers seemed to be having. What really helped me was to start comparing myself with myself and all the progress I have done.</blockquote><p><strong>What do you like most about being a developer?</strong></p><p>That I can learn something new every day. There are so many programming languages, resources and projects out there that even if you spent every second of your life on them, there would still be new things to learn. It’s constantly evolving!</p><p><strong>What would you advise aspiring developers?</strong></p><p>Just start! There are many free resources available that can help you practice and learn!</p><p><strong>Do you have any pet projects? If yes, which one?</strong></p><p>I am working together with a friend on a project called Dateros, which is a data-based project following the political situation in Peru.</p><p><strong>What do you enjoy doing apart from programming?</strong></p><p>Many, many things, but to narrow it down I really love doing pottery and drawing!</p><p>–</p><p><em>Thank you Leoni for this very encouraging and insightful interview! If you enjoyed this chat and want to join us, check out our </em><a href="https://door2door.io/en/about-us/jobs/"><em>job postings</em></a><em> and see you in the next episode!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=43f7b29d0e00" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-leoni-43f7b29d0e00">Introducing our Female Software Developers: Leoni</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The most extraordinary hiring process: Meet Joshua!]]></title>
            <link>https://medium.com/tech-blog-door2door/the-most-extraordinary-hiring-process-meet-joshua-b52c7e1ec213?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/b52c7e1ec213</guid>
            <category><![CDATA[tech-jobs]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[tech-startups]]></category>
            <category><![CDATA[team-culture]]></category>
            <category><![CDATA[berlin]]></category>
            <dc:creator><![CDATA[door2door]]></dc:creator>
            <pubDate>Tue, 09 Nov 2021 08:01:19 GMT</pubDate>
            <atom:updated>2021-11-09T08:01:13.924Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*auRB0ie6-foXz1NV6SqgKg.png" /></figure><p><em>In this blog post interview, we would like to give you some insights into a quite extraordinary recruitment process. High bureaucratic hurdles led to a two-year hiring process for Joshua from Lagos (Nigeria), but neither Joshua nor door2door gave up and today he works as a software engineer in the Analytics team. But let him tell his story:</em></p><p><strong>How long have you been working for door2door and what is your position?</strong></p><p>I started on 1 July 2021. My position is that of a software engineer in the Analytics team.</p><p><strong>What are you doing there exactly?</strong></p><p>It’s a mix of web development, data engineering and analysis. The Analytics team is responsible for providing insights about the operations of our services. We know there’s a wealth of information locked in the driver and passenger behaviour. This data is relevant to understanding the market and the service, and as a member of the team, I build pipelines that ingest and transform this data and also visualize it on dashboards.</p><p><strong>How did you find out about door2door?</strong></p><p>That was in July 2019. I had gotten involved in big data technology and was looking for a full-time role in that space. So I went searching on Twitter and I saw a tweet that said: “We’re hiring at door2door.” I read it and it sounded good. I sent a direct message and said: “I love the purpose of the company and the description of the job. I am interested.” I got a response encouraging me to apply. I did, and the HR department contacted me the next day to start the process with interviews.</p><p><strong>What happened then?</strong></p><p>I had an introduction interview. I was given a test which I had to submit in seven days. I did it, I got a positive review, and I was moved to the final interviewing round with the Engineering team. That also went well and I got the offer which came with relocation to Berlin. The company tried to get me an appointment at the German consulate, but the website stated the wait period was 16 months. I also tried myself but I saw the same message, 16 months! door2door had to fill the position and couldn’t wait that long, which I totally understood. We both had to move on.</p><p><strong>Why does it take so long to get an appointment at the consulate?</strong></p><p>I think there are a lot of appointments in the queue and it just takes time to get through all of them. This was also pre-pandemic. I’m sure the lockdown must have further slowed down the process.</p><p><strong>Did you really make an appointment for 16 months later?</strong></p><p>Yes, back in 2019, I booked the appointment hoping things would get sorted out once it was registered in the system. I waited for a while, but nothing changed. Months later, I accepted another job offer.</p><p>Fast forward to 2021: I received an email from the consulate stating that my appointment would be in 30 days. So I had 30 days to reach out to door2door, find out if there was an opening, and see if I could get a renewed offer. I contacted door2door’s HR department via email saying: “Hello, do you remember me from 2019? I have an appointment now.” I wasn’t confident things would work out, but I got a reply saying: “Yes, we remember you, the role is gone. But we can talk to the teams to know if there is a need.” Then they came back and said: “Yes, there is a position in the Analytics team”. I was happy, my initial role was a full stack developer role; this role was similar, but revolved around big data, which was perfect. The team sent me a test, I completed the test, they liked the result and I had interviews with members of the team.</p><p><strong>You had to do the whole process again?</strong></p><p>Yes, I was tested on data pipelines, analysis and modelling, which is different from the test I had initially done. It was a fast process. The company understood that I was on a tight schedule, so the process was done in good time. I remember Thiago had to take a call with me while he was away on vacation.</p><p><strong>Did everything go smoothly after you got the new job at door2door?</strong></p><p>Not right away. One of the required documents for the interview was an affidavit. I didn’t have the document and the courts in Nigeria were on strike at the time. I was troubled by this, because I didn’t want my application to get rejected on this basis. Fortunately, a week before the appointment, the judicial chapter in my city relaxed the strike and partially resumed activities and I got the document.</p><p>There was another scare. After the interview, there was an issue with the recognition of my university degree, and the process of getting a degree recognized starts with sending the transcript from the university. The transcript process is usually slow in many Nigerian universities. It sometimes takes months to process, and I got worried. But fortunately again, the issue was resolved, my degree was recognized in the database and I received the visa.</p><p><strong>Did you ask the consulate why it all took so long?</strong></p><p>No I didn’t, the funny thing was during the interview, the visa officer saw my contract and said: “This contract is from 2019. What took you so long?” It was amusing, because I felt that was what I should have been asking.</p><p><strong>Why were you so keen to join door2door?</strong></p><p>I was sold on the mission “A city for people, not cars”. Where I’m from, it’s not unheard of to spend three hours commuting between home and the workplace. Vehicle congestion is the culprit most times, and this affects many things: the environment, quality of life, productivity, etc. I believe what we do as a company is important for sustainable life and I am happy that my love for building data-driven solutions is aligned with this mission.</p><p>–</p><p>Thanks, Joshua, for the interview! We are very happy to finally have you here on the door2door team.</p><p>Would you also like to apply at door2door? Have a look at our <a href="https://door2door.io/en/about-us/jobs/">job postings</a>. We’re also happy to help you with a move, and we really hope it doesn’t take as long as it did for Joshua :)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b52c7e1ec213" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/the-most-extraordinary-hiring-process-meet-joshua-b52c7e1ec213">The most extraordinary hiring process: Meet Joshua!</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introducing our Female Software Developers]]></title>
            <link>https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-8e191a7e8ff?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/8e191a7e8ff</guid>
            <category><![CDATA[tech-jobs]]></category>
            <category><![CDATA[berlin]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[female-engineers]]></category>
            <category><![CDATA[engineering]]></category>
            <dc:creator><![CDATA[door2door]]></dc:creator>
            <pubDate>Mon, 01 Nov 2021 08:01:11 GMT</pubDate>
            <atom:updated>2021-11-01T08:01:11.255Z</atom:updated>
            <content:encoded><![CDATA[<p>Episode 5: Bowie</p><p><em>We would like to take a moment and shine the light on our female engineering colleagues, in a series of short interviews. We hope that this will inspire more women to join the tech world and make a difference! This week we are introducing #{name}</em></p><pre>name = Bowie</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NE6Q4voUuuKeUBoy57QAIQ.jpeg" /></figure><p>Having worked as a full-stack developer for four years, Bowie joined door2door in October 2019 as a Software Engineer. She graduated from the University of Waterloo in Canada with a Geomatics degree as her major and Computer Science as her minor. After graduating, she decided to follow her passion and chose computer science as her future career.</p><p><strong>When did you first start writing code?</strong></p><p>My first computer science course was in VBA in high school.</p><p><strong>What was the first language you started in?</strong></p><p>The first language I seriously started in was C language.</p><p><strong>Do you have a preference between frontend or backend development?</strong></p><p>I prefer the backend.</p><p><strong>What was your biggest struggle when starting out?</strong></p><p>It was hard to connect the theories I learned in class to real-life applications, especially web applications. In class, we learned things like network, database, data structure and so on individually. It was a bit hard for me to figure out how they could be grouped together in applications. Also, because we use pre-defined wireframes and packages in real-life web development, we sometimes don’t pay attention to the actual implementation behind them.</p><p>Another factor was that because I only minored in Computer Science, I never actually took a course on operating systems. So understanding the server side was initially an obstacle for me.</p><p><strong>Do you have any learning tips?</strong></p><p>Ask for recommendations from senior developers, who can usually give you better learning materials. Keep learning consistently. Computer science is a field in which you always have to learn new things. Always try to spend some spare time on self-improvement in the direction you are most interested in.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/258/0*NgcFmaZP5DO2aMdT.png" /></figure><p><strong>What made you want to become a developer?</strong></p><p>I love the fact that it can be challenging, motivates me to keep learning and can be applied in every field.</p><p><strong>What do you like most about being a developer?</strong></p><p>The feeling of accomplishment I get when I finish a project, fix a bug or solve a problem, no matter if it is from working independently or team work.</p><p><strong>What would you advise aspiring developers?</strong></p><p>Pair coding. Through my experience at door2door, I found that pair coding is actually a great way to motivate developers in a team to share ideas, learn from one another, and surprisingly, is much more efficient than I had imagined.</p><p><strong>Do you have any pet projects? If yes, which one?</strong></p><p>No, currently I am not working on any side projects.</p><p><strong>What do you enjoy doing apart from programming?</strong></p><p>Travelling. Because I come from a geography background, I really enjoy exploring different terrains and experiencing different cultures.</p><p>–</p><p><em>Thank you Bowie for this very fun and insightful interview! If you enjoyed this chat and want to join us, check out our </em><a href="https://door2door.io/en/about-us/jobs/"><em>job postings</em></a><em> and see you in the next episode!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8e191a7e8ff" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-8e191a7e8ff">Introducing our Female Software Developers</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introducing our Female Software Developers]]></title>
            <link>https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-zofia-4dbb7cf7403?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/4dbb7cf7403</guid>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[female-engineers]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[berlin]]></category>
            <category><![CDATA[tech-jobs]]></category>
            <dc:creator><![CDATA[door2door]]></dc:creator>
            <pubDate>Thu, 28 Oct 2021 08:46:56 GMT</pubDate>
            <atom:updated>2021-10-28T08:49:58.011Z</atom:updated>
            <content:encoded><![CDATA[<p>Episode 4: Zofia</p><p><em>We would like to take a moment and shine the light on our female engineering colleagues, in a series of short interviews. We hope that this will inspire more women to join the tech world and make a difference! This week we are introducing #{name}</em></p><pre>name = Zofia</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4olTEB5Axo6_Im1SeRwgZA.jpeg" /></figure><p>Zofia joined the door2door Passenger Team in March 2021 as an iOS-focused Software Developer. After graduating in Physics in Warsaw, she started her iOS journey at the Apple Developer Academy in Naples.</p><p><strong>When did you first start writing code?</strong></p><p>I wrote my first lines of code during my first year of university (I studied physics), which was really my first experience with any code at all. And it wasn’t easy, as most students already had some coding experience. So we covered topics pretty quickly.</p><p><strong>What was the first language you started in?</strong></p><p>C++, at my university.</p><p><strong>Do you have a preference between frontend or backend development?</strong></p><p>I don’t have too much experience with backend, as I’m an iOS developer. I’d like to try it one day. But as I have a creative soul I believe I still would prefer frontend.</p><p><strong>What was your biggest struggle when starting out?</strong></p><p>Comparing myself to others. Compared to other physics students, I started coding pretty late. Also, I didn’t know how to learn or which resources to use — as learning to code through “typical learning methods” like books isn’t really a good idea.</p><p><strong>Do you have any learning tips?</strong></p><p>Don’t compare yourself to others, just do your job. Find a problem that interests you and start creating small pet projects to solve that problem. Finding a mentor can be very motivating and helpful. You can find one in Facebook groups or maybe at local events (after the pandemic ends).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/0*2ko2b8J_KCC8Wods.jpg" /><figcaption>Mind the Gap Google event in Warsaw</figcaption></figure><p><strong>What made you want to become a developer?</strong></p><p>I had no clue what coding was when I was in high school. I didn’t even think about becoming a programmer back then. I had already decided that I would study physics and math. After high school, I was invited to Google in Warsaw for an event called “Mind the Gap”. They invited girls that got at least to the second round of the Mathematical Olympiad. At that event, I participated in a live coding session with Python. I became really interested in it, and it was my first trigger to find out more about coding. Later, at university, I took C++ classes. They were quite difficult for me, but there was one moment when I solved some additional homework and got really excited about it. I think what made me want to become a developer were these small moments of excitement while solving problems — moments when I was proud of myself and I just enjoyed it.</p><p><strong>What do you like most about being a developer?</strong></p><p>Solving and discussing problems. I like creating new things. Surprisingly, a developer job is all about working with people and I love that.</p><blockquote>“Find a problem that interests you and start creating small pet projects to solve that problem.”</blockquote><p><strong>What would you advise aspiring developers?</strong></p><p>If you like it, don’t give up. If you need to, find a mentor. Don’t be afraid to ask questions. We’ve all been there, and we still are…</p><p><strong>Do you have any pet projects? If yes, which one?</strong></p><p>I have one at a very early stage. It’s a platform for learning a language from authentic materials. We’ll see how it goes, as from experience I know I need to rest and pursue other hobbies as well. It’s just not possible to code every day for hours at a time.</p><p><strong>What do you enjoy doing apart from programming?</strong></p><p>I love dancing, particularly ballroom dancing. I’m also into brush lettering and I have my own bullet journal. I also like to learn languages. As I’ve moved to Berlin my focus is currently German, but I am also slowly getting back into Italian.</p><p>–</p><p><em>Thank you Zofia for this very encouraging interview and welcome to the team, we are so happy to have you! If you enjoyed this chat and want to join us, check out our </em><a href="https://door2door.io/en/about-us/jobs/"><em>job postings</em></a><em> and see you in the next episode!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4dbb7cf7403" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-zofia-4dbb7cf7403">Introducing our Female Software Developers</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introducing our Female Software Developers]]></title>
            <link>https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-7e720c30ba8?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/7e720c30ba8</guid>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[female-developer]]></category>
            <dc:creator><![CDATA[door2door]]></dc:creator>
            <pubDate>Thu, 28 Oct 2021 08:46:23 GMT</pubDate>
            <atom:updated>2021-04-13T11:02:51.989Z</atom:updated>
            <content:encoded><![CDATA[<p>Episode 3: Laras</p><p><em>We would like to take a moment and shine the light on our female engineering colleagues, in a series of short interviews. We hope that this will inspire more women to join the tech world and make a difference! This week we are introducing #{name}</em></p><pre>name = Laras</pre><figure><img alt="picture of Laras" src="https://cdn-images-1.medium.com/max/800/1*dsWepFJKuJEvy3cX_nTlfg.png" /></figure><p>Laras joined door2door in March 2018 as a Software Developer Intern, focusing on developing and maintaining the company’s static website. For 6 months, she worked on the frontend with HTML, CSS, and JavaScript. After finishing her internship, she started working on a machine learning project for her thesis with Python and its amazing libraries like pandas, geopandas, and scikit-learn. This ignited her passion for (geospatial) data analysis and visualisation. After graduating, she was given the chance to further develop her skills at door2door as a Junior Software Developer. About 1 and a half years later, she is now a Software Developer in the Analytics Team.</p><p><strong>When did you first start writing code?</strong></p><p>When I was studying at an international student college in Berlin (pre-university) in 2015.</p><p><strong>What was the first language you started in?</strong></p><p>Pascal (yeah, I know it’s old).</p><p><strong>Do you have a preference between frontend or backend development?</strong></p><p>When I started doing my thesis project, I found out I have a deeper love for backend development. I basically love both, just like I love both data analysis and visualisation. But data analysis, feature engineering, and data cleaning are things that intrigue me the most.</p><p><strong>What was your biggest struggle when starting out?</strong></p><p>I started with Pascal, which is a very procedural programming language, so it kind of surprised me in a way that I felt forced to be a very well-structured person (well yes, we do have to structure our code well as a programmer/developer, but Pascal is just… something, in my opinion). It took time for me to be able to think of well-structured algorithms that make sense. I was always the slowest in my class every time the professor gave us a task to solve, because I wasn’t used to thinking in a structured way. It was very hard for me to translate logic into code.</p><p><strong>Do you have any learning tips?</strong></p><p>Just get your hands dirty with something and keep doing it, one step at a time. Try different tutorials or crash courses; there are a lot of free ones online. Start a small project. That’s how I first got to know about machine learning during my thesis project (I had never learned about it before). Also, actually doing a project with someone as a mentor who can guide you is really helpful.</p><figure><img alt="A picture of Laras at a Python conference" src="https://cdn-images-1.medium.com/max/1024/1*WfHq9b1Fkj9mdIU5IwAbpA.jpeg" /><figcaption>Laras at PyCon</figcaption></figure><p><strong>What made you want to become a developer?</strong></p><p>The fact that I failed a lot in my programming class during my pre-university year made me even more curious and triggered me to practice more and more, so that I became obsessed with the feeling of satisfaction when you are finally able to solve a problem. But in all honesty, it’s probably because of J.A.R.V.I.S from Iron Man.</p><p><strong>What do you like most about being a developer?</strong></p><p>It’s gonna sound like a cliché, but the fact that you can learn a lot of new things everyday as the industry evolves, and at the same time you can be a part of it too. And you can actually contribute your learnings to help the industry evolve even further.</p><p><strong>What would you advise aspiring developers?</strong></p><p>Don’t be afraid or feel bad about making mistakes. Failing sucks, yes, but it is inevitable. Especially because being a software developer IS about being challenged by interesting problems and constant learning. Even senior developers still make mistakes. When you make mistakes, you will actually discover something new and learn from that for the next time. If everything feels slow right now (like it did for me) or like it’s not going anywhere or it feels like you aren’t as smart as Mark Zuckerberg — no matter what, if you love it, don’t stop. Find a mentor whom you can learn from. And write down your goals, so you can also track your progress.</p><p><strong>Do you have any pet projects? If yes, which one?</strong><br>Currently I don’t have any pet projects, because I prefer to use my free times for my other passions as much as I can to refresh 🙂 but I did a project in data science back then during my bachelor days to analyse the correlation between diversity in lead actors and the gross revenue of Hollywood movies</p><p><strong>What do you enjoy doing apart from programming?</strong><br>I’m all about films and acting 🎬 🎭 I really enjoy making self-tape videos and make a dedicated playlist for each character I play. I also spend a lot of time reading books 📚 (occasionally reading them out loud as if I’m doing a monologue haha) and, whenever possible, doing karaoke 🎤</p><p><em>Thank you Laras for sharing your awesome story with us, let’s be honest J.A.R.V.I.S </em><strong><em>is</em></strong><em> really cool! If you enjoyed this chat and want to join us, check out </em><a href="https://door2door.io/en/about-us/jobs/"><strong><em>our job postings</em></strong></a><em> and see you in the next episode!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7e720c30ba8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-7e720c30ba8">Introducing our Female Software Developers</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introducing our Female Software Developers]]></title>
            <link>https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-a4648dafe2f0?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/a4648dafe2f0</guid>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[female-developer]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[door2door]]></dc:creator>
            <pubDate>Mon, 01 Mar 2021 08:05:04 GMT</pubDate>
            <atom:updated>2021-03-01T08:05:04.673Z</atom:updated>
            <content:encoded><![CDATA[<p>Episode 2: Nadine</p><p><em>We would like to take a moment and shine the light on our female engineering colleagues, in a series of short interviews. We hope that this will inspire more women to join the tech world and make a difference! This week we are introducing #{name}</em></p><pre>name = “Nadine”</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*DiomdLiietIZxVj6.jpeg" /></figure><p>After graduating in the summer of 2016 with a Computer Science degree, Nadine joined door2door in September 2016 as an Android developer. She has since worked in multiple teams building and maintaining apps, and is currently working as a Senior Software Developer in the Passenger team, a cross-functional development team building all the backend and mobile components that allow our users to search for multi-modal routes and book rideshare trips.</p><p><strong>When did you first start writing code?</strong></p><p>If you count static websites with HTML and CSS then probably around the age of 13. I learned by looking at the source code of websites and copy-pasting bits that I liked to create my own websites. Then I started writing code “for reals” when I was 21.</p><p><strong>What was the first language you started in?</strong></p><p>HTML and CSS and then JavaScript as the first actual programming language.</p><p><strong>Do you have a preference between frontend or backend development?</strong></p><p>Mobile, which is frontend I guess 🙃 I don’t dislike backend, but I just can’t seem to warm up to Ruby, which is used in most backend services 🤷‍♀️ I quite enjoy working in our Kotlin service though.</p><p>For me there is no better way to learn than just playing around with a technology and trying stuff out until I understand what is happening.</p><p><strong>What was your biggest struggle when starting out?</strong></p><p>Not getting overwhelmed by the amount of things I didn’t know and filtering what is important to learn and what isn’t.</p><p><strong>Do you have any learning tips?</strong><br>Just dive in and get your hands dirty. For me there is no better way to learn than just playing around with a technology and trying stuff out until I understand what is happening.</p><p><strong>What made you want to become a developer?</strong></p><p>I was really bored in my old job and started to learn JavaScript through Codecademy when I had nothing to do at work. I really enjoyed it and decided to study Computer Science and learn more.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*AFMDdC5xgDrz16Tq.jpeg" /></figure><p><strong>What do you like most about being a developer?</strong></p><p>It’s challenging, creative and you get to work with amazing people.</p><p><strong>What would you advise aspiring developers?</strong></p><p>Don’t be afraid not to know things and ask questions, even if they might seem stupid. Also don’t let anyone make you feel like you don’t belong, because you absolutely do 🙂</p><p><strong>Do you have any pet projects? If yes, which one?</strong></p><p>No, I don’t seem to find the time… too many other hobbies!</p><p><strong>What do you enjoy doing apart from programming?</strong></p><p>I spend a lot of time playing video games 🎮, knitting things 🧶 and climbing up walls 🧗‍♀ .️</p><p><em>Thank you Nadine for sharing your story with us, we will definitely start playing around with things! If you enjoyed this chat and want to join us, check out </em><a href="https://door2door.io/en/about-us/jobs/"><strong><em>our job postings</em></strong></a><em> and see you in the next episode!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a4648dafe2f0" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-a4648dafe2f0">Introducing our Female Software Developers</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introducing our Female Software Developers]]></title>
            <link>https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-4e92850b417d?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/4e92850b417d</guid>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[female-developer]]></category>
            <dc:creator><![CDATA[door2door]]></dc:creator>
            <pubDate>Mon, 30 Nov 2020 15:06:58 GMT</pubDate>
            <atom:updated>2020-11-30T15:06:57.873Z</atom:updated>
            <content:encoded><![CDATA[<p>Episode 1: Nicky</p><p><em>We would like to take a moment and shine the light on our female engineering colleagues, in a series of short interviews. We hope that this will inspire more women to join the tech world and make a difference! This week we are introducing #{name}</em></p><pre>name = &quot;Nicky&quot;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*YqkhkSWK8EXEQ25c.jpeg" /></figure><p>Nicky joined door2door in January 2020 as a System Administrator Working Student and so far it has been a wild ride. But even with all the challenges that have taken place this year, the team has been supporting her to graduate and become a software developer. Now she’s working with the Passenger team as a Junior Software Developer.</p><p><strong>When did you first start writing code?</strong></p><p>I wrote my first lines of code around the age of 11 in my school’s informatics class. We started with Turbo Pascal and the class quickly became my favourite — not just because of the adrenaline rush of a code that compiles, but also because we could stay programming after school and if we wrote enough compilable code, we were allowed to browse the internet! It was early noughties so we craved that sweet sweet online freedom.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fgiphy.com%2Fembed%2FzOvBKUUEERdNm%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fgiphy.com%2Fgifs%2Fcoding-zOvBKUUEERdNm&amp;image=https%3A%2F%2Fmedia0.giphy.com%2Fmedia%2Fv1.Y2lkPTc5MGI3NjExOXgwdHkxZWlqbTZmbXp1OHRuandmazRlemIzaGh1djFiNmkwMGsweiZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9Zw%2FzOvBKUUEERdNm%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="244" frameborder="0" scrolling="no"><a href="https://medium.com/media/d6a57a70a87599fcd4c6c2cccd289fb8/href">https://medium.com/media/d6a57a70a87599fcd4c6c2cccd289fb8/href</a></iframe><p><strong>What was the first language you started in?</strong></p><p>My first language (at least, the first that I remember) was Turbo Pascal. As I had no idea how it could be applied outside of the classroom, it took me a while to graduate to new languages that I could use — those were HTML and CSS.</p><blockquote><em>Ask questions, even if you feel extremely embarrassed!</em></blockquote><p><strong>Do you have a preference between frontend or backend development?</strong></p><p>Having just started professional development, I can’t say that I do. Both of them are fulfilling in different ways for me and both are challenging. But if I <em>had </em>to choose, I’d probably choose frontend development.</p><p><strong>What was your biggest struggle when starting out?</strong></p><p>Having worked in a professional setting before and having a strong understanding of the key processes as well as approaches, it is challenging to start from the beginning again. I keep on expecting to perform as if I have 3 years of experience (which I do, but not in development). As much as it frustrates me sometimes, it’s also a great way of learning to be kind to yourself — because there’s no other way to continue!</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fgiphy.com%2Fembed%2FU5Dhg98ryYysg9cqMz%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fgiphy.com%2Fgifs%2Fwoofwaggers-struggle-struggles-dog-U5Dhg98ryYysg9cqMz&amp;image=https%3A%2F%2Fmedia0.giphy.com%2Fmedia%2FU5Dhg98ryYysg9cqMz%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="242" frameborder="0" scrolling="no"><a href="https://medium.com/media/36ad1068f5fe87d1786333d8b76ff415/href">https://medium.com/media/36ad1068f5fe87d1786333d8b76ff415/href</a></iframe><p><strong>Do you have any learning tips?</strong></p><p>If you are just starting, don’t expect to grasp everything in a week. Ask questions, even if you feel extremely embarrassed! It’s always better to ask the same thing twice then look at the code for three hours on the verge of a breakdown. Trust me, I’ve been there.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fgiphy.com%2Fembed%2F12QASni5AUN1SM%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fgiphy.com%2Fgifs%2Fthings-successful-interactions-12QASni5AUN1SM&amp;image=https%3A%2F%2Fmedia0.giphy.com%2Fmedia%2F12QASni5AUN1SM%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="300" frameborder="0" scrolling="no"><a href="https://medium.com/media/ceda9597d9ad61b11c32c0baa042e2e7/href">https://medium.com/media/ceda9597d9ad61b11c32c0baa042e2e7/href</a></iframe><p><strong>What made you want to become a developer?</strong></p><p>I’ve gravitated towards it since I was a child, so when the time came for me to consciously choose what I wanted to do professionally, I thought “Well, I was kinda good at programming… What could go wrong?” How young and naive I was!</p><p><strong>What do you like most about being a developer?</strong></p><p>It’s a constant challenge. For me, it is important to shake things up all the time — let yourself grow in maybe the most unexpected ventures. With development, the fact that you can’t know everything really excites me.</p><p><strong>What would you advise aspiring developers?</strong></p><p>Be nice to people. Work on what excites you and not what you think can sell. Take breaks! Be kind to yourself.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fgiphy.com%2Fembed%2F3ofSAZQdz2CQZgRpwA%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fgiphy.com%2Fgifs%2Fkpop-k-pop-heart-3ofSAZQdz2CQZgRpwA&amp;image=https%3A%2F%2Fmedia1.giphy.com%2Fmedia%2F3ofSAZQdz2CQZgRpwA%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="244" frameborder="0" scrolling="no"><a href="https://medium.com/media/eb04a156267f5ede33c53d5019228808/href">https://medium.com/media/eb04a156267f5ede33c53d5019228808/href</a></iframe><p><strong>Do you have any pet projects? If yes, which one?</strong></p><p>I just finished uni, so I’m taking a break from weekend coding ;)</p><p><strong>What do you enjoy doing apart from programming?</strong></p><p>I write my own music and record song covers. Sometimes I post them on YouTube.</p><p><em>Thank you Nicky for this inspiring interview, we will remember to be patient :wink ! If you enjoyed this chat and want to join us, check out </em><a href="https://door2door.io/en/about-us/jobs/"><strong><em>our job postings</em></strong></a><em> and see you in the next episode!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4e92850b417d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/introducing-our-female-software-developers-4e92850b417d">Introducing our Female Software Developers</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Preparation to migrate a domain from a monolith into its dedicated service]]></title>
            <link>https://medium.com/tech-blog-door2door/preparation-to-migrate-a-domain-from-a-monolith-into-its-dedicated-service-328782418c7b?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/328782418c7b</guid>
            <category><![CDATA[migration]]></category>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[monolith]]></category>
            <category><![CDATA[software-architecture]]></category>
            <category><![CDATA[domain-driven-design]]></category>
            <dc:creator><![CDATA[Yone Lacort]]></dc:creator>
            <pubDate>Thu, 18 Jun 2020 08:34:39 GMT</pubDate>
            <atom:updated>2020-06-20T11:25:43.704Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BbPCLtTAHIUxpcfTQsSAXw.jpeg" /><figcaption>Birds migrating — Photo by <a href="https://unsplash.com/@rayhennessy?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Ray Hennessy</a> on <a href="https://unsplash.com/s/photos/migration?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplay.ht%2Fembed%2F%3Farticle_url%3Dhttps%3A%2F%2Fmedium.com%2F_p%2Fpreparation-to-migrate-a-domain-from-a-monolith-into-its-dedicated-service-328782418c7b&amp;display_name=Play&amp;url=https%3A%2F%2Fplay.ht%2Farticles%2F328782418c7b&amp;image=https%3A%2F%2F1%2AQsrTLX8BLdPKwBVSRSBw5A.png&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=play" width="700" height="185" frameborder="0" scrolling="no"><a href="https://medium.com/media/08802e3fac9d68ec1794ea93f7a3666b/href">https://medium.com/media/08802e3fac9d68ec1794ea93f7a3666b/href</a></iframe><p>When starting to build a new product as an MVP (minimum viable product), it’s hard to foresee how its development will evolve and what features will become relevant after releasing it to the public. Once the product is out for consumption, users or clients will help guide what functionality is important and how the product should evolve.</p><p>As a small group of developers initiating the project, while the product is in its validation stage, it is a completely valid approach to start simple and allocate all the logic under a single service to reduce costs, simplify operations and move at a faster pace to deliver a working product quickly, sometimes even hastily.</p><p>The fact that the product should be released in a short period of time, doesn’t imply developers should not care about establishing a good and extensible foundation for the application growth. Otherwise at a certain point in the future, they might have to pay a higher price back when the product needs to be extended with new features and functionality needs to be extracted into new services.</p><p>Over time, the more functionality is added to the service, the more it will grow in complexity, business logic and different kinds of data to manage. If that preliminary foundation is not solid, unless measures are taken at the right time, the service could end up convoluted, with coupling in the code base producing an entanglement between several domains.</p><p>Having more than one domain within one service could complicate the application management. As the number of developers that work on the product grows, new developer teams will be spawned, with more than one team working simultaneously on the service’s development.<br>Ideally, the system’s architectural design should reflect the organizational structure, as stated in <a href="https://en.wikipedia.org/wiki/Conway%27s_law">Conway’s law</a>. This will prevent unnecessary communication overhead between teams, reduce cross-domain dependencies in the code base and lessen overlapping in deployment processes and other factors, that can ultimately affect the development teams’ autonomy and velocity in delivering changes that improve the product.</p><p>Once this situation is given, the aforementioned issues could be alleviated by extracting certain functionality from the monolithic service in which all these domains live to its dedicated domain-specific service. This newly born service should have a single, clear responsibility and ideally one team should be responsible for its development and maintenance.</p><p>Until the new service’s birth and its production usage, there is an often tedious and complicated process that requires the preparation of a clear vision and strategy before its development can even start.</p><ul><li>The <strong>vision</strong> will help to think ahead about the evolution of the architecture and also reduce potential pitfalls.</li><li>On the other hand, the <strong>strategy</strong> will aid to minimize failures along the way, smoothing the execution process and avoiding the unexpected emergence of unplanned work. With that said, there will always be those unforeseen tasks that will need to be addressed throughout the journey.</li></ul><p>The following 10 steps should assist you in preparing the migration of a domain to its dedicated service.</p><h3>1. Define the components and domains contained within the monolith</h3><p>As a first step, ensure the monolith’s components are well understood by all developers who work on it. Even if developers have been working on a service for a long time, this doesn’t mean that everyone is able to zoom out to fully understand other parts than the ones they are used to working on. And even if someone did, the understanding might diverge between different people until there’s a conversation to reach consensus.</p><p>Therefore, as a good point of departure, identify and document the components that live in the monolithic service in order to obtain a shared understanding with other engineers and group them in domains.</p><ul><li><em>A </em><strong><em>component</em></strong><em> is a building block of the application, with a clear role and responsibility and generally, constituted by multiple classes.</em></li><li><em>A </em><strong><em>domain</em></strong><em> could be interpreted as a context in the real world in which the software solves a problem.</em></li></ul><p>A good way to start having an open discussion, in the path to find an agreement on the shared vision within the team, could be hosting an <a href="https://en.wikipedia.org/wiki/Event_storming">Event Storming</a> workshop to convey ideas and different points of view in an interactive manner. The domain definition should be a collaborative process, in which even non-technical people who are domain experts can participate.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*a5x18IgmVsK_KkZBLkF4lg.jpeg" /><figcaption>Event domain workshops should end covering the wall with post-its — Photo by <a href="https://unsplash.com/@joszczepanska?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Jo Szczepanska</a> on <a href="https://medium.com/s/photos/workshop?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>Another great tool, in addition to an interactive workshop, is a written proposal. Written proposals are a great way to collectively agree on solutions in an asynchronous manner, giving people more time to quietly think at their own pace without the social pressure to construct justifications for their reasonings. Furthermore, proposals can also serve as documentation of the outcome as well as for the discussion process.</p><h4>Component interdependencies</h4><p>At this stage, it’s also worth investing time on analysing the interdependencies that exist between components and domains. In case of a domain extraction, these dependencies will eventually be translated into either data duplication across services or communication between services.</p><h3>2. Identify the domain to be extracted</h3><p>Knowing all domains that comprise the monolith and having a clear overview about how they interconnect, it’s a good foundation for making a better-informed decision on which domain and its subjacent components will constitute the new service.</p><p>A domain doesn’t necessarily need to be reduced to a single component. Don’t fall into the trap of spawning a micro-service per component, just because it seems to be a tendency in popular companies in the industry. Instead, give it enough thought before concluding, with strong reasons, that there are benefits for going with this solution. Otherwise, one side effect is that the system’s architectural and operational complexity could increase in an undesirable manner for the team that has to maintain it.</p><p>In this decomposition process, it would be advisable to start with macro-services, referring to service-based architecture, where more than one component can be aggregated into a single domain for reasons such as logic sharing, semantic cohesion, component entanglement, etc.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fWPTwnLBSgrKVb3uqAvdvw.jpeg" /><figcaption>Extracting a component carefully from the monolith — Photo by <a href="https://unsplash.com/@mparzuchowski?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Michał Parzuchowski</a> on <a href="https://medium.com/s/photos/jenga?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>After the service is created, it will be clearer whether there is an actual need to further decompose it into subdomains or whether the architectural state is good enough to remain as it is.</p><p><strong>Architecting for change</strong></p><p>If there’s a need to reduce complexity in a monolithic service by extracting a domain, an important aspect to account for is the likelihood that the domain’s code will be changed.</p><p>Ask yourself: <em>“Will the code that belongs to the domain be subjected to frequent changes or will it remain untouched, simply requiring some maintenance from time to time?”</em></p><p>Once the service is running, features and improvements relating to that domain should be delivered to end users at a higher velocity, reducing the time to market. This is often a good reason for working in such a migration, as it promotes strategic advantages for the business.</p><h3>3. Outline the business value</h3><p>Re-architecting services and extracting a domain from a monolith into its own service is a tedious and complex task. This work might require weeks and sometimes even months, and therefore, this labour should be backed by justified business value that the company will benefit from.</p><blockquote><em>Start focusing on the goals, then on the process.</em></blockquote><p>The decomposition of a monolith into several services may aid to improve certain system quality attributes such as maintenance, velocity, autonomy and many others. But are those aspects relevant for the business at the moment in time when the work is being proposed?</p><p>There may be other, more appropriate topics to tackle that could bring either more economic value or strategic advantage over business competitors at a given time. Besides, system quality attributes are technical terms that might seem abstract and not easily understood by non-technical stakeholders.</p><p>Consequently, it’s worth investing effort in finding the right words to describe the impact that the labour will have on the business. If such a statement is backed with numbers, even better.</p><p>As an example, imagine a hypothetical case in which a company has the need to double its user base and currently is at its capacity for dealing with its existing user base load. What would stakeholders think if the following statement were presented?</p><p><em>It is estimated that this migration will take around 2 months. It will help us to support twice the number of users we do now, as well as accept 10,000 orders more per minute than the 15,000 we are currently supporting. This could potentially translate into $400,000 in increased revenue.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*U0GRwuOlCtEXOdzqHTfuoQ.jpeg" /><figcaption>Present the migration to stakeholders as a great idea given its business value — Photo by <a href="https://unsplash.com/@pavement_special?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Riccardo Annandale</a> on <a href="https://medium.com/s/photos/business?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>A sentence of this kind will help depict the benefits to everyone. However, once non-technical stakeholders see the impact on the business, it will help convince them about the importance of this task, making it easier to buy them into the idea of working with the migration.</p><p>It’s definitely not easy to come up with such a statement and the numbers for this kind of work. Nonetheless, the takeaway is to simply think about translating them into a language that everyone understands and that clearly illustrates the business value.</p><h3>4. Envision the service evolution</h3><p>Thinking about the service evolution could be a speculative process, since it’s not possible to anticipate future business requirements with a high degree of certainty, although it will help establish a more solid foundation for the new service, along with measures to prevent eventual pitfalls.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wJQLLw_YE0zvqptvS6mvGw.jpeg" /><figcaption>Image by <a href="https://pixabay.com/users/madartzgraphics-3575871/?utm_source=link-attribution&amp;utm_medium=referral&amp;utm_campaign=image&amp;utm_content=1903318">Darwin Laganzon</a> on <a href="https://pixabay.com/?utm_source=link-attribution&amp;utm_medium=referral&amp;utm_campaign=image&amp;utm_content=1903318">Pixabay</a></figcaption></figure><p>Ask yourself questions such as:</p><ul><li><em>Will the service need to scale to thousands of requests per second at some point?</em></li><li><em>Will the service need to be extensible to support new features or connectivity with other services?</em></li></ul><p>The purpose of these questions is to identify risks that could affect the architecture’s evolution. It is recommended to document and analyse potential problems with the purpose of finding mitigation strategies while preventing dead ends in the future when problems could arise.</p><h3>5. Assess trade-offs</h3><p>It’s impossible to come up with an architectural solution that complies in the best manner with all <a href="https://en.wikipedia.org/wiki/List_of_system_quality_attributes">non-functional requirements</a> (a.k.a. system quality attributes), such as scalability, cost-efficiency, extensibility, etc. within a finite amount of time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*V9oP-wKCYrfkE9V3ue7uWg.jpeg" /><figcaption>Make a good analysis and discuss with your team and stakeholders - <a href="http://www.freepik.com/">Designed by slidesgo / Freepik</a></figcaption></figure><p>Instead of trying to comply with every single system quality attribute, be pragmatic and understand what are the most important ones. Then, propose a solution that complies with them as much as is feasible. It’s not necessary to invest time in making an API respond under 50ms for all its endpoints, if performance is not such a critical factor.</p><blockquote>Software architectures are never perfect — they were built by assessing trade-offs that might have been valid compromises in the past and whose aim was to promote the focus on what was important for the business at the time when those decisions were made.</blockquote><h3>6. Specify technologies for the new service</h3><p>When creating a new service, it’s a good moment to re-consider technologies. New technologies should be introduced if they bring a tangible benefit and not just because they are trendy and well known companies are using them.</p><p>Introducing a new technology that most developers are unfamiliar with will require a period of learning, practicing and mastering, slowing down the development’s velocity at the beginning. Furthermore, the more technologies are introduced in a company, the harder it gets for existing staff and new hires to maintain them.</p><p>Make sure the team that will build and maintain the new service will feel comfortable with the technical decisions made in advance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4iQxbNzrJQBlciTQ2sPrsg.png" /><figcaption>Endless number of technologies to choose from nowadays — <a href="https://landscape.cncf.io/">CNCF landscape</a></figcaption></figure><h4>Programming language</h4><p>Engineers might decide it is time to try that new trendy language without having thought about the implications that this decision will entail. There’s nothing bad about being innovative and wanting to introduce a new and more modern language, although there are trade-offs to be aware of.</p><p>In the scenario in which big portions of code from the monolith can be reused and extracted to the new service, it might be advisable to keep using the same programming language in order to save a substantial amount of effort.</p><p>Replacing the programming language will generally imply rewriting the code, unless a decision that allows a gradual transition is taken, such as transitioning from Java to Kotlin or from C to C++.</p><p>On the other hand, there might be strong reasons for ditching the current language and applying a different one at the expense of a potential code rewrite.</p><p>If developers that will work on the project are unfamiliar with the new language, this will most likely decelerate its delivery. Under these circumstances, it will be convenient to have at least one expert who can instruct others along the way and ensure the quality of the software delivered.</p><p>Carefully evaluate if the delay of rewriting everything into another language is going to bring benefits worth considering, versus the time it will delay the delivery of the project.</p><h4>Frameworks and libraries</h4><p>If you decide to stick to the same programming language, it might be worth re-evaluating other frameworks and libraries to obtain certain benefits that were lacking in the ones being used in the monolithic service.</p><p>Considering switching to a new programming language, if the company relies heavily on open-source code, ensure that the language is mature enough and has a strong community that builds good libraries and frameworks around it.</p><h4>Patterns</h4><p>This is a good moment to evaluate new patterns, since they influence the internal structure and design of a service. New patterns could bring advantages, such as making the project more maintainable and understandable, the code more extensible and reusable and other facets which will improve the overall development process.</p><p>If you decide to use the same programming language, the introduction of a new pattern should allow the reusability of code that was planned to migrate to the new service, although it might need slight modifications before it can be plugged into the new pattern interface.</p><h4>Underlying hosting infrastructure</h4><p>Changing the way a new service will be hosted, compared to how the monolithic service is, could bring certain advantages. Considering a few different hosting options, such as:</p><ul><li><a href="https://en.wikipedia.org/wiki/Platform_as_a_service">PaaS (Platform as a Service)</a> to reduce operational work and focus on application development to evolve the product at a faster pace.</li><li><a href="https://en.wikipedia.org/wiki/Function_as_a_service">FaaS (Function as a Service)</a> to have a cost-effective, scalable solution out-of-the-box and hassle-free server management.</li><li>Self-managing infrastructure components, if there’s a need to have more control over them for particular business reasons.</li><li>Migrating to a container orchestration solution, such as Kubernetes, in which all services can benefit from <a href="https://landscape.cncf.io/">powerful DevOps tooling</a> and be managed homogeneously.</li><li>Moving from on-premises to a cloud hosting provider with the objective to leverage on-demand computing and don’t have the need to pay up-front for hardware resources.</li></ul><p>Ultimately, this change might result in additional work that should neither be underestimated nor forgotten. This effort could include the following kinds of tasks:</p><ul><li>infrastructure resources monitoring and visualization,</li><li>collection of application metrics,</li><li>distinct alerting mechanism setup,</li><li>different service management procedures,</li><li>ability to improve deployment processes,</li><li>software patching, or</li><li>introduction of a new technology into the company.</li></ul><h4>Database</h4><p>A domain is broadly constituted of entities and their associated data, but in certain cases it could be purely functional without requiring data storage. If the former applies, data will eventually need to be migrated to its dedicated domain service.</p><p>It might be favourable to reuse the same database technology, unless there is a solid argument not to. Using the same technology will reduce entropy in the data migration process, in which data will be transferred from the monolith database into the service-dedicated database.</p><p>Ideally the data migration process should be conducted without incurring downtime and with the ability to roll back use from the new to the old database in case of certain malfunctions. There are tools and cloud services that help deal with the migration of existing data and the synchronization of new transactions without causing downtime.</p><h3>7. Forecast costs</h3><p>Systems should always be built in the most cost-efficient possible manner and consequently, cost-awareness should be an important constituent in the decision process. Even if the budget is unlimited or a process isn’t in place for approval of the cost allocation by the finance department, costs that the system re-architecture will entail should be forecast and made transparent to stakeholders.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MXvPtxDTbU_vcDjgTPqDCg.jpeg" /><figcaption>Burning money is what you shouldn’t be doing — Photo by <a href="https://unsplash.com/@jpvalery?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Jp Valery</a> on <a href="https://medium.com/s/photos/money?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>Increasing the amount of services generally increases the overall costs. Nevertheless, nowadays more modern hosting and cost-effective measures exist, such as serverless architectures, which may ultimately lead to lower overall costs.</p><h3>8. Define the strategy</h3><p>Describing the steps that can be foreseen in chronological order will certainly improve the execution process and reduce surprises along the way. In order to define the steps concisely and with an acceptable level of detail, an allocation of resources will be necessary to prepare the execution plan. This effort includes analysing the code, drawing diagrams, writing proposals, discussing different approaches and many other activities that should be conducted collaboratively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-w6CEgdvvVclAc6N2oTdjQ.jpeg" /><figcaption>Prepare a clear strategy to win — Photo by <a href="https://unsplash.com/@felix_mittermeier?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Felix Mittermeier</a> on <a href="https://medium.com/s/photos/strategy?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>Some steps that are generally applicable to most cases:</p><ul><li>The domain’s extractable code might be entangled with other domains. As groundwork, ensure that the domain’s code is decoupled before starting the migration.</li><li>Evaluate how reusable the code is to reach a decision on either migrating it, migrating parts of it or a full code rewrite.</li><li>Document in detail the new service’s API before starting its implementation.</li><li>Define the order in which the logic can be migrated if there’s a possibility of tackling this gradually.</li><li>Investigate the complexity of toggling between functionality in the new service and in the monolith, aiming for a gradual rollout plan rather than an <em>“all at once”</em> approach.</li><li>Prepare how the functionality of the new service can be tested independently and in conjunction with the monolith.</li><li>Ideate a database migration strategy.</li><li>Define how the new service will be managed, accounting for monitoring, alerting, deployment strategies and all other related DevOps processes.</li></ul><p>There are definitely more steps or tasks to consider, and every migration will demand its own particular ones. Be aware that even after having defined all the steps in as much detail as possible at an early stage, unforeseen issues and additional labour will surely arise as the migration goes on.</p><h3>9. Estimate the effort and plan the work</h3><p>It’s not easy to estimate the effort that will take a migration from a monolithic service to a new dedicated service, considering the amount of steps and unsought obstacles that may be encountered down the road. Having prepared a thorough strategy in advance will undoubtedly serve as a good foundation for estimating the effort. The more precise the strategy is, the better the estimations will be.</p><p>As a general rule of thumb, when estimating effort, evaluate with a pessimistic rather than an optimistic inclination. We all know that tasks will usually take longer than one had anticipated.</p><p>Every team operates with its preferred estimation units to measure effort or complexity, using fibonacci sequences, T-shirt sizes, story points, etc. However, for the purpose of high-level task estimation, I personally find developer days a better measurement, given it is a widely understood unit.</p><p>After the task estimation is completed, the next step is planning, where tasks should be scheduled in the chronological order defined in the strategy. Important factors to ponder in this phase will be the parallelization of those tasks as well as the availability of those developers who will be allocated to the project.</p><p>It is important to account for unforeseen work by adding a buffer to the different migration phases that is proportional to their complexity.</p><p>A <a href="https://en.wikipedia.org/wiki/Gantt_chart">Gantt chart</a> will be a good ally in visualizing the planned work, seeing what tasks can be parallelized and the amount of people that will work on it. The visualization will contribute to an understanding on how long a presumably homeric migration will be expected to take and when certain benefits can start having impact.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*I8UCO6pmtZ3AjMCDlJZM3w.jpeg" /><figcaption>Gantt chart visualization — <a href="http://www.freepik.com/">Designed by Freepik</a></figcaption></figure><h3>10. Document decisions</h3><p>Architecture decisions are trade-offs for system quality attributes that aim for the best compromise solution that satisfies service needs after taking their business importance into account. These decisions should be documented and chronologically ordered to assist with the understanding of how the architecture of a system has evolved over time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LwRic5M_dmmL-JI7G3KfsA.jpeg" /><figcaption>Photo by <a href="https://unsplash.com/@sctgrhm?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Scott Graham</a> on <a href="https://medium.com/s/photos/document?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>Decision records should be created for architecturally significant resolutions, such as those that affect the structure, non-functional characteristics, dependencies and service interfaces.</p><p>There are <a href="https://github.com/joelparkerhenderson/architecture_decision_record">different templates which can be chosen to create decision records</a> depending on how detailed the decision needs to be portrayed.<br>Regardless of the template selected and the level of detail to be logged, these records should encapsulate the following core concepts:</p><ul><li>the context in which the decision was made,</li><li>the problem solved,</li><li>the solution taken,</li><li>and lastly, the consequences and downsides of that solution.</li></ul><h3>Conclusion</h3><p>Before starting the extraction of a domain from a monolith to its dedicated service, it’s worth investing time in the preparatory work and being thorough enough. This will enable more informed decisions, thereby preventing eventual pitfalls that could turn out expensive for the company.</p><p>Stakeholders should understand and be aware of the business value this labour will bring, as well as the estimated amount of time and effort it will take.</p><p>All the steps should be conducted as a collaborative process in which decisions are discussed and agreed upon with relevant teammates who can contribute to a better outcome.</p><blockquote>Good luck is the result of good planning, hence you should plan thoroughly ahead of time to ensure a smoother and more trouble-free migration process.</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=328782418c7b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/preparation-to-migrate-a-domain-from-a-monolith-into-its-dedicated-service-328782418c7b">Preparation to migrate a domain from a monolith into its dedicated service</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Recording Network Requests for iOS UI Tests with Succulent]]></title>
            <link>https://medium.com/tech-blog-door2door/recording-network-requests-for-ios-ui-tests-with-succulent-e17809c4197c?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/e17809c4197c</guid>
            <category><![CDATA[ios]]></category>
            <category><![CDATA[succulents]]></category>
            <category><![CDATA[ui-testing]]></category>
            <category><![CDATA[swift]]></category>
            <dc:creator><![CDATA[Dylan Maryk]]></dc:creator>
            <pubDate>Wed, 15 Apr 2020 09:08:33 GMT</pubDate>
            <atom:updated>2020-04-15T09:08:33.549Z</atom:updated>
            <content:encoded><![CDATA[<p>Over the last few months at door2door on the iOS team, we have begun to write UI tests to cover the core flows of our app in an attempt to catch any serious issues before we do a release. For the most part we have been utilising Apple’s own tools for this, but there is one third-party library we have been using in addition to these: <a href="https://github.com/cactuslab/Succulent">Succulent</a>.</p><p>Succulent is a simple but effective Swift library for recording and replaying network requests, which means that after the first time your app performs a request, subsequent requests to the same endpoint will return the same response, enabling faster and more stable testing. There’s no need to take a cumbersome approach of manually mocking responses yourself. So, let’s dive into a tutorial covering how you can get going with Succulent in your project!</p><p>Start by installing Succulent using Cocoapods. Add this to your Podfile: pod “Succulent”. Then just go ahead and run pod install, and done!</p><p>Next you need to do some preparation before you can start making use of the library. The path at which you want your recordings to be stored has to be specified in the Info.plist of your <strong>UI testing directory</strong>. For example:</p><pre>&lt;key&gt;TraceRecordPath&lt;/key&gt;<br>&lt;string&gt;$(PROJECT_DIR)/{YOUR-UI-TESTS-DIRECTORY}/Succulent/&lt;/string&gt;</pre><p>You also need to give the target you are testing permission to connect to a local server, as it will need to connect to the local server Succulent runs on your machine to record requests. This is done by adding the following to the Info.plist of the <strong>target you are testing against</strong>:</p><pre>&lt;key&gt;NSAppTransportSecurity&lt;/key&gt;<br>&lt;dict&gt;<br> &lt;key&gt;NSAllowsLocalNetworking&lt;/key&gt;<br> &lt;true/&gt;<br>&lt;/dict&gt;</pre><p>Lastly for the preparation, the base URL to which your app makes requests has to be replaced with the URL to Succulent’s local server. You can do this by using the Succulent base URL if an environment variable for it exists, which we will come to adding a little later. Otherwise, default to using your real base URL.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/80ba618101ce9991e618b831c664da54/href">https://medium.com/media/80ba618101ce9991e618b831c664da54/href</a></iframe><p>Now we can start to use the library itself! At this point we can mostly rely on the official documentation, so this portion of code is mostly based on the example provided in Succulent’s README. It should be placed inside your XCTestCase. If you have multiple test cases, you could also place it in a super class that each test case inherits from.</p><p>One very important detail to note is the / at the end of the real base URL, as without this, <strong>requests will fail to be recorded</strong>.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/069ad346d6fc8446a885e50dd89c38b3/href">https://medium.com/media/069ad346d6fc8446a885e50dd89c38b3/href</a></iframe><p>Now run your UI tests, and afterwards you should notice a .trace file at the TraceRecordPath you specified earlier for each endpoint that was called. You can optionally add this directory of traces to your project, if you want to be able to access them within Xcode. If you ever want to re-record a request to an endpoint, just delete its corresponding trace file and run your UI tests again.</p><p>If you are using a version control system, make sure all your trace files are committed to ensure alignment with your team and/or continuous integration setup.</p><p>That’s it, we’re done! Enjoy your easier, faster and more reliable UI testing process!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e17809c4197c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/recording-network-requests-for-ios-ui-tests-with-succulent-e17809c4197c">Recording Network Requests for iOS UI Tests with Succulent</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Predicting Ridepooling Searches using POIs]]></title>
            <link>https://medium.com/tech-blog-door2door/predicting-ridepooling-searches-using-pois-94e6b6cddc2f?source=rss----b94e3e744719---4</link>
            <guid isPermaLink="false">https://medium.com/p/94e6b6cddc2f</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[transportation]]></category>
            <category><![CDATA[gis]]></category>
            <category><![CDATA[regression]]></category>
            <dc:creator><![CDATA[Mariam M.]]></dc:creator>
            <pubDate>Thu, 14 Mar 2019 12:56:39 GMT</pubDate>
            <atom:updated>2019-03-14T14:35:04.444Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZnzHXxc7OOyd2PLVEsKceg.png" /><figcaption>Existing Ridepooling Searches in the operating area in Berlin, grouped by roads they’re routed through</figcaption></figure><p>A very common question I think a lot of B2B companies get nowadays (regardless of industry) by clients that want to start doing business with them is: <strong>how do I know this will succeed?</strong></p><p>In the <a href="https://www.door2door.io/insights.html">Mobility Intelligence team</a> at door2door, we try to answer that question for our potential Ridepooling partners. We do so by analysing and visualising a number of indicators (like <a href="https://blog.door2door.io/what-is-public-transport-coverage-ptc-d4b7635875cc">Public Transport Coverage</a> for example), and simulating possible Ridepooling scenarios based on existing demand data to get <a href="https://en.wikipedia.org/wiki/Performance_indicator">KPI</a>s that help clients design their service.</p><p><strong>But what if we don’t have demand data for a given region? What then?</strong></p><p>We could try and simulate with random demand (and we do), but how much does that tell us, really? We needed something more solid. Something <strong>better than random</strong> to give us more confidence in our simulations.</p><h3>Existing Demand Data</h3><p>What we define as “demand data” is actually movement data. Where people want to move from/to and when. One of our demand data sources is the Searches from our existing Ridepooling services. It’s nothing too complex. Every Search consists of:</p><ul><li><strong>location_from</strong>: where the person wants to depart from.</li><li><strong>location_to: </strong>where the person wants to go to.</li><li><strong>timestamp:</strong> when the search happened.</li></ul><p>For the purpose of Insights, we also add a layer of aggregation to this raw data, to be able to analyse and therefore draw conclusions from it more easily. We divide every region we have in Insights into hexagon cells, and all the hours of the week into 1-hour bins. So the data we end up getting from the raw data is:</p><ul><li><strong>cell_from</strong>: which cell the trip originated from.</li><li><strong>cell_to</strong>: which cell the trip ends up going to.</li><li><strong>time_id</strong>: the 1-hour bin in the week when the trip happened.</li><li><strong>weight</strong>: how many trips, on average, happen from/to the same cells in the same 1-hour bin.</li></ul><h3>What we want to predict</h3><p>The idea is to predict how people want to move within a region, in general. But more specifically, we want to predict how they move between Insights cells (and at which hours/weekdays). Therefore, we want to predict, in a completely new region where we don’t have any Search data yet:</p><p>For every combination of <strong>cell_from, cell_to</strong>, and <strong>time_id: </strong>how many trips will happen on average (weight)?</p><h3>How will we make those predictions?</h3><p>The process to get to a prediction consists of three main steps:</p><ol><li>Define what a cell <em>has</em> or <em>is</em> that contributes to the decision of taking a trip to or from it (<strong>“Features”</strong>). This includes how many points of interest it has, how many people live there, etc.</li><li>Teach the algorithm we’re writing to translate those features to a number of searches (weight) at a given time (<strong>“Training”</strong>).</li><li>Based on the previous two steps, predict what the searches will look like in a new region.</li></ol><h4>Features</h4><p>What can a given cell have or be to contribute to the decision of taking a trip to or from it? We boiled that down, for now, to certain POIs (points of interest) and population density. The set of features we use right now are:</p><ul><li><strong>Population density</strong>: number of inhabitants per square km (not taking into account demographic data like age distribution, etc).</li><li><strong>Sustenance</strong>: number of restaurants/bars/etc that exist in a given cell.</li><li><strong>Residential buildings</strong>: number of buildings that are marked as “residential” in a given cell.</li><li><strong>Shops:</strong> number of shops in a given cell.</li><li><strong>Offices:</strong> number of offices in a given cell.</li><li><strong>Entertainment:</strong> number of entertainment POIs (cinemas/theatres/etc) in a given cell.</li></ul><p>We get all of this data (except population data) from <a href="http://openstreetmap.org/">OpenStreetMap</a>.</p><p>We take those features into account for both the <strong>cell_to</strong> and the <strong>cell_from</strong>.</p><h4>Training</h4><p>Now that we can define what a connection means in terms of features, we have to teach the algorithm how to translate those features into a number of searches. We do this by:</p><ul><li>Preparing a training data set based on Searches from our Ridepooling operations in Berlin.</li><li>Feeding this data set to a <strong>model</strong>, which will then map the features of the cells in Berlin to the number of Ridepooling Searches we have there and come up with a way to translate features into searches.</li></ul><h4><strong>Preparing the (training) data set</strong></h4><p>We add two new features to every connection:</p><ul><li>First, a feature called <strong>target_value</strong> (= weight of Ridepooling Searches in that connection) based on real data, and also</li><li>A <strong>time_category</strong>: this indicates in which interval of time did those Ridepooling Searches happen.</li></ul><p><strong>Wait. “Time Category”?</strong></p><p>For the purpose of this prediction, we thought a 1-hour bin might be too granular. We wanted broader bins to hold that info, which is where <strong>time_category </strong>comes in. We divided the day into bigger bins (like “morning”, “afternoon”, “evening”, and “night”). Therefore, we will also predict in those bigger bins instead, and assume that within every time_category, trips are distributed evenly between individual hours.</p><h4>Scaling the data</h4><p>To be able to compare features in different regions in a more accurate way, we <em>scale </em>the data so it’s relative to its region, and <strong>instead of representing an absolute number for cell features</strong>, we transform that to represent<strong> how many features this cell has <em>relative to other cells in the same region.</em></strong> For example:</p><ul><li>Berlin has<strong> 3.7 million people</strong> living in it. Therefore if a <strong>cell A</strong> has <strong>3.7 million</strong> people, it has 100% of the population<strong> (=1 on the scale from 0 to 1).</strong> If it has<strong> half</strong>, then it’s a <strong>0.5 on a scale from 0 to 1</strong>, and so on.</li><li>Duisburg has <strong>500K people</strong> living in it. Therefore if a <strong>cell B</strong> has <strong>500K people, it has 100% of the population (=1 on the sale from 0 to 1).</strong></li><li><strong>With scaling</strong>, cells A (Berlin) and B (Duisburg) are <strong>equal.</strong> They both represent a very high density (100% density) of the people in that area.</li></ul><p>This also applies to other features, like say, shops. 20 shops in Berlin might be comparable to only 5 in Duisburg because of how the shops are distributed in a region. This gives us the power to analyse/predict based on <em>relative features</em> because every region is different.</p><p><strong>Note:</strong> when we first predict a value for the searches, we get a value from 0 to 1, we then <strong>un-scale </strong>this value by multiplying it by the maximum number of Searches we could have in a connection to get a real/absolute value again.</p><h3>Finding the right model</h3><p>Every type of model has its own algorithm (= way of translating features into values). There are many models out there that could be potentially used to solve this type of problem: predicting a <strong>numeric target value</strong> based on a set of features.</p><p>Since the target value is numeric, and can be any number from 0 to infinity, we use a general technique called <a href="https://en.wikipedia.org/wiki/Regression_analysis"><strong>Regression</strong></a>. The other general technique would be Classification, but this only applies to data sets where the target value is a closed set and not an infinite range. Under Regression, there are several models we could use: Linear, Polynomial, and Random Forest are just three examples.</p><p>We also tried to use <a href="https://en.wikipedia.org/wiki/Deep_learning">Deep Learning</a> using a <a href="https://en.wikipedia.org/wiki/Artificial_neural_network">Neural Network</a> with different configurations, but that did not yield results we are satisfied with.</p><h4>But how do we measure results, anyway?</h4><p>We used a combination of different metrics in order to compare different models to each other (and to give us a general sense of the performance of each one with our data). Those metrics are applied on <strong>testing regions</strong> (non-Berlin) where we predict searches <em>and </em>have real data to compare how accurate the predictions are:</p><ul><li><a href="https://en.wikipedia.org/wiki/Root-mean-square_deviation"><strong>Root Mean Square Error</strong></a><strong>:<br></strong>For every prediction we make, we compare <strong>predicted_searches </strong>to <strong>ridepooling_searches </strong>(actual searches). We calculate the squared difference (“Square Error”) = (predicted — actual)². We then get the mean of all the Square Errors, and get the square root of that mean to get an average error in the entire dataset. If we get, for example, an actual RMSE of 0.2 this means that on average, we predict searches that are +/- 0.2 from the actual searches.</li><li><strong>Value Distribution:<br></strong>We plot the actual and predicted values on a chart, where every point on the<strong> x-axis</strong> is a connection, and every value on the <strong>y-axis</strong> is the number of Searches. We then visually compare the distribution of values between the connections (predicted vs. actual).</li><li><strong>Spatial Distribution:<br></strong>We plot the actual and predicted values to/from a certain cell on a (heat) map and visually compare the distributions. This helps us see if we predict demand in places like city centers, for example, or rural areas correctly or not.</li></ul><h4>Results</h4><p>Based on all the results combined, we decided to go with <a href="https://en.wikipedia.org/wiki/Random_forest"><strong>Random Forest Regression</strong>.</a> We then generated more in-depth insights of the results of that model by predicting demand in two regions: Duisburg and Munich, for which we already have real data.</p><p>After training the chosen model, we analysed the <strong>importance</strong> of the POI features on the number of actual searches (Ridepooling Searches) in the training region (Berlin):</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Nu-t0QbyEDhp0RsigvQAkg.png" /><figcaption>Importance of every feature as determined by the model</figcaption></figure><p>Great. Now we have successfully predicted Searches. We can use this everywhere and make absolute decisions with it, right?</p><p>Err..</p><h3>Limitations</h3><p>Since we use a training data set, we are limited by its limitations. Since our training data set is the Ridepooling ops in Berlin, our limitations are:</p><ul><li><strong>The operating period:</strong> we will not be able to predict searches outside that operating period because we simply don’t know how people behave the rest of the day(s).</li><li><strong>The features in the operating area:</strong> if different combinations of features yield different results, we won’t know. Best we can do is compare the new region to Berlin’s operating area and try to predict based on that.</li><li><strong>The service itself:</strong> pricing, advertising, etc all affect how many people use a service to move. If different services have different models that might affect who uses their service, we are not able to detect that difference or take it into account.</li></ul><h3>Existing Research and Studies</h3><p>In the scientific community what we refer to as connections (trips from one cell to another) is usually called <strong>Origin-Destination (O-D) matrices</strong>. There exists already quite a a bit of research on how to predict O-D matrices using machine learning. We’ve reviewed some of this research to decide on how to solve our problem. Here’s a quick overview of some of what we reviewed:</p><ul><li><strong>ADEWALE 2017: </strong><a href="http://ds.cs.ut.ee/courses/course-files/Ayobami-project.pdf"><strong>Deep Learning Based Origin Destination Matrix Prediction Through GPS DATA</strong></a></li><li><strong>SAADI et al. 2017</strong>: <a href="https://www.sciencedirect.com/science/article/pii/S2352146517306087"><strong>A bi-level Random Forest based approach for estimating O-D matrices: Preliminary results from the Belgium National Household Travel Survey</strong></a></li><li><strong>ASHOK 1996: </strong><a href="https://its.mit.edu/sites/default/files/documents/AshokPhd_0.pdf"><strong>Estimation and Prediction of Time-Dependent Origin-Destination Flows</strong></a></li><li><strong>HELD et al. 2018: </strong><a href="https://doi.org/10.3929/ethz-b-000266653"><strong>Future mobility demand estimation based on sociodemographic information: A data-driven approach using machine learning algorithms</strong></a></li></ul><h3>Conclusion</h3><p>In a parallel universe, this would be a lot easier: just find someone who can see into the future, train them to gather data efficiently, and hire them to come back with accurate predictions. We’re stuck with Maths for now. But, while it does have its limitations, especially with a limited dataset, it certainly gives us a pretty good <em>insight</em> (ha!) into one of the many possible futures.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=94e6b6cddc2f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/tech-blog-door2door/predicting-ridepooling-searches-using-pois-94e6b6cddc2f">Predicting Ridepooling Searches using POIs</a> was originally published in <a href="https://medium.com/tech-blog-door2door">door2door Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>