<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by chris viau on Medium]]></title>
        <description><![CDATA[Stories by chris viau on Medium]]></description>
        <link>https://medium.com/@christopheviau?source=rss-a1206bba9736------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*Os02YyqJ5QMAEO6bA0l0-A.jpeg</url>
            <title>Stories by chris viau on Medium</title>
            <link>https://medium.com/@christopheviau?source=rss-a1206bba9736------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 02:23:35 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@christopheviau/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[VisMTL10: La recherche]]></title>
            <link>https://medium.com/@christopheviau/vismtl10-la-recherche-a018e4cb64ee?source=rss-a1206bba9736------2</link>
            <guid isPermaLink="false">https://medium.com/p/a018e4cb64ee</guid>
            <category><![CDATA[dataviz]]></category>
            <dc:creator><![CDATA[chris viau]]></dc:creator>
            <pubDate>Wed, 30 May 2018 19:06:38 GMT</pubDate>
            <atom:updated>2018-05-30T19:06:38.861Z</atom:updated>
            <content:encoded><![CDATA[<p>Le meetup <a href="https://www.meetup.com/vismtl/events/249642994/">VisMontreal #10</a> a profité du passage de conférenciers de réputation internationale qui étaient à Montréal pour présenter leur recherche à <a href="https://chi2018.acm.org/">CHI</a>, une des plus importantes conférences en datavis et Interaction Humain-Machine. Nous vous invitons à découvrir ces chercheurs qui <strong>repoussent les limites de la datavis</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/450/1*WTiu5LgSMIs_xCdheGH_6A.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uxgJSxfyEGmalLIEwYDijA.jpeg" /></figure><h3>Remplacer des nombres par des grandeurs connues</h3><p><strong>Jessica Hullman, </strong>University of Washington<br><a href="https://dl.acm.org/citation.cfm?id=3173608">Improving Comprehension of Measurements Using Concrete Re-Expression Strategies</a></p><p>De nombreux articles nous présentent des nombres qui sont difficiles à saisir, bien souvent parce que ces valeurs ne se rapporte à rien de concret à quoi se comparer. Par exemple, au lieu d’affirmer que «l’usine d’épuration de Montréal traite 68567.05 m³ d’eau par jour», on peut remplacer ce chiffre par une <strong>valeur de comparaison</strong> qui permet de mieux se la représenter, par exemple «l’usine d’épuration de Montréal traite l’équivalent de 18 piscines olympiques d’eau par jour». Jessica Hullman et son équipe ont conçu un plugin qui remplace ainsi automatiquement les nombres par des grandeurs de comparaison.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fvideoseries%3Flist%3DPLH3opJw-q4ksBQ7Qp0vsB2zS39rJVnpis%26start%3D0&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D9H3fFpkuz64&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F9H3fFpkuz64%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/114258cf63fadebc5227e60ba7b071f2/href">https://medium.com/media/114258cf63fadebc5227e60ba7b071f2/href</a></iframe><h3>Comparer des diagrammes à barres</h3><p><strong>Arjun Srinivasan, </strong>Georgia Institute of Technology<br><a href="https://dl.acm.org/citation.cfm?id=3173878">What’s the Difference?: Evaluating Variations of Multi-Series Bar Charts for Visual Comparison Tasks</a></p><p>Il existe plusieurs types de <strong>diagrammes à barres</strong> (groupées, empilées, etc.). Arjun explore la façon d’en analyser l’efficacité par rapport à certaines tâches, comme la visualisation des différences entre deux barres, la compréhension des valeurs négatives et autres tâches analytiques</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fvideoseries%3Flist%3DPLH3opJw-q4ksBQ7Qp0vsB2zS39rJVnpis%26start%3D0&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DsMlfrng-Mho&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FsMlfrng-Mho%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/599bf76cb1995a4dc826a2ec6c9eef10/href">https://medium.com/media/599bf76cb1995a4dc826a2ec6c9eef10/href</a></iframe><h3>Dessiner des données</h3><p><strong>Fanny Chevalier</strong>, University of Toronto &amp; Inria<br><strong>Nathalie Henry Riche</strong>, Microsoft<br><a href="https://dl.acm.org/citation.cfm?id=3173797">DataInk: Direct and Creative Data-Oriented Drawing</a></p><p>L’application que Fanny et Nathalie ont développée permet d’utiliser des données pour faire <strong>varier des propriétés graphiques</strong>, comme la couleur, la taille et la distribution dans l’espace, sur des dessins faits à la main.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fvideoseries%3Flist%3DPLH3opJw-q4ksBQ7Qp0vsB2zS39rJVnpis%26start%3D0&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dq_ylWhkqklw&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fq_ylWhkqklw%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/16a3efc2a882c619a2916da6b7ab03b7/href">https://medium.com/media/16a3efc2a882c619a2916da6b7ab03b7/href</a></iframe><h3>Opérer un rover</h3><p><strong>Matthew Conlen</strong>, University of Washington<br><a href="https://idl.cs.washington.edu/files/2018-Meridian-CHI.pdf">Towards Design Principles for Visual Analytics in Operations Contexts</a></p><p>Matthew explique les contraintes techniques d’un de ses projets fait pour optimiser <strong>la position de l’antenne d’un rover</strong> sur mars pour lui permettre d’envoyer des donner sur terre. En plus d’explorer d’intéressantes avenues en terme de datavis et d’expérience usager, l’ensemble des contraintes et des solutions envisagées forment un bel exemple de démarche de datavis pour répondre à un besoin industriel précis.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fvideoseries%3Flist%3DPLH3opJw-q4ksBQ7Qp0vsB2zS39rJVnpis%26start%3D0&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DkGXZtM8BYIU&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FkGXZtM8BYIU%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/0c2dc59838b386f4fcdf0e638badfc25/href">https://medium.com/media/0c2dc59838b386f4fcdf0e638badfc25/href</a></iframe><h3>Data comics</h3><p><strong>Benjamin Bach</strong>, University of Edinburgh<br><a href="https://www.research.ed.ac.uk/portal/files/57948259/Bach2018designpatterns.pdf">Design Patterns for Data Comics</a></p><p>Benjamin nous fait découvrir le monde du Data Comics, des infographies utilisant les principes narratifs de la <strong>bande-dessinée</strong>.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fvideoseries%3Flist%3DPLH3opJw-q4ksBQ7Qp0vsB2zS39rJVnpis%26start%3D0&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DDNwKzElA0XQ&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FDNwKzElA0XQ%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/ccb2a24d6eb24e428a749d53dfa9ecff/href">https://medium.com/media/ccb2a24d6eb24e428a749d53dfa9ecff/href</a></iframe><h3>Illustration vectorielle à partir de données</h3><p><strong>Zhicheng Liu</strong>, Adobe Research<br><a href="http://www.data-illustrator.com/papers/DataIllustratorCHI18.pdf">Data Illustrator: Augmenting Vector Design Tools with Lazy Data Binding for Expressive Visualization Authoring</a></p><p>Zhicheng nous présente Data Illustrator, un outil de création de graphiques sans avoir besoin de programmer. Comme DataInk mentionné plus haut, cet outil me rappel le travail fondamental de Bret Victor <a href="https://www.youtube.com/watch?v=PUv66718DII">Inventing on Principle</a> où des variables visuelles sont associées a des données pour former des visualisations cohérentes et intuitives. Un des aspects les plus intéressants de <strong>Data Illustrator</strong> est qu’il va plus loin que le prototype de recherche, <a href="http://data-illustrator.com/">vous pouvez l’utiliser maintenant</a>.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fvideoseries%3Flist%3DPLH3opJw-q4ksBQ7Qp0vsB2zS39rJVnpis%26start%3D0&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D0ezweiYnsTI&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F0ezweiYnsTI%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/94a3b0c6f498c92625b20070bd2154f9/href">https://medium.com/media/94a3b0c6f498c92625b20070bd2154f9/href</a></iframe><h3>Visualisations physiques</h3><p><strong>Alice Thudt</strong>, University of Calgary<br><a href="https://research-repository.st-andrews.ac.uk/bitstream/handle/10023/13267/CPV_CameraReady.pdf?sequence=1">Self-Reflection and Personal Physicalization Construction</a></p><p>Alice nous fait part de son expérience avec un groupe diversifié qui a accepté de se prêter au jeu d’encoder d’une façon personnelle des données tirées de leur quotidien. Par exemple, une participante enfilait une bille de collier chaque fois qu’elle se trouvait déconcentrée dans son travail, l’espace entre les billes montrant les périodes de travail. La recherche nous a permis de découvrir de nombreuse datavis «physique» ainsi que l’<strong>histoire personnelle</strong> derrière chacune.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fvideoseries%3Flist%3DPLH3opJw-q4ksBQ7Qp0vsB2zS39rJVnpis%26start%3D0&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D9Cho8jvlDFY&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F9Cho8jvlDFY%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/ef6cedae0d2be92b8b68b223f125dddc/href">https://medium.com/media/ef6cedae0d2be92b8b68b223f125dddc/href</a></iframe><h3>Conclusion</h3><p>La recherche académique en datavis permet d’en explorer les limites, de développer des outils, de découvrir de nouveaux champs d’application et de réfléchir sur la pratique même. Nous espérons que ces chercheurs vous ont inspirés à <strong>explorer de nouvelles avenues</strong>, ou, pourquoi pas, à retourner aux études!</p><p>Merci à <strong>MapD</strong>, notre précieux commanditaire.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/477/1*jiqjFlEXkfrYim3QZ--_FA.png" /></figure><p>Ainsi qu’à toute l’équipe de VisMTL. <strong>Joignez-vous à nous</strong> et organiser d’autres <a href="https://www.meetup.com/vismtl/">meetups</a> comme celui-ci!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZyHEbpKJJfi3QOkDsfgnFw.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a018e4cb64ee" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VisMTL9: des nouvelles de l’industrie]]></title>
            <link>https://medium.com/@christopheviau/vismtl9-des-nouvelles-de-lindustrie-ef47608d3669?source=rss-a1206bba9736------2</link>
            <guid isPermaLink="false">https://medium.com/p/ef47608d3669</guid>
            <category><![CDATA[dataviz]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[meetup]]></category>
            <category><![CDATA[montreal]]></category>
            <dc:creator><![CDATA[chris viau]]></dc:creator>
            <pubDate>Mon, 13 Nov 2017 13:40:22 GMT</pubDate>
            <atom:updated>2017-11-15T13:24:25.612Z</atom:updated>
            <content:encoded><![CDATA[<p>Le meetup <a href="https://www.meetup.com/vismtl/events/244654682/">VisMontreal #9</a> portait sur <strong>l’écosystème de la datavis</strong>, rapportant les dernières nouvelles de l’industrie, les studios, les outils, les tendances, les conférences, etc.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/450/1*ndA6MdQRHn8ivfC9vc7opg.jpeg" /></figure><h3>Live from SF!</h3><p>La conférence a commencé avec une présentation en direct des bureaux de Google à San Francisco avec <strong>Ian Johnson</strong>, un des organisateur les plus importants de la communauté de la datavis. Il est le fondateur des meetup D3 dans la Silicon Valley, de la D3 Unconference et de nombreux outils collaboratifs. Il nous a entre autres présenté un outil de partage et une galerie comprenant plus de <a href="http://blockbuilder.org/search">26000 visualisations</a>, très utile pour s’inspirer et pour apprendre à programmer en D3.js. Voici ses <a href="https://docs.google.com/presentation/d/1Q3Tet-ktGU4szCDLIbepBfwWqrIXzfTARnmyUTkMhho/edit?usp=sharing">diapos de présentation</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*a4TMHBk61P7ZpA9rNn4cvA.jpeg" /><figcaption>Ian Johnson</figcaption></figure><h3>Des nouvelles de l’industrie</h3><p>Ensuite <strong>Christophe Viau</strong> donnait un aperçu de ce qu’il percevait, de sa position de développeur en datavis pour <a href="https://www.mapd.com/platform/immerse/">MapD</a> à San Francisco, comme étant les dernières tendances de l’industrie. La présentation était basée sur un déluge de liens. Les voici donc.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*o8-vZJj129BGbf9SWXEKnw.jpeg" /><figcaption>Christophe Viau</figcaption></figure><h3>Conférences</h3><p>Il y a des <strong>meetups</strong> comme VisMTL, <a href="https://www.meetup.com/Quebec-Numerique/">VisQUEBEC</a> et les Data Viz Jam Sessions un peut partout dans le monde. D’autres conférences <strong>académiques</strong> permettent de se tenir à la fine pointe. Mais on voit apparaitre de plus en plus de conférence plus près du <strong>design</strong> d’interaction, du «creative computing» et des arts génératifs. Une des plus intéressante, OpenVisConf, a un avenir incertain après la fermeture du département de datavis de la compagnie <a href="https://bocoup.com/services/datavis">Bocoup</a> qui l’organisait.</p><h4>Meetups</h4><ul><li><a href="https://www.meetup.com/vismtl/events/244654682/">VisMTL 9: Survol du domaine de la datavis</a></li><li><a href="https://www.meetup.com/Data-Viz-Jam-Sessions-MTL/">Data Viz Jam Sessions - MTL</a></li></ul><h4>Académique</h4><ul><li><a href="http://www.wikicfp.com/cfp/call?conference=visualization">Visualization Call For Papers for Conferences, Workshops and Journals at WikiCFP</a></li><li><a href="http://ieeevis.org/">IEEE VIS 2017</a></li><li><a href="http://eurovis.swansea.ac.uk/">Eurovis 2014 | The Eurographics Conference on Visualization | Home</a></li><li><a href="http://chi2014.acm.org/">Welcome | CHI 2014</a></li><li><a href="https://theinnovationenterprise.com/summits/data-visualization-summit-boston-2017">Data Visualization Summit, Boston</a></li></ul><h4>Design, infographique, creative computing</h4><ul><li><a href="http://www.malofiejgraphics.com/world_summit/">Malofiej World Summit - About - Malofiej · Infographic World Summit · Infographic awards</a></li><li><a href="http://eyeofestival.com/">Eyeo Festival | Converge to Inspire</a></li><li><a href="http://visfest.com/d3unconf-2017/">2017 - viSFest</a></li><li><a href="https://www.see-conference.org/en/?noredirect=en_US">see-Conference #11</a></li><li><a href="https://openvisconf.com/#videos">OpenVisConf 2017</a></li></ul><h3>Compagnies</h3><p>Les grands <strong>journaux</strong> états-uniens et anglais continue d’être la locomotive de l’industrie de la datavis. Mais de grandes innovations sortent aussi de petits <strong>studios</strong> comme, FFunction, une entreprise de Montréal. D’intéressantes discussions récentes entre plusieurs acteurs de l’industrie tentent de mieux définir la place de l’expert en datavis dans l’industrie.</p><h4>Studios</h4><ul><li><a href="https://ffctn.com/en/index">Data visualization company - FFunction</a></li><li><a href="https://medium.com/accurat-studio/the-architecture-of-a-data-visualization-470b807799b4">The Architecture of a Data Visualization</a></li><li><a href="http://www.periscopic.com/">Periscopic: Do Good With Data</a></li><li><a href="https://stamen.com/">Stamen DesignStamen Design: Data Visualization and Map Design Studio</a></li><li><a href="http://pitchinteractive.com/">Pitch Interactive - A Data Visualization Studio</a></li><li><a href="https://www.visualcinnamon.com/portfolio">Professional &amp; personal data visualization projects - Visual Cinnamon</a></li><li><a href="http://www.bestiario.org/">bestiario_</a></li></ul><h4>Industrie</h4><ul><li><a href="https://medium.com/visualizing-the-field/why-people-leave-their-data-viz-jobs-be1a7ab5dddc">If Data Visualization is So Hot, Why Are People Leaving?</a></li><li><a href="https://www.rgd.ca/insights/news_post/3219.php">Industry Insights from Montreal, Quebec | RGD</a></li></ul><h3>Livres</h3><p>Une bonne façon de se lancer en datavis est de trouver le bon type de livres. Pour ce qui est du développement, il y a beaucoup de livres pour <strong>apprendre D3.js</strong>. En voici une sélection. Ensuite, il y a les livres de type «table à café», qui sont à la fois <strong>inspirant</strong> et plus axé sur le design que sur la programmation. D’autres livres plus <strong>académiques</strong> permettent de bien comprendre l’espace conceptuel de la datavis.</p><h4>D3</h4><ul><li><a href="https://www.amazon.ca/s/ref=nb_sb_noss_2?url=search-alias%3Daps&amp;field-keywords=d3.js">Amazon.ca: d3.js</a></li><li><a href="https://github.com/d3/d3/wiki/tutorials#books">d3/d3</a></li><li><a href="https://www.packtpub.com/web-development/d3js-4x-data-visualization-third-edition">D3.js 4.x Data Visualization - Third Edition | PACKT Books</a></li><li><a href="https://www.manning.com/books/d3js-in-action-second-edition">Manning | D3.js in Action, Second Edition</a></li><li><a href="http://shop.oreilly.com/product/0636920026938.do">Interactive Data Visualization for the Web</a></li><li><a href="http://shop.oreilly.com/product/0636920037057.do">Data Visualization with Python and JavaScript</a></li></ul><h4>Listes de livres</h4><ul><li><a href="https://www.visualcinnamon.com/resources/learning-data-visualization/books">Reviews of Data Visualization books - Visual Cinnamon</a></li><li><a href="https://policyviz.com/better-presentations/data-viz-resources/data-viz-books/">DataViz Books - Policy Viz</a></li></ul><h4>Livre de table à café</h4><p><a href="https://www.amazon.com/Dear-Data-Giorgia-Lupi/dp/1616895322">https://www.amazon.com/Dear-Data-Giorgia-Lupi/dp/1616895322</a></p><p><a href="https://www.amazon.com/Book-Trees-Visualizing-Branches-Knowledge/dp/1616892188">https://www.amazon.com/Book-Trees-Visualizing-Branches-Knowledge/dp/1616892188</a></p><ul><li><a href="https://www.amazon.com/Book-Circles-Visualizing-Spheres-Knowledge/dp/1616895284">The Book of Circles: Visualizing Spheres of Knowledge</a></li><li><a href="https://www.amazon.com/Data-Visualisation-Handbook-Driven-Design/dp/1473912148">Data Visualisation: A Handbook for Data Driven Design</a></li><li><a href="https://www.amazon.com/Knowledge-Beautiful-Impossible-Invisible-Connections-Visualized/dp/0062188224">Knowledge Is Beautiful: Impossible Ideas, Invisible Patterns, Hidden Connections--Visualized</a></li></ul><h4>Académique</h4><p><a href="https://www.amazon.com/Elements-Graphing-Data-William-Cleveland/dp/0963488414">The Elements of Graphing Data</a></p><p><a href="https://www.amazon.com/Visualization-Analysis-Design-Peters-Series/dp/1466508914">https://www.amazon.com/Visualization-Analysis-Design-Peters-Series/dp/1466508914</a></p><h3>Education</h3><p>Le présentateur suivant parlera de l’éducation au niveau universitaire à Montréal. Mais plusieurs cours sont aussi disponibles en ligne.</p><ul><li><a href="https://www.mooc-list.com/tags/data-visualization">Data Visualization</a></li><li><a href="https://fr.coursera.org/learn/datavisualization">Data Visualization | Coursera</a></li><li><a href="https://www.udacity.com/course/data-visualization-and-d3js--ud507">Data Visualization and D3.js | Udacity</a></li></ul><h3>Outils et tendances</h3><p>En travaillant dans l’industrie, au coeur de la Silicon Valley, il est facile de voir certaines tendances se dessiner. Un grand nombre de librairies de graphiques sont apparues, ce qui montre un besoin d’utiliser des graphiques prêtes à l’emploi plutôt que de les développer. Les librairies développées avec React et D3.js sont les plus populaires. On voit aussi la montée de l’accélération graphique, utilisant le WebGL et la puissance des GPUs. Ces technologies permettent l’exploration interactive de plus gros ensembles de données. Du côté des applications, Tableau est encore un des leaders, mais Google et autres gros noms font une forte compétition avec des outils nuagiques.</p><h4>Librairies de graphiques</h4><ul><li><a href="https://github.com/wbkd/awesome-d3">GitHub - wbkd/awesome-d3: A list of D3 libraries, plugins and utilities</a></li><li><a href="https://plot.ly/">Interactive Data Visualization Tools &amp; Software | Plotly</a></li></ul><h4>React+D3</h4><ul><li><a href="https://emeeks.github.io/semiotic/#/">Semiotic Data Visualization</a></li><li><a href="http://www.reactd3.org/">A Javascript Library For Building Composable And Declarative Charts | React-D3</a></li><li><a href="https://vx-demo.now.sh/">vx | visualization components</a></li><li><a href="http://formidable.com/open-source/victory/">Victory</a></li><li><a href="https://uber.github.io/react-vis/">react-vis</a></li></ul><h4>WebGL, GPU = exploration BigData</h4><ul><li><a href="https://uber.github.io/deck.gl/#/">Large-scale WebGL-powered Data Visualization</a></li><li><a href="https://www.mapbox.com/mapbox-gl-js/example/heatmap-layer/">Create a heatmap layer</a></li><li><a href="https://www.mapd.com/demos/">MapD</a></li></ul><h4>BI exploration tools</h4><ul><li><a href="https://www.tableau.com/fr-fr">Tableau Software</a></li><li><a href="https://www.google.com/analytics/data-studio/gallery/">Interactive Reporting &amp; Dashboard Examples - Google Data Studio - Google</a></li></ul><h3>Tendances du design</h3><p>Une des grandes tendances du design est de concevoir des outils qui sont faits pour donner de l’information <strong>en un coup d’oeil</strong>. On peut aussi voir un intérêt grandissant pour les outils «<strong>invisibles</strong>», les commandes vocales, les assistants qui ne communiquent que par vibration, les visualisations ambiantes, etc. Un sujet qui pose encore de bons défis et font l’objet de beaucoup de discussions est la façon de visualiser l’<strong>incertitude</strong>. Il est encore tôt pour voir ce que la <strong>réalité virtuelle</strong> amènera de bon pour la datavis. Mais c’est un sujet en plein essor, qui fera peut-être l’object de notre prochaine conférence en direct avec San Francisco.</p><h4>Glanceable/invisible design</h4><p><a href="https://www.pinterest.ca/search/pins/?q=glanceable&amp;rs=typed&amp;term_meta[]=glanceable%7Ctyped">Pinterest</a></p><h4>Visualizing uncertainty</h4><ul><li><a href="http://www.visualisingdata.com/2015/02/references-visualising-uncertainty/">References for visualising uncertainty - Visualising Data</a></li><li><a href="https://flowingdata.com/2016/11/15/showing-uncertainty-during-the-live-election-forecast/">Showing uncertainty during the live election forecast</a></li><li><a href="https://blogs.scientificamerican.com/sa-visual/visualizing-uncertain-weather/">Visualizing Uncertain Weather</a></li></ul><h4>VR</h4><ul><li><a href="http://studio.knightlab.com/results/exploring-data-visualization-in-vr/uncharted-territory-datavis-vr/">Uncharted Territory</a></li><li><a href="http://store.steampowered.com/app/551960/DatavizVR_Demo/">DatavizVR Demo on Steam</a></li></ul><h3>Le journalisme informatique</h3><p>Jean-Hugues Roy, professeur à l’École des médias de l’UQAM, a ensuite parlé de l’état de l’<strong>éducation</strong> et du journalisme de données au Québec. Il constate la décroissance des projets de journalisme computationnel et le peu d’offre du côté des institutions. Cependant, la place est ouverte pour des journalistes talentueux, bien qu’ils soient généralement isolés, alors que les projets à rayonnement international sont généralement fait en équipe.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZaemeHZKq4cQ-1OCki0l0w.jpeg" /><figcaption>Jean-Hugues Roy</figcaption></figure><p>Voici le support visuel de sa présentation:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F242506208&amp;dntp=1&amp;url=https%3A%2F%2Fvimeo.com%2F242506208&amp;image=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F666382879_1280.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=vimeo" width="1280" height="720" frameborder="0" scrolling="no"><a href="https://medium.com/media/e7aa3ec721ed88b57038a02add5bc200/href">https://medium.com/media/e7aa3ec721ed88b57038a02add5bc200/href</a></iframe><h3>Conclusion</h3><p>Nous voulions montrer que cette industrie est de plus en plus organisée. Merci de venir aux meetups et de participer au déploiement de notre propre écosystème de la datavis québécois. Surveillez les <a href="https://www.meetup.com/vismtl/">prochains événements</a>!</p><p>Merci a notre commanditaire, <a href="https://www.wework.com/fr-CA/buildings/place-ville-marie--montreal--QC">WeWork Place Ville-Marie</a>, pour le superbe espace et pour tout ce qui était fourni, de la bière au projecteur à l’ambiance décontractée.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*5I1IBycTav244tsCN7JcLg.png" /></figure><p>Jessica, Christophe, Francis et Santiago</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1e8WC0ViW9NxKR1AGhyUEQ.jpeg" /><figcaption>L’équipe de VisMTL</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ef47608d3669" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A common language for product specifications]]></title>
            <link>https://medium.com/@christopheviau/a-common-language-for-product-specifications-a2f2d07e8?source=rss-a1206bba9736------2</link>
            <guid isPermaLink="false">https://medium.com/p/a2f2d07e8</guid>
            <category><![CDATA[unit-testing]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[chris viau]]></dc:creator>
            <pubDate>Tue, 01 Aug 2017 15:51:05 GMT</pubDate>
            <atom:updated>2017-08-01T15:51:05.184Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DBa885KYOnZduh7ryFa_xQ.jpeg" /></figure><h4>Pipeline and deliverables</h4><p>I like developing software products. I like the whole <strong>product development pipeline</strong>. My favourite workflow starts from product managers gathering what we call “user stories”, which we could define as a complete set of use cases to describe how to satisfy a need. These stories can be used to describe some deliverables or “epics” in task manager jargon, which will then be broken down further into bite-size tasks ready for the development team to work on.</p><p>A complete product development flow involves multiple people in interaction. Each stage of the process can sync-up by sharing specific <strong>deliverables</strong>. Typically, product managers will write down a set of user stories and feature lists, Designers will turn them into wireframes and flow diagrams, project managers will break the features down into tasks, etc. It can be hard to translate these documents across the pipeline. What if we could find a common language, so everybody can collaborate on building a shared vision of the product?</p><p>Let me share my experience using unit tests syntax to achieve this <strong>common operating picture</strong>.</p><h4>User Story</h4><p>A product manager collects user stories describing a user need. They are phrased in a way that doesn’t bother with design. For example “as a user, I want to click on a button” is not a user story. Here is an example:</p><blockquote>“As a user, I want to be able to annotate a dashboard with links, pictures and basic text formatting.”</blockquote><h4>Feature list</h4><p>From these stories a product manager can translate it into a user task and a set of features that need to be there to accomplish it.</p><blockquote>* Adding link and picture to an annotation widget<br>* Formatting text with bold, italic, underline</blockquote><h4>Design</h4><p>A UX/UI designer can decide what actual UI elements and interaction patterns could be used to expose these features.</p><blockquote>* Switching the widget to edit mode by clicking on it<br>* Adding link, picture, format text with bold, italic, underline by entering some markdown markup</blockquote><h4>Unit tests specs</h4><p>This level of specifications is where a lot of different people can collaborate since it’s the place to decide exactly how each features will be implemented. It may require sketching some UIs, revisiting some assumptions, maybe even getting back to the client. But once these specs are defined, everything should be clear.</p><blockquote>* it switches to a textarea on click<br>* it saves text and switches to formatted text display on blur<br>* it changes mode to non-editable on blur<br>* it should automatically focus on textarea in edit mode<br>* it should format markdown text to html<br>* it supports multiple markdown features</blockquote><h4>Unit tests</h4><p>The developer can then turn these specs into actual tests that will run against some code. The implementation details are coupled with the tech stack and not important here. But they rest on solid grounds, on a set of fine-grained specs validated by the whole team.</p><blockquote>it(“should format markdown text to html”, () =&gt; {</blockquote><blockquote>mountComponent({text: “# test”})<br> const converted = fixture.querySelector(“.formatted-text”).innerHTML<br> expect(converted).to.equal(“&lt;h1 id=\”test\”&gt;test&lt;/h1&gt;”)</blockquote><blockquote>})</blockquote><p>That’s it. I just wanted to show that the deliverables coming out of each stage of a development pipeline are not so different, but going from general to specifics, from a user need to a technical implementation. And the unit test specs can be a place where decisions are validated and everybody can sync around them, from product managers to designers to developers. Have fun unit testing!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a2f2d07e8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Progressive refactoring]]></title>
            <link>https://medium.com/@christopheviau/progressive-refactoring-979c2c7e40b1?source=rss-a1206bba9736------2</link>
            <guid isPermaLink="false">https://medium.com/p/979c2c7e40b1</guid>
            <category><![CDATA[testing]]></category>
            <category><![CDATA[dataviz]]></category>
            <category><![CDATA[javascript-development]]></category>
            <category><![CDATA[refactoring]]></category>
            <dc:creator><![CDATA[chris viau]]></dc:creator>
            <pubDate>Thu, 20 Jul 2017 18:21:37 GMT</pubDate>
            <atom:updated>2017-07-21T16:03:33.055Z</atom:updated>
            <content:encoded><![CDATA[<p>Refactoring is hard. And the value it brings, like stability and scalability, is not easy to sell to a non-developer. But here I want to share a strategy for progressive refactoring, one which doesn’t block delivering business value, but which helps <strong>transitioning your code</strong> to a target architecture.</p><h3>When should I stop fixing and start refactoring?</h3><p>To answer this, I’m always watching for symptoms of <strong>Fear-Driven-Development</strong> (<a href="http://wiki.c2.com/?FearDrivenDevelopment">FDD</a>): when you are afraid to touch the codebase because unrelated things can break in unexpected ways.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xnzZC9f1rjQEBL0ZXdo_LQ.jpeg" /><figcaption>When you are diagnosed with FDD, get ready for an invasive surgery.</figcaption></figure><p>When weird bugs seem to appear from other dimensions, you know it’s time to fight back. But where to start? I suggest starting by <strong>designing an API</strong> (Application Programming Interface) to split your code into modules, then running as fast as possible to <strong>get there</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/955/1*Cnw5UZpfp4gnXa5WWRCRHg.jpeg" /><figcaption>Aliens are not supposed to come and kill your dog.</figcaption></figure><h3>Stop the bleeding</h3><p>I usually start by sketching a diagram of the current architecture to try to find where to draw the line between modules. For example, in my datavis work, I try to always keep a separation between data, rendering, events and states. Here is an example of an architecture diagram. The current architecture has some intensive coupling between all the modules. So the goal is to extract components, for example a state manager and a data store, to start seeing a path to a proper dataflow.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WJL6rLUSRqdLWV0NgLrE9w.png" /><figcaption>Diagram of current and target architecture</figcaption></figure><p>Once I identified the coupling in the current code and how it can be encapsulated into the target components, I need a strategy to start building walls around the new modules.</p><h3>Designing the API</h3><p>When wrapping code inside modules, the most important is to <strong>design around APIs</strong>. What do I mean by API?</p><p>The term “API” can mean a lot of different thing. I use it to describe the “interface” that a module exposes and uses to communicate with other modules. Let’s take a chart module as an example. The API is what is exposed, what you would document for other developers to use your charts. It’s also what should be unit tested. Here is an <strong>example of a chart API</strong>:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/2594115669b2fcda939a8f859b71c5e7/href">https://medium.com/media/2594115669b2fcda939a8f859b71c5e7/href</a></iframe><p><strong>Committing to an API</strong> is the key. So you need to carefully evaluate if you have the right one. For example, here I want to decouple data from rendering. The current API (inspired by Dc.js) takes a dimension and a group, expecting the chart to take care of transforming the dataset from this information. But what I want is to delegate this to a data manager, which will provide clean data to the chart. I like charts to be dumb containers: pass data and config in, get a chart out.</p><p>I also want the chart to <strong>expose events</strong>. An external event manager could centralize the interaction logic. So here I expose a generic .on() method to the chart API.</p><p>Another change that seems cosmetic is to replace the chain of setter functions by a single .config() method. This simpler API will help me clean-up a lot of internal boilerplate code. But it is also more <strong>declarative</strong> than functional, which is consistent with the target API I have in mind for the other modules.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/cc0ee092e64f90e03b39b4c06313f42a/href">https://medium.com/media/cc0ee092e64f90e03b39b4c06313f42a/href</a></iframe><h3>Coding the interface, unit tests, docs</h3><p>Once you have a clear separation of concerns between all your modules and you have an API you can commit too, it’s time to write down the contract. The best way to do it is with <strong>unit tests</strong>. I wrote a short chapter about unit tests for charts in this <a href="http://backstopmedia.booktype.pro/developing-a-d3js-edge/chapterx/">little book</a>.</p><p>Once you have the right API, you can code an interface. Here is an example of an empty shell that could match with a set of unit tests. Start porting over all features until all tests pass.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/929a7b36d060ba77080fcea8c6cb82e2/href">https://medium.com/media/929a7b36d060ba77080fcea8c6cb82e2/href</a></iframe><p><strong>Documenting</strong> the code can be rewarding too. I really like <a href="http://documentation.js.org/">documentation.js</a> for that. Writing docs is a good way to think about your code as modules. What you see, test and document, is only what is exposed. That’s also what other modules should see. The rest is a blackbox to them, but it’s your playground! Once a strong API is in place, you are free to experiment as much as you want, as soon as all your tests pass.</p><h3>Watertight modules</h3><p>Now that you defined your APIs, one tricky part is to find how to consolidate the various part of the code behind each module API. Start from the areas that already respect the API contract. Find a piece that should not stick out and bring it back in. Decide which part should stay messy while others are getting rectified. It could involve a bit of whac-a-mole, but keep hitting until <strong>everything is hidden</strong> under the APIs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/535/1*xMQ33NyG1t0VBpmGrPJQ8A.jpeg" /><figcaption>Keep hitting until nothing moves</figcaption></figure><h3>Continue delivering value</h3><p>Refactoring takes time and hard work. Sometimes you need a scalpel, sometimes a chainsaw. But while you are working on your Frankenstein, how do you <strong>keep delivering value</strong>? One answer is to identify the part of the code that would fix a batch of issues in a single refactoring session faster than it would take to fix them one by one. Another one is to implement a new feature that clearly shows what the refactoring made possible. Also, work can still happen on the legacy part of the code, even if you know it will be replaced soon.</p><p>The strategy depends on a lot of factors (deliverables, resources, replacing vs refactoring). The whole process can take many forms. But one thing must absolutely be in the plan: a clear idea of where the <strong>refactoring session ends</strong>. So carefully plan a target, grab your chainsaw and make your way there as fast as possible.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/736/1*Z0fHOInHUo1K672Pcre2nw.jpeg" /><figcaption>Aggressive refactoring</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=979c2c7e40b1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Measuring to Manage: A New Food Security Visualization Initiative by Planet OS]]></title>
            <link>https://medium.com/planet-os/measuring-to-manage-a-new-food-security-visualization-initiative-by-planet-os-efa853693084?source=rss-a1206bba9736------2</link>
            <guid isPermaLink="false">https://medium.com/p/efa853693084</guid>
            <category><![CDATA[environment]]></category>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[security]]></category>
            <category><![CDATA[climate-change]]></category>
            <category><![CDATA[food]]></category>
            <dc:creator><![CDATA[chris viau]]></dc:creator>
            <pubDate>Thu, 08 Jun 2017 00:09:51 GMT</pubDate>
            <atom:updated>2017-09-06T18:49:41.671Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7NWGWNWFH-njF3jelkDOuA.jpeg" /></figure><p><em>The article was first published on </em><a href="https://planetos.com/"><em>https://planetos.com/</em></a><em> in October 2015.</em></p><p>We are delighted to present a new initiative to create and publish a series of data visualizations that highlight the most alarming environmental trends contributing to the global decrease in food security. The visualizations to be debuted today at the Zoological Society of London <em>The future of food the future of biodiversity?</em> conference in London, United Kingdom will seek to leverage our big data platform for the closer analysis of the global food security crisis. Our partners at <a href="https://www.intertrust.com/">Intertrust Technologies</a> will also present their work digging through the data to secure our global food supply.</p><p>Environmental trends to be highlighted include how climate change-induced ocean acidification is threatening marine organisms; how abnormal weather events, such as the California drought, are impacting regional food production; and how economic growth affects our natural ecosystems, such as forests. Reductions in global marine organism counts, curtailed regional food production, and reduced forest area are just some of the many trends threatening to exacerbate the global food security crisis.</p><blockquote>We call this series of visualizations Measuring to Manage because we believe one cannot manage what one cannot measure. Access to data — be it food data, climate data, rainfall data, population data — is paramount to ensuring that good food policy decisions are made.</blockquote><p>To create this first set of visualizations, our team tried to work with only readily-available public data to help audiences better understand what is possible (and what isn’t) with currently accessible open data. All visualizations here were thus created using existing public data available from reputable sources. With this in mind we also hope to spur discussion around how more data can be made available to further analyze these environmental trends and help solve the food security crisis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cAj23qawhEMN_wPnN2R0QA.jpeg" /><figcaption>“Measuring to manage” poster</figcaption></figure><p>Our team is honored to be able to debut these images at the Zoological Society of Londons conference alongside world-renowned researchers, academics, students, industry professionals and our partners at Intertrust Technologies Corporations. Intertrust has written more extensively about their work digging through the data to secure our food supply <a href="https://web.archive.org/web/20170224225628/http://www.intertrust.com/blog-other/2015/10/13/digging-data-to-secure-our-food-supply">in this blog</a>. Subsequent activities to further advance the initiative will soon be announced.</p><p>Here below are three of the first visualizations and a brief description for each.</p><h3>Ocean Acidification</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tpmd2tjRZ9MWlXGOib4-wQ.jpeg" /><figcaption>Ocean acidification affects marine organisms and fishing communities</figcaption></figure><p>Using a dataset and model developed by the Woods Hole Oceanographic Institution, we have created a visualization showcasing the increasing acidification of the worlds oceans. Increasing acidity lowers carbonite ion concentrations, an essential ingredient necessary for marine organisms such as clams, corals, planktons and mollusks to develop hard shells and skeletons.</p><p>The inability of these organisms to properly form internal structures causes them to spend more energy regulating internal chemistry, reproduce improperly, die earlier, or migrate to nontraditional habitats. These effects on marine organisms have consequences on delicate fishing stocks and the communities that rely on them for economic sustenance.</p><h3>Drought In California</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7ekf1hFJStr-DMFh29aq6g.jpeg" /><figcaption>The historic drought in California threatens agricultural production</figcaption></figure><p>California is amidst a historic drought that sees no signs of letting up. These California map visualizations reproduced here were originally created by the National Drought Mitigation Center at the University of Nebraska-Lincoln to track the progress of drought intensity from 2000 to 2014.</p><p>We have further analyzed drought intensity levels for all California counties to showcase a stark portrayal of just how severe the drought has become at a localized level in 2014. Although total production losses have not been conclusively determined, the decrease of California agricultural production due to water shortages would affect consumer prices and threaten the livelihood of farming communities in California.</p><h3>GDP And Forest Area</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*p7N_3s_4lpFEpP5Tj_1w3A.jpeg" /><figcaption>How economic growth may affect global forest area</figcaption></figure><p>Our team wanted to better understand the relationship between global forest area and gross domestic product (GDP), as a proxy for economic growth. Vietnam, for example, grew its GDP by 6.96% annually between 1990 and 2011, and increased its forest area by 17.2% in the same time. It ranks along with China, Spain, and Italy as four of the top five countries with the largest annual net gain in forest cover. Vietnam’s economic growth has allowed it to fund various national reforestations efforts and introduce legislation to incentivize improved forest management policies.</p><p>However, Indonesia and certain other countries listed in red, have seen their forest area further reduced while increasing GDP. More data is needed to conclusively define the exact correlation between GPD and forest area. Nevertheless, economic growth inevitably plays some type of role in how we will manage this current food security crisis.</p><p><em>If you’re curious about renewable energy, the role of data in the energy industry and tips and tools for data-driven businesses, then you are most certainly in the right place. I invite you to look at what we are doing with </em><a href="http://bit.ly/2jZqAs0"><em>big data analytics in the renewable energy industry</em></a><em>. Also, we are sending our blog posts also via email once a month, so simply </em><a href="http://bit.ly/2jXBBJn"><em>add yourself to the list</em></a><em> and find the next issue in your mailbox.</em></p><h4>Previous article→ “<a href="https://medium.com/planet-os/a-scalable-way-to-access-weather-data-to-address-climate-risk-19b62b3b5886">A Scalable Way To Access Weather Data To Address Climate Risk</a>”</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=efa853693084" width="1" height="1" alt=""><hr><p><a href="https://medium.com/planet-os/measuring-to-manage-a-new-food-security-visualization-initiative-by-planet-os-efa853693084">Measuring to Manage: A New Food Security Visualization Initiative by Planet OS</a> was originally published in <a href="https://medium.com/planet-os">Planet OS (by Intertrust)</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Graphiques Dégénérés: An Exhibition of Datavis Failures]]></title>
            <link>https://medium.com/@christopheviau/graphiques-d%C3%A9g%C3%A9n%C3%A9r%C3%A9s-an-exhibition-of-datavis-failures-5bcb7f3bc705?source=rss-a1206bba9736------2</link>
            <guid isPermaLink="false">https://medium.com/p/5bcb7f3bc705</guid>
            <category><![CDATA[d3js]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[generative-art]]></category>
            <dc:creator><![CDATA[chris viau]]></dc:creator>
            <pubDate>Wed, 17 May 2017 13:44:07 GMT</pubDate>
            <atom:updated>2017-05-17T14:11:27.945Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Graphiques Dégénérés/Degenerate Graphics is an exhibition of failed datavis, inspired by </em><a href="https://twitter.com/search?q=%23d3brokeandmadeart"><em>#d3brokeandmadeart</em></a><em>, that took place at La Chambre Blanche in Québec City, Québec, Canada, organized by </em><a href="http://www.chambreblanche.qc.ca/en/"><em>La Chambre Blanche</em></a><em>, </em><a href="http://www.04h11.com/"><em>04h11</em></a><em> and </em><a href="https://twitter.com/d3visualization?lang=en"><em>Chris Viau</em></a><em>.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YM_K8fVczf2HLuUReW1U1g.jpeg" /><figcaption>Micah Stubbs</figcaption></figure><p>Data visualization is the art of turning data into meaningful graphics. But sometimes, this process doesn’t go as planned. When it fails in style, the results can be surprisingly beautiful. I come from an art background, so I asked myself the question: can we present <strong>datavis failures</strong> in an art gallery and treat it as a form of <a href="https://www.fastcompany.com/3034296/inside-the-bizarre-phenomenon-known-as-glitch-art">Glitch art</a>? That’s what we did. And it was fun.</p><h3>VISquebec on Twitter</h3><p>17h à la @CHAMBREBLANCHE, 8ème de #VISQuebec avec @04h11. @micahstubbs en meetup et l&#39;expo Graphiques dégénérés #d3brokeandmadeart #dataviz</p><p>It all started from an idea mentioned on the <a href="https://d3js.slackarchive.io/d3brokeandmadeart/page-2">#d3brokeandmadeart</a> <a href="https://d3-slackin.herokuapp.com/">D3 Slack</a> channel. We curated some of the best datavis failures, printed them, pinned them on the wall of an arts center and organized a <a href="https://twitter.com/visquebec?lang=en">VisQuébec</a> <strong>meetup</strong> for the opening.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Pd-7D5zyI_XwdqQbUWEYJA.jpeg" /><figcaption>VisQuebec meetup at exhibition opening</figcaption></figure><p>I started by presenting the concept. First, <strong>the venue</strong>. As François Vallée describes it:</p><blockquote>La Chambre Blanche is dedicated to the experimentation of visual and digital arts, in particular practices exploring new technologies within an international network of exchanges and residencies. This laboratory is built around research, creation, production, documentation training and broadcasting.</blockquote><p>I have collaborated with La Chambre Blanche since 1998 and together we organize the VisQuébec meetup, so it was natural to host the exhibition there.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FHvRExD-MahcSWvkNqCqzw.jpeg" /></figure><p>I then described what we mean when we talk about a <strong>datavis failure</strong>.</p><h3>VISquebec on Twitter</h3><p>Ouverture du meetup avec @d3visualization sur les Graphiques Dégénérés. #d3brokeandmadeart #D3js #8emeVISQuebec #dataviz</p><p>The <strong>data visualization process</strong> can be illustrated as a pipeline from data to graphics (and then interactivity), going through a lot of steps, like data transform, visual abstraction, graphical transform, etc. At each of these steps some code can go rogue.</p><p>I like to see failures from multiple datavis practitioners because it tells a lot about their process. For example, <a href="https://medium.com/u/f50f8c4bbcbd">Moritz Stefaner</a> is well known for his design skills, always pushing the boundaries of data representation. So here is what happens when he fails:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*AQRjIMU_vbOh4FYgZ2FQ-w.gif" /><figcaption>Moritz Stefaner</figcaption></figure><p>A projector was beaming this live rendering of Moritz’s Processing code on the wall. What was he trying to do?</p><blockquote>I wanted to demonstrate that the “hexagonal” structures bees build are actually just squeezed circles, so the hex structure is really just an emergent property of the fact they are tightly packed together when building and some basic geometry. So, I thought, I’ll just build a couple of “wheels” with one particle in the center, a set of particles — all connected in a ringwise fashion by springs, and also connected to the center node. If those would all gravitate to the center, and repel each other, I should get a nice squeezing behavior, right?</blockquote><p>After this introduction, I had the pleasure to introduce <a href="https://twitter.com/micahstubbs">Micah Stubbs</a>, who helped organize the show and was visiting from the Bay Area.</p><h3>LA CHAMBRE BLANCHE on Twitter</h3><p>Merci à Micah Stubbs pour sa présentation à VISQuébec! @micahstubbs @d3visualization @04H11 @ejacquin #meetup #datavis #chambreblanche</p><p>He talked about <a href="https://twitter.com/search?q=%23D3brokeandmadeart">#d3brokeandmadeart</a>, <a href="https://twitter.com/accidental__aRt">@accidental__aRt</a>, and ways to share code, modify it and share it, especially when it fails. Here are some links he mentioned:</p><p><a href="https://bl.ocks.org/micahstubbs">Micah Stubbs’s Blocks</a> <br><a href="https://bl.ocks.org/micahstubbs/3f439df92579c5bb2902fab15742ba87">Graph community clustering</a> on bl.ocks<br><a href="http://blockbuilder.org/micahstubbs/3f439df92579c5bb2902fab15742ba87">Graph community clustering</a> on Blockbuilder<br><a href="https://twitter.com/search?f=tweets&amp;vertical=default&amp;q=d3brokeandmadeart%20|%20(15)%20d3brokeandmadeart">#d3brokeandmadeart</a><br><a href="https://bl.ocks.org/curran/57d3608636fa0515b5f1b84f606d939a">Animated fractal pie chart</a> by Curran Kelleher<br><a href="https://twitter.com/accidental__aRt">@accidental__aRt</a><br><a href="https://www.reddit.com/r/prettybugs/">Museum of Pretty Bug</a><br><a href="https://www.reddit.com/r/glitch_art/">Glitch_art</a> reddit: dedicated to the art of databending<br><a href="https://badsamples.tumblr.com/">Bad Samples</a><br><a href="https://photos.google.com/share/AF1QipOFIpZkDLxW6RV098fK1w3m9oQNNtzrzh8J6y2HbPxvdPvu3F28M4jc8pvyEjBhYw?key=enhTUEM4aDhhbVVpcms2TVRuTE9SRDZqNE1GbFhB">Datavis failures selection</a></p><p>I then told the story of some of the prints that were exhibited.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kTT3jlgujAkjhMk-rdMRpg.jpeg" /></figure><p>For example, this is a map of oceans surface temperature. But I needed a projection that would better show the poles, which is problematic with mercator. So I tried a mercator transverse. And failed.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PaoBHqn64r-8a7nd7pCG3A.jpeg" /><figcaption>Chris Viau</figcaption></figure><p>This one from <a href="https://medium.com/u/89d2e252d523">Kai Chang</a> represents the <a href="https://en.wikipedia.org/wiki/Molniya_(satellite)">Molniya satellites</a> trajectories in a variant of this same projection.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*THVSix-YlsT5rgpyaL_0yg.jpeg" /><figcaption>Kai Chang</figcaption></figure><p>I like how this one from <a href="https://ffctn.com/en/index">FFunction</a> looks like an old polaroid of a matrix printer attempt at drawing a map.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ad00p2T3qEPdh0I20MVY4Q.jpeg" /><figcaption>FFunction</figcaption></figure><p>This one is different. It’s from a generative artist: <a href="http://www.chaoscopia.com/demarche_en.html">Etienne Saint-Amant</a>. He is also an expert in scientific visualization, but I wanted to show that being data-driven was not the point of this particular exhibition. The question here is: what will a generative artists consider to be an error and not a happy accident? Datavis designers usually have a very specific goal. So anything that goes against it, even if visually interesting, is a problem. But an artist has a whole creative space opened for deciding what is junk and what will become art. On this piece, the algorithm was run too deep, rendering some chaotic layering of over-detailed textures in a somewhat confused composition. Which is nice.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*qQ-14GCWVTnhT4J7hf3pmg.jpeg" /><figcaption>Étienne Saint-Amant</figcaption></figure><p>The exhibition was a success. I really think it’s a concept that is easy to implement and that deserves to happen again. If you want to give it a try, we would be glad to help you organize one in your own city.</p><h3>@micahstubbs on Twitter</h3><p>@CHAMBREBLANCHE @VISquebec beginning now ✨ #d3brokeandmadeart @CHAMBREBLANCHE works from @syntagmatic @currankelleher @Elijah_Meeks @mf_viz and more 🌅</p><p>Here is the complete video of the meetup. My part is in French and Micah’s part is in English. Have fun!</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F217426773&amp;url=https%3A%2F%2Fvimeo.com%2F217426773&amp;image=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F634650169_1280.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=vimeo" width="1280" height="720" frameborder="0" scrolling="no"><a href="https://medium.com/media/cf39b2079ec296cc39da5ff392f1e0f5/href">https://medium.com/media/cf39b2079ec296cc39da5ff392f1e0f5/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5bcb7f3bc705" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[6 weird tricks for insane chart performance]]></title>
            <link>https://medium.com/@christopheviau/6-weird-tricks-for-insane-chart-performance-afd29f90f271?source=rss-a1206bba9736------2</link>
            <guid isPermaLink="false">https://medium.com/p/afd29f90f271</guid>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[performance]]></category>
            <category><![CDATA[charts]]></category>
            <category><![CDATA[d3js]]></category>
            <dc:creator><![CDATA[chris viau]]></dc:creator>
            <pubDate>Fri, 12 May 2017 22:03:26 GMT</pubDate>
            <atom:updated>2017-05-13T14:44:23.844Z</atom:updated>
            <content:encoded><![CDATA[<p>How do you like my clickbait title? While you are here, I want to share some tricks to handle larger datasets without locking the UI or slowing down your chart interaction.</p><h3>Trick 1: Progressive rendering</h3><p>A common problem is UI locking while a piece of graphics is rendering. One solution for freeing the UI is to <strong>chunk the rendering</strong> and render each piece in a loop, handing over the thread each time to the UI. For this trick, I often use a little tool called r<a href="https://github.com/biovisualize/render-slicer">enderSlicer</a>.</p><p>In this <a href="http://bl.ocks.org/biovisualize/f84111e34edc6d594216">example</a>, I misused the rendering to illustrate the point. Instead of only rendering the section visible in the window, I re-render the whole line chart on every brush events. I also set the render-slicer parameters to be very slow, so we can see the <strong>effect of progressive rendering</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*h8m29jUBvQPPyRI9rT7jQQ.gif" /></figure><p>It would be fun to try other patterns for progressive rendering, like iteratively refining the resolution of a grid, or like the example from the original library called <a href="http://bl.ocks.org/syntagmatic/raw/3341641/">renderQueue</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*OIh5djRrCuqWoEhGERsd9w.gif" /></figure><p>Render-slicer uses <strong>requestAnimationFrame</strong>, which is a timer that will go as fast as the browser can go. Other optimizations I used on this chart is <strong>throttling</strong> on the brush events, so it doesn’t try to re-render, for example, more than once each 300ms.</p><h3>Trick 2: Buffer canvas</h3><p>One challenge I often meet is the handling of <strong>streaming charts</strong>: new data is coming periodically, the whole chart is shifting to leave room for the new data points.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*zzbG6by4Ynv7h-iqTK81QA.gif" /><figcaption>Basic streaming example</figcaption></figure><p>In this basic <a href="http://blockbuilder.org/biovisualize/ad7f1a572b526af1e068">streamgraph example</a>, the I’m just <strong>redrawing everything</strong> at every tick. It doesn’t matter for this toy example. But what if you need to stream more data points, at higher frequency, or if your graphic computation is expensive, or if you have tons of these charts on the same panel?</p><p>My first reaction to this situation would be to <strong>find a better abstraction</strong>, like downsampling, focus+context, etc., more suitable to what the human visual bandwidth can handle. But I saw some cases where data overload was a necessary evil (which I could share in a future blog post). For now, let’s assume that the goal is to <strong>scale up</strong> a streaming chart to more data points than the browser can handle without noticeable UX penalty.</p><p>While designing the <a href="https://github.com/boundary/firespray">Firespray</a> streaming charts library, I made an example inspired by side-scrolling games and infinite-scrolling tables like <a href="https://clusterize.js.org/">clusterize.js</a> or <a href="http://airbnb.io/infinity/">∞.js</a>. The idea is to always have a <strong>buffer canvas</strong> preloaded, ready to fill the gap when the main canvas shifts left. As soon as the main canvas is translating out of the view, it jumps back to the right of the second canvas, which is now acting as the main one. And the cycle continues.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*ZZkMQePiJRB6cOBof94-Sg.jpeg" /><figcaption>Canvas buffer swapping</figcaption></figure><p>There are multiple advantages to this technique. First, instead of redrawing everything on each tick, you only draw everything once a cycle and <strong>shift the canvases</strong> using a simple scrollLeft or even a nice CSS animation (which is <a href="https://www.sitepoint.com/introduction-to-hardware-acceleration-css-animations/">hardware accelerated</a>). I also like that you can lazy-load the data, draw it on the canvas and then leave it to the GC to free the memory.</p><p>Here is an <a href="http://blockbuilder.org/biovisualize/549bfceb0938859fea34">example</a> with a canvas buffer on both sides, ready for scrolling in <strong>both directions</strong>. I’m sure you will excuse my prototype-quality code.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*lxhafbmYGDa79e_uveFfTw.gif" /></figure><p>One trade-off is that you need to be able to afford this buffer zone, but it can be as short as a single sample. I realized that it often is more important for the user to see a chart <strong>ticking exactly on clock</strong>, even if it’s one sample behind. Real-time visualization can mean many things, but for me it’s all about trust in the system being, not fast, but dependable.</p><h3>Trick 3: Quadtree and bisection</h3><p>Using the previous idea of sliding canvases, here is another <a href="http://bl.ocks.org/biovisualize/raw/04aba0475e5b2a41f0a6/">example</a> to show one solution for fast interactivity. Even with 100 line charts, each showing 300 data points, the hovering is totally smooth. The trick is to look for data and layout in the data space using <strong>bisection</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/1*xEWZBQDF6ZBR0kYBge0feg.gif" /><figcaption>Fast tooltip using bisection</figcaption></figure><p>If you are in a 2D space, the equivalent would be to use a <strong>quadtree</strong>. Here’s an <a href="https://bl.ocks.org/mbostock/4343214">example</a> by <a href="https://medium.com/u/c918c478bb49">Mike Bostock</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/494/1*ILUqqCiLcyr5fVOIQ2bO-A.gif" /><figcaption>Rectangular selection using a quadtree</figcaption></figure><h3>Trick 4: Breaking free from the thread</h3><p>When the graphic calculation is expensive, one solution is to compute it on a <strong>Web Workers</strong>. Here is another nice <a href="https://bl.ocks.org/mbostock/01ab2e85e8727d6529d20391c0fd9a16">example</a> from <a href="https://medium.com/u/c918c478bb49">Mike Bostock</a> for computing a static graph layout.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eHvbDfzz8zuSSmz9s5MMgQ.png" /></figure><p>Another way of breaking free from the single Javascript thread is to use the <strong>GPU</strong>. I was playing with OpenGL a lot at some point before D3.js came out.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fplayer.vimeo.com%2Fvideo%2F13979287&amp;url=https%3A%2F%2Fvimeo.com%2F13979287&amp;image=https%3A%2F%2Fi.vimeocdn.com%2Fvideo%2F81209254_640.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=vimeo" width="640" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/9deaa340d39825648afde788b851ab5d/href">https://medium.com/media/9deaa340d39825648afde788b851ab5d/href</a></iframe><p>But don’t take this as an example of a good datavis. <a href="https://www.mapbox.com/mapbox-gl-js/api/">Mapbox GL</a> offers a way better example of the power of <strong>webgl</strong>, in this case<strong> </strong>for rendering interactive maps. Another very impressive framework is <a href="https://eng.uber.com/deck-gl-framework/">deck.gl</a> by the Uber datavis team.</p><h3>Trick 5: Stream loading</h3><p>Another process that can block the UI is loading data. My favorite solution is <strong>streaming data</strong> with <a href="http://papaparse.com/">Papaparse</a>. Papaparse is not just a csv parser. It can also stream data by chunking it and transferring each data parts sequentially. This can even use Web Workers with a simple config switch. I should really bring one of my examples online, but the fun part is that you can render each data chunk directly to a canvas as soon as it arrives, and then discard it from memory. So you get the same idea as with renderSlicer but across the whole data pipeline.</p><h3>Trick 6: Fast data layer</h3><p>I often see the data layer as being the bottleneck. There are of course tons of powerful backend technologies to get things done efficiently. But sometimes, we can forget details that can have a big impact, for example <strong>data compression</strong>. I liked how <a href="http://nytimes.github.io/tamper/">Tamper</a> takes care of finding the best compression algorithm for your categorical data, especially when coupled with <a href="http://nytimes.github.io/pourover/">PourOver</a>, also by the NYTimes kind of a <a href="http://square.github.io/crossfilter/">Crossfilter</a> for categorical data.</p><p>Crossfilter and PourOver make filtering, aggregating and other data operations <strong>client-side</strong> very efficient. <a href="https://github.com/mapd/mapd-charting">MapD-charting</a> goes one step further in connecting to their very <a href="https://www.mapd.com/products/immerse/">fast database</a> leveraging the power of GPUs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LRl4yhoRZryaSiA4XAk-FQ.gif" /><figcaption>MapD Charting example from their Github repo</figcaption></figure><h3>Conclusion</h3><p>A lot can be done to optimize client-side rendering, and there’s a lot more we could discuss (caching, server-side rendering, aggregating, LOD, blitting, using shaders, using <a href="http://superconductor.github.io/superconductor/">webcl</a>, etc.). But I just wanted to share some solution I had to use so far in my datavis work. Thanks for your attention!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=afd29f90f271" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Insights from designing Datahub-ui in D3.js]]></title>
            <link>https://medium.com/planet-os/insights-from-designing-datahub-ui-in-d3-js-d8f0f5da0b82?source=rss-a1206bba9736------2</link>
            <guid isPermaLink="false">https://medium.com/p/d8f0f5da0b82</guid>
            <category><![CDATA[3d]]></category>
            <category><![CDATA[ui]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[javascript]]></category>
            <dc:creator><![CDATA[chris viau]]></dc:creator>
            <pubDate>Tue, 21 Feb 2017 04:44:51 GMT</pubDate>
            <atom:updated>2017-02-22T03:47:48.368Z</atom:updated>
            <content:encoded><![CDATA[<p>In a previous post, I talked about D3 modularity.</p><p><a href="https://medium.com/@christopheviau/d3-js-modularity-d5eed78ba06e">D3 is now modular</a></p><p>Now I want to share some modular patterns that I’ve been using for a while in my datavis engineer job at <a href="https://medium.com/planet-os">Planet OS</a>. On top of our environmental data platform called <a href="http://data.planetos.com/">Datahub</a>, we build nice dashboards, mostly for clean energy companies like wind farms, that we call <a href="https://planetos.com/powerboard/"><strong>Powerboard</strong></a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K9ObEK-9kOLcdtsgdVVDbg.png" /><figcaption>One of the Powerboards we developed</figcaption></figure><p>I already wrote about a <strong>pipeline pattern</strong> that I was using for Cirrus.js.</p><p><a href="https://planetos.com/blog/cirrus-js/">Why I Built Cirrus.js - Planet OS</a></p><p>I continued using this pattern for <a href="https://github.com/planet-os/ui-components/tree/master/datahub-ui"><strong>Datahub-ui</strong></a>, the internal UI toolkit we are developing to quickly build Powerboards. Here are some lessons I learned, after almost 2 years using this pattern.</p><h3>Patterns vs Abstractions</h3><p>I like this pipeline pattern because it’s just a way to structure code. The same way I like D3 because it encourages the use of declarative and functional patterns to describe how data should be mapped to graphics. I prefer to focus on patterns like MV* architecture and unidirectional data flow than on abstractions like digest cycles and Virtual DOM.</p><p>I like to be able to <strong>encapsulate</strong> functional components at a very low level. A component can be as simple as configuring a D3 scale:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/6e216c0bfd7f577dbbca13a618cc5fda/href">https://medium.com/media/6e216c0bfd7f577dbbca13a618cc5fda/href</a></iframe><p>These tiny modules are piped together using a simple <strong>pipeline function</strong>. Each module receives a clone of the global object config and can add items to it to pass it down the pipe.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/d70cfa7bd3e946f427e96d42fbf6fa69/href">https://medium.com/media/d70cfa7bd3e946f427e96d42fbf6fa69/href</a></iframe><h3>It All Starts with Data</h3><p>One of the main tool we provide at Planet OS is called <a href="http://data.planetos.com/"><strong>Datahub</strong></a>. It’s basically a consistent interface to access Earth science data. Most of these datasets are already available online, but are usually pretty hard to access and manipulate. Having a consistent interface really shines when you need to search and compare datasets, merge data from multiple sources and use common tools to directly use clean data in a simple format.</p><blockquote>Embedding maps with multiple layers of raster and point data, streaming multidimensional timeseries, computing alerts by comparing finance data with weather forecast, that’s the kind of data merging we do everyday at Planet OS. Modularity at the dataset level is an important subject that I don’t see discussed so often in the D3 community.</blockquote><p>In my experience, the data straight from an API is rarely in the best format for the way the charts may want to receive it. Sometimes a chart has to aggregate some values, split them into layers, group into a hierarchy, just in the process of mapping to graphical elements. When I design charts, I try to choose a data format that makes sense from the datavis pipeline standpoint. Then I write adapters to transform from data API format to chart input format. So the first modules in my pipeline are usually <strong>data adapters</strong> and validators.</p><p>With each charts library, I also write a <strong>data generator</strong>. It can be used as dummy data for developing the charts, but also for testing various cases, for smoke testing, etc. Datahub-ui has generateTimestamps and generateTimeSeries for example, that are made from lower-level data type generators.</p><h3>Template Loading</h3><p>The next module you see in this example pipeline is the template. One trick that I often do when I start a new chart is to implement it in <strong>plain html/svg/css</strong> first, without touching a single line of D3. Then I extract the base template from this prototype so I don’t have to generate it with D3. Since this base template is static, I just load it once and apply it to the DOM on init. Then I replace all other pieces of my prototype by D3 code that generates it.</p><p>Here is part of my template module code:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/86ad8484d05319919775cfb5cd166fce/href">https://medium.com/media/86ad8484d05319919775cfb5cd166fce/href</a></iframe><p>Yes, I know that html in a variable is ugly. Maybe loading the template from an external file would make it a little cleaner. But even there, I realized that I have less trouble reading this than the equivalent description in D3, especially when I need multiple levels of nesting.</p><h3>Generic vs Multi-specific</h3><p>In this pipeline, you can see some items coming from the “common” namespace. That’s where I place items that are easy to <strong>reuse</strong> across multiple charts.</p><p>When I start a new chart, I usually copy one of my chart file and just <strong>replace the modules</strong> I need to change. I can for example replace an ordinal scale by a time scale, adding a data validation layer, changing some default config. When I’m tempted to add a conditional switch somewhere in my code, I replace the whole module by a new one instead. After a couple of years of working like this, adding modules instead of overloading an existing one, I’m still surprised by the benefits. It was way more complicated before, when I was trying to be generic, adding conditionals everywhere, multiplying config options and fixing all side-effects every time I introduced a small changes.</p><p>I really like the freedom to make a new pipeline, replace some modules, just make my specific chart work the way I want, then take the time to unify some modules if I need too. That’s what I call <strong>multi-specific</strong> instead of generic.</p><h3>Modular File Structure</h3><p>The pipeline pattern is not the only example of modularity in the Datahub-ui code. We also need a <strong>modular file structure</strong>. Here is an example of a boilerplate I often use:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/29da9b43b728d02060370495b6592a82/href">https://medium.com/media/29da9b43b728d02060370495b6592a82/href</a></iframe><p>Using this wrapper, I can load files one by one with html script tags or I can use <strong>npm</strong> to distribute it as a bundle.</p><p><a href="https://www.npmjs.com/package/datahub-ui">datahub-ui</a></p><p>Splitting the code in multiple files, into namespaces, into functions that are assembled to form higher-level functions are just some examples of modular patterns I like to use. I also <strong>split my css files</strong> in at least two types. I keep a separation between the css describing the layout, all the features that are needed for the charts to work, and the theme, all colours and styling that you would want to eventually override. Even the semantic colouring of internal elements of the charts, like the color of each line of a line chart, which are typically set in D3 with inline styles from accessor functions, is left to the css.</p><h3>External API</h3><p>Whatever strategy is used for keeping the code modular, it also needs to expose a consistent API to the outside world. There’s a difference between the internal API, for example this pipeline pattern, and the external one. I like to design simple <strong>external APIs</strong> that doesn’t need much on initialization, usually just the DOM container. The options could be passed on initialization, but can also be set after.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/b220ae12a18fe8c909502af1a5cb7767/href">https://medium.com/media/b220ae12a18fe8c909502af1a5cb7767/href</a></iframe><p>One feature that can be challenging to get right is <strong>exposing events</strong>. I use <a href="https://github.com/d3/d3-dispatch">d3-dispatch</a> internally with a simple way to expose it to an .on() method for listening to events.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/c1d58c3218adf78110e4b744eb364b7d/href">https://medium.com/media/c1d58c3218adf78110e4b744eb364b7d/href</a></iframe><p>In the example pipeline, there’s an “eventsPanel” module that is in fact a rectangle on top of all other elements that catches all user events and collect the appropriate information to return to the event listener. For example, I don’t rely on SVG or DOM events for detecting mouseover. I grab the mouse position and look-up what’s under the cursor directly from the data space. That trick works well for example when using canvas, or when you have a chart that combines a line and a stack of bars. The <strong>eventsPanel</strong> will grab all data available at the closest timestamp in a consistent way.</p><h3>Testing</h3><p>Both the internal modules and external APIs could be unit tested. I only wrote <a href="https://planet-os.github.io/ui-components/datahub-ui/test/">external API tests</a> for Datahub-ui, because my main concern was to make sure the chart usage is clear and exposes a consistent contract with the outside world. I wrote a bit about unit testing D3 charts in a small book: <a href="http://backstopmedia.booktype.pro/developing-a-d3js-edge/3b-the-reusable-api/">Developing a D3.js Edge</a>. But I also often use another form of testing: every time I develop a new feature, I add it to a giant <a href="https://planet-os.github.io/ui-components/datahub-ui/example/index.html"><strong>visual test sheet</strong></a>. I can use it as a demo of all the features or to tweak the style with the designer, but it also makes it easy to visually find bugs that are otherwise hard to catch with unit tests.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cELkTBjJOoo4USXKLsUcuA.png" /><figcaption>Datahub-ui visual test sheet</figcaption></figure><h3>Conclusion</h3><p>Datahub-ui is open source, but we didn’t design it as a generic toolkit for everyone to use. It has all the features we need to connect to Datahub data, quickly build Powerboards, and in a close future, it could become a self-serve tool to build one yourself. But we want to share it early, even as an internal tool, to show how using simple patterns is a key to quickly <strong>deliver products</strong> at Planet OS. I invite you to look at what we are doing with <a href="http://data.planetos.com/">Datahub</a> and <a href="https://planetos.com/powerboard/">Powerboard</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d8f0f5da0b82" width="1" height="1" alt=""><hr><p><a href="https://medium.com/planet-os/insights-from-designing-datahub-ui-in-d3-js-d8f0f5da0b82">Insights from designing Datahub-ui in D3.js</a> was originally published in <a href="https://medium.com/planet-os">Planet OS (by Intertrust)</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[D3 is now modular]]></title>
            <link>https://medium.com/@christopheviau/d3-js-modularity-d5eed78ba06e?source=rss-a1206bba9736------2</link>
            <guid isPermaLink="false">https://medium.com/p/d5eed78ba06e</guid>
            <category><![CDATA[3d]]></category>
            <category><![CDATA[components]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[charts]]></category>
            <dc:creator><![CDATA[chris viau]]></dc:creator>
            <pubDate>Wed, 15 Feb 2017 14:41:31 GMT</pubDate>
            <atom:updated>2017-02-15T17:42:56.765Z</atom:updated>
            <content:encoded><![CDATA[<h3>D3 is Now Modular</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RMDQcFWBux1Gw_hoABizSg.png" /></figure><p>“D3 is now modular”. That’s how the D3.js v4.0.0 <a href="https://github.com/d3/d3/releases/tag/v4.0.0">release note</a> starts. What does it mean? Why should I care? It’s important as a community that we discuss it so we can make the best out of it. So at the <a href="https://medium.com/@enjalot/d3-unconf-2016-recap-b3af91a65f00#.8e3xk7oms"><strong>D3 Unconf 2016</strong></a>, one <a href="http://[https://docs.google.com/document/d/1X9ceTjvZjQmVrp51aABvbPBvApE9cBuJoeHKJP-dSkY/edit]">session</a> was about this concept of modularity. Here’s my summary.</p><p>D3 v4 is a collection of modules you can use independently, with minimal dependency between them, all neatly isolated in their own <a href="https://github.com/d3">repo</a>. It’s a good illustration of one definition of modularity: having code as a collection of <strong>minimal functional units</strong>, tightly encapsulated and exposing a clearly defined API.</p><p>One advantage of this new modular approach is to encourage developing and owning more granular modules. The community was already providing a lot of <a href="https://github.com/wbkd/awesome-d3">charts libraries</a>, some d3 layouts and <a href="https://github.com/d3/d3/wiki/Plugins">plugins</a>, some code structures like <a href="https://github.com/misoproject/d3.chart">d3.chart</a> and <a href="http://kotojs.org/">Koto</a>, some higher-level abstractions like <a href="http://visible.io/">D4</a>, but not much core or lower-level features. The closest we have from core feature extensions are some pretty cool monkey-patch and extensions like <a href="https://github.com/1wheel/d3-jetpack-module">d3-jetpack</a> and <a href="https://github.com/twitter/d3kit">D3 Kit</a>. But now, contributing to <strong>core features</strong> or any other abstraction levels just got easier.</p><h3>Bundle</h3><p>When a javascript developer talks about modularity, he often refers to the mechanism of wrapping and <a href="https://addyosmani.com/writing-modular-js/">loading modules</a>, like using AMD, CommonJS, ES6 imports, etc. Indeed, one characteristic of modules is how they can be bundled together. D3 modules can be loaded independently from each Github repos or from <a href="https://www.npmjs.com/search?q=d3&amp;page=1&amp;ranking=optimal">npmjs</a>. Or the <a href="https://d3js.org/d3.v4.min.js">official bundle</a> can be loaded at once. But you can also build your own version. Why would you want to do that? Maybe you want to add some modules that are not in the official package, like the useful <a href="https://github.com/d3/d3-selection-multi">d3-selection-multi</a>. Maybe you want a smaller package only containing what you actually use. A good example of a <strong>custom package</strong> would be to bundle all modules that are independent from the DOM, to use on servers without using a dom implementation, on <a href="https://bl.ocks.org/mbostock/01ab2e85e8727d6529d20391c0fd9a16">Web Workers</a> or even with <a href="http://www.macwright.org/2016/10/11/d3-and-react.html">React</a> without interfering with its virtual DOM lifecycle. D3 v4 uses <a href="http://rollupjs.org/">Rollup</a> to bundle using <a href="https://github.com/rollup/rollup/wiki/ES6-modules">ES6 modules</a> syntax. Here’s a <a href="https://bl.ocks.org/mbostock/bb09af4c39c79cffcde4">simple example</a> of how to do it and a <a href="https://medium.com/@richavyas/d3-js-the-custom-modular-bundle-now-bebd6f25bc8b#.b9bnmrj3s">more complex</a> one. I especially like this tutorial about how to <a href="https://537.io/a-simple-d3-plugin/">make a plugin</a>, and of course the <a href="https://bost.ocks.org/mike/d3-plugin/">official one</a> from Mike Bostock.</p><p>I’m not a big fan of scaffolding: compilers/transpilers, code loaders/bundlers, etc. I often use a simple and well documented <a href="http://www.adequatelygood.com/JavaScript-Module-Pattern-In-Depth.html">module pattern</a> that I simply concatenate/minify on deployment for production use. But whatever technology you use, IMO the power of modularity is not about loading and bundling, but about encapsulating in a way that small modules can form bigger modules on another level of functional abstraction. For example, D3 is a “<a href="http://vis.stanford.edu/files/2011-D3-InfoVis.pdf">visualization kernel</a>”, a very thin layer on top of the DOM API and other web technologies. Then the community is building <a href="http://blockbuilder.org/search">tons</a> of <a href="http://christopheviau.com/d3list/gallery.html">examples</a> with it. But between low-level D3 modules documentation and standalone examples, there’s a whole range of modularization that is mostly missing. Let’s look at these <strong>levels of encapsulation</strong> that are harder to find example of: components, charts, libraries, views, apps.</p><h3>Components</h3><p>A common pattern to encapsulate charts is the <a href="https://bost.ocks.org/mike/chart/">Reusable Pattern</a>, which is nothing more than a “closures with getter-setter methods”. NVD3 is one of the most popular charts library using it. But IMO the best example of this reusable pattern is the <a href="https://github.com/d3/d3-axis/blob/master/src/axis.js">d3-axis</a>. I described this pattern, how to write unit tests for it and how to use it in a real-world app in a short book we wrote a while ago: <a href="http://backstopmedia.booktype.pro/developing-a-d3js-edge/3b-the-reusable-api/">Developing a D3.js Edge</a>. Interesting modifications to this pattern <a href="https://www.toptal.com/d3-js/towards-reusable-d3-js-charts">where</a> <a href="https://bocoup.com/blog/reusability-with-d3">proposed</a> . I recently wrote a minimalistic example of a simpler pattern I use for encapsulating charts and widgets, in that case a simple <a href="https://github.com/biovisualize/d3.todo">todo app</a> inspired by TodoMVC. But other <strong>patterns</strong> can work well too, like traditional <a href="https://ejb.github.io/2016/05/23/a-better-way-to-structure-d3-code.html">object-oriented</a> code used by one of the oldest D3 charts library around called <a href="http://code.shutterstock.com/rickshaw/">Rickshaw</a>.</p><p>Many patterns can be used, but what I want to emphasize is how encapsulation can happen at the “component” level. I define a component as any reusable <strong>building block</strong> to build charts. An axis is a good place to start, as it sometimes is the most complex part of the chart. For example, when I designed <a href="http://micropolar.org/">Micropolar</a> for <a href="https://plot.ly/javascript/">Plotly</a>, I realized that most of my code was describing a very configurable polar axis, the rest being simply graphical marks and helpers around it. <a href="http://plottablejs.org/components/">Plottable.js</a> is working at this type of component level. <a href="http://visible.io/">D4</a> proposes an interesting way to mix in and out components and features. Interstingly, some of these patterns look a lot like <a href="http://mbostock.github.io/protovis/">Protovis</a>, the project from which D3 derived.</p><p>During my PhD, I studied ways to think all visualizations as assemblies of components. Someday I should make something out of all the material I have about the design space of hybrid visualizations. Here’s <a href="http://profs.etsmtl.ca/mmcguffin/research/2012-viau-connectedCharts/viau-eurovis2012-connectedCharts.pdf">one paper</a> we published at Eurovis about chart components.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kWh9iOEI48C3F3dtusuFPQ.png" /><figcaption>ConnectedCharts, Eurovis 2012</figcaption></figure><p>When working on <a href="https://github.com/boundary/firespray">Firespray</a> for Boundary, and then on <a href="https://planetos.com/blog/cirrus-js/">Cirrus.js</a> for Planet OS, I experimented with a nice <strong>pipeline pattern</strong> that I still use today to encapsulate low-level components and pipe them together to describe charts. I used it in multiple contexts, I even tried a version using <a href="https://github.com/biovisualize/piper.js/tree/rollup">Rollup</a>. I will describe how I solve various challenges using this pipeline pattern in a follow-up blog post.</p><h3>Charts, Libraries, Views, Apps</h3><p>Various patterns can be used to describe how the components will be assemble to form a chart, from factories to mixins to inheritance. Sometimes the assembly involves a more general framework, like <a href="http://www.reactd3.org/">React</a>, <a href="https://github.com/AdRoll/backbone.d3">Backbone</a> or <a href="https://krispo.github.io/angular-nvd3/#/">Angular</a>. I prefer to keep my charts code in plain javascript and D3, then wrap it with React for example. But that’s personal preference. What is important here is to emphasize that there are multiple levels of encapsulation to take into accounts: toolkits, components, charts, chart libraries, UIs (e.g.: dashboards), applications as a collection of UIs, software ecosystem, etc. At each level, there are choices to make that depends on a lot of different factors. Data piping, events passing, error handling, state managing, are all bringing different constraints to the architectural choices. But at every level, the only feature I always keep in mind is modularity, encapsulating functional blocks that can be assembled to form <strong>larger functional blocks</strong>, until I reach the final functional block: the actual end-user product.</p><p>In conclusion, I’m glad the community is sharing more patterns, code architecture and modules at various levels, adding diversity to the already awesome collection of examples, chart libraries and plugins ecosystem.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d5eed78ba06e" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>