All Posts

Data Visualization Video Inspiration + Learning

This post was prompted by Andy Kirk’s call for sources of video inspiration

One exciting yet daunting feature of a broad and rapidly changing field like interactive data visualization is that it can be difficult to find the frontier. If you don’t know which way is up, it is hard to advance. In my pursuit of the horizon, I watch videos regularly. As I progress as a data visualization engineer, the challenge shifts from “how do I build this idea I have” to “what things should I try to build”? In order to avoid getting stuck in a local optima of familiarity, I try to reflect on new ideas regularly.

This post is a compilation of video sources I find useful.

Applied Conferences

  • OpenVis by Bocoup: Each year, the team produced a custom webpage (and sometimes a dataviz about the conference itself!) about the presentations. (I’ve made some notes about what was distinctive each year.)
    • 2013 - Playlist only. However, OpenNew’s Erin Kissane wrote some excellent dispatches and an overall recap for the event.
    • 2014 - Talks segmented by dominant color
    • 2015 - Talks stamped with visual encodings whenever visualizations or specific words occur. Segments are sized by average word length or speech density within that segment. Some talks have direct slide links.
    • 2016 - Filmstrip visualizations accompanied by bigrams of common phrases, extracted using TF-IDF. Bigrams can be used to visually annotate where in the video they occurred. Many talks have direct slide links too. On a personal note, Mariko Kosaka’s How Computers Draw Pixels was the first time I heard about Openvis, courtesy of the inaugural D3 Bay Area Meetup Movie Night - it set the bar very high!
    • 2017 - Speakers linked in a network diagram based on similarity of their transcripts. See Ben Garvey’s post for recaps.
    • 2018 - No “official” dataviz this year, but others have taken a pass at it. One learning exercise could be to get transcripts from this year, and apply the code from a previous year’s interactive playlist to the new year’s content instead. Full playlist is here.
  • Eyeo: 2011 - 2019: Extends into many other tech/culture/visualization adjacent fields beyond dataviz
  • Tapestry (Tableau): 2013-2018
  • PlotCon: 2016 NYC / 2017 Oakland.
  • d3.unconf / visfest unconf. Many of the workshops were not recorded, but keynotes and highlights from 2014 - 2019 are available. Kudos to the event organizers for making a concerted point of making a point to seek out world class female practioners every year.

Academic Conferences

The line between academia and applied can be blurry as academics also apply their work, and people in industry may be involved with academic research. I separated this section because these conferences primarily feature papers to be published in a journal.

  • IEEE Vis - the premiere academic data visualization conference. In 2020, registration was free as the event was online! Some of the workshops are publicly recorded online, and many of the papers have 30 second “teasers”. I found Robert Kosara’s guide useful as I had not previously participated in this conference.
  • ACM SIGCHI - Youtube Channel - One of the main Computer Science special interest groups adjacent to data visualization.

Other fields

  • Data Journalism: Not everything is public, but NICAR and Malofiej are good starting points.
  • Visual explainers on Youtube: (3blue1brown, Primer, Vox Earworm, etc) - For ideas about how to weave data into a narrative and present data to broad audiences in an engaging way.
  • Data Science: I attended ODSC East in 2015 and 2018. I don’t refer back to recordings often, but Randy Olsen’s AutoML (Automated Machine Learning) overview stood out.
  • Game Design Confs / Teardown Channels (GDC, GMTK, Architect of Games) - Similar motivation/rational as Youtube visual explainers.

Livestreams

It’s valuable to watching people build in a format less rehearsed than a formal presentation, as you get to see how experienced practioners work their way through sticking points and consider multiple options. See Cedric Chin’s post for more on acquiring tacit knowledge by observing experts.

  • Ian Johnson wrote a bit about streaming vs meetups here.
  • Swizec Teller: Frontend, data visualization, serverless
  • Shirley Wu: Data visualization, data art, react
  • Keyframers: David Khourshid and Stephen Shaw take mockups from Behance, Dribbble, etc, and implement them in JS (or sometimes pure CSS).
  • The Coding Train: The legendary Daniel Shiffman’s channel. He posts both raw footage and edited clips. Code samples and community submitted implementations of his coding challenges are on this website and on github.
  • The Pudding - Patreon patrons get behind-the-scene videos about how their articles are made.
  • Time lapses of visual artists painting/sketching, both those drawing from reference and from imagination (the processes are slightly different, but I think the process of making multiple passes and focusing on different levels of abstraction/detail has indirect parallels with the software writing process).

Podcasts

  • Data Viz Today - Hosted by Alli Torban, with well-edited show notes and actionable takeaways at the end of every episode. Good for bite sized listening.
  • Data Stories - hosted by Enrico Bertini. Longer form, episodes around 40-60 minutes. Note the recent recap of IEEE Vis 2020 was an in depth supplement to reading IEEE vis recaps from publications like Multiple Views.

Upshot: Beyond Watching

Just hearing about techniques isn’t enough to change your work or improve your craft. I try to improve my craft by producing an artifact after every video. Depending on time and interest, this may be a short pro/con tradeoffs sheet for my notebook, a discussion with a fellow practitioner, or a demo in Storybook / ObservableHQ notebooks.

I’d be glad to hear from you if you’d like to share a way you internalize learnings from watching videos, or have another resource to recommend.