top of page
  • Writer's pictureT. @ Data Rocks

Do We Take Data Visualisation Too Seriously? Featuring Questions in Dataviz by Neil Richards.



If you are reading this, I take you are, at the very least, curious about data and how we communicate it. You may be a seasoned data visualiser, or maybe you’re just starting out in the field or someone trying to figure out what this is all about. Whatever brought you here, I can probably make a bold assumption that you may have a question about dataviz. Or multiple. Why do we visualise data? How do I use colours in dataviz? Which types of shapes should I use to visualise data? Is there a right type of chart for certain types of data? Why does everyone seem to hate pie charts? The list goes on. There are probably as many questions to ask about dataviz as there are data points in the world.


Data Visualisation is a field built on curiosity. We ask, therefore we viz. I have talked about the importance of asking good questions before. It is also an uncomfortable feeling for many: not all of these questions have definitive answers. Most of the time, there will be multiple correct answers, or the conclusion will be the dreaded “it depends”. Other times there will be questions that lead to more questions and no answer at all.


One fellow curious mind is Neil Richards. Coming from a background in Mathematics and having worked in BI roles, he decided to embark on a dataviz learning journey, recording his questions and sharing his findings in his blog, which turned into today’s book recommendation - the aptly named Questions in Dataviz - A Design-driven process for Data Visualisation.



Community challenges


Data visualisation has a few guideline principles that most practitioners are aware of. We call them best practices. They usually are postulates that come from learnings shared by fellow visualisers along the way - such as don’t use unnecessary embellishments in your viz, always start your Y-axis at zero, use colour carefully to draw attention, place items on the screen with hierarchy principles in mind and many others.


But if there’s one truth about all those principles is that they can and will be broken at some point. If you’re starting out, they can be useful to help you get the hang of what to do. They offer direction and guardrails. But once you get the basics and start advancing into communicating data visually, you’ll begin to wonder: is it ever ok to flex or ignore these principles? Will I open some chaos chasm at the core of the Earth if - god forbid! - I decide that I want to add an illustration that doesn’t have much of a purpose to my viz because it looks cool?


If you’ve ever tried posting a more creative piece of data visualisation work to social media, you’ll know what I mean. There is a portion of the data visualisation community that will fiercely defend that everything should have been a bar chart. That’s partly why I believe Questions in Dataviz is a very welcome addition to the field’s literature - it questions not only the principles we hold in high praise but the attitude we have towards them as information designers in a candid and open-minded way.


Data Rocks started in October 2019. In early 2020 I joined Twitter and discovered the amazing community challenges of the Tableau #datafam. I then decided to join Makeover Mondays. Initially, my idea was to join a whole year of challenges, which would help me hone my Tableau skills while building an online portfolio in Tableau Public. A great idea, and I wish I had stuck with it for a while, but I didn’t. The main reason was the disconnect I felt between my desire to try new things - different from the kinds of visualisations I would do for my clients - and the community's expectation - which seemed to be more of what I did for clients.


I confess I have low social energy and bandwidth to deal with the unsolicited feedback that my tentative of a novel chart type would be better using a more traditional approach. So, with the pandemic, lockdowns and my luck of working with a Health Insurance company during that time, I decided to focus on the tasks I had in my job and in dealing with all the challenges of transitioning into entrepreneurship.


One of the greatest things about putting yourself out there, though, despite the naysayers, is that it opens doors you wouldn’t otherwise believe to be there. Because of my brief posting history of Makeover Mondays, I ended up helping Alex Waleczek run the Auckland Tableau User Group from 2020 to 2022.


I had never seen myself as a community leader. I love Tableau as a tool and have been using it since 2016. But to be invited to help with the local TUGs? It was one of the most awesome moments of my career. I helped run the TUG when they were happening mostly virtually over the pandemic years - we also ran some New Zealand-wide TUGs. Bringing people together around dataviz is always a fun thing! But life sometimes throws us unexpected events, and I had to move to a much smaller city. My tenure as Auckland Tableau User Group’s co-leader came to an end late last year.


This is an opportunity I wouldn’t have had if I didn’t put myself out there, albeit briefly. If you’re starting out now, joining community initiatives will not only help you improve your technology skills but will also help you get seen, and new opportunities may arise from that. And don’t fret if the feedback of what you’re sharing is negative, despite your best efforts. Learn from it, and move on. The only person who needs to give you permission to try new things is yourself.


Neil Richards makes a similar case with his book. It was written initially as a series of blog posts based on his experience sharing personal projects online, joining challenges and engaging with the dataviz community, primarily on Twitter and the Data Visualization Society’s Slack.


Each chapter is titled as a question, and a few of the questions come from a place of arguing and justifying his choices for experimentation. The overall arching theme is about questioning the establishment and the rules we come up with or adopt in dataviz. Neil often concludes that as long as you can justify your design choices and they serve your purpose well, then it’s ok to bend the rules.



Do we take data visualisation too seriously?


This is the title of chapter 2.6 of Questions in Dataviz. The book is divided into 3 sections, where Neil Richards embarks on his quest for answers using his own personal experiences and works as a reference.


  • Section I: First Questions

Here, the author brings up basic questions that everyone must have wondered about at some point when dealing with data visualisation. My personal favourite is Chapter 1.6 - Is white space always your friend?. We’re invited to challenge the assumptions behind the term chartjunk, coined originally by Edward Tufte. If you’re like me, someone who defends that minimalism is a stylistic choice and shouldn’t always be the norm, this chapter is a treat!


  • Section II: Challenging Questions

The core of the book dives a little bit deeper into the practice of design-driven data visualisation - that is, the type of visualisation work where the idea for the design comes before the data - something almost sacrilegious when we’re talking about business dataviz. The author begins this section by wondering why do we visualise data? And then, he wonders further into amusingly specific design choices such as Chapter 2.3 - Does it matter if shapes overlap? Or Chapter 2.2 - Why do we visualise using triangles? My personal favourite, though, is his take on Chapter 2.4 - What is Data Humanism. If you’ve followed some of my previous blog posts, you’ll recognise the mentions of some of his references in this chapter - such as Giorgia Lupi, Nadieh Bremer and Shirley Wu.


  • Section III: Idea Questions.

This is where the book dives even deeper into design-driven visualisations and where we can see the author’s style surfacing. Chapter 3.3 - Why do I use flowers to visualise data? is a brilliant example. And if you’ve ever visited my website’s About Me page, it will come as no surprise that my favourite chapter here is Chapter 3.4 - What are Data Portraits? Honourable mention goes for the opening chapter in this section, which discusses what is the third wave of data visualisation, or in other words: what will be the next big trend in the field?



Without spoiling it too much, many of these questions have no definite answer, and that’s exactly the point - this is a unique book in the sense that its purpose is to ask. It also reinforces the concept that we don’t always must have all answers. It is alright to embrace a little bit of uncertainty. dataviz is an ever-evolving field, and what we may hold true right now may turn out to be different in a few years' time. My main takeaway from the book is this: whatever the answers you may find, just make sure you have fun in the process.





Should you read it?


This book is a new-ish addition to the Data Viz Bookshelf; I got it as a gift for having helped run the Auckland TUGs - thanks, Alex! I love it!


I have reviewed other books with big question marks on the cover before - but although they are also about asking questions, they weren’t about dataviz. Questions in Dataviz: A Design-Driven Process for Data Visualisation goes deeper into the practical and more philosophical thoughts that pop into our minds when working in the field.


If you’re new to dataviz, this will be a good reference book to go back to as you move further in your journey whenever you’re caught up wondering something, and you’re not quite sure if someone else has also wondered that.


If you’re more seasoned, you’ll still love it for the fact that it is very much a historical record of what our field looks like today. Many of these questions exist in our current context, challenging the way we used to think about the field until very recently - it will be awesome to go back to it in 20 years and see if we’re still asking the same things and what will have changed.


I also must say that I had a lot of fun reading it. The chapters are bite-sized, engaging and conversational. It felt like I was discussing my favourite subject with a friend. It is very high on my recommendations list.


Always check your local library first to see if any of the books I recommend are available. If they’re not, consider donating a copy!

 

Get a copy at your local library | Amazon


 

If you subscribe to my monthly Newsletter, you’ll get a summary of all recommendations, plus more of my data viz musings.


You can also follow Data Rocks on LinkedIn or read this and other articles on Medium.

 


コメント


bottom of page