SHOWCASE
SurĽoh
Google News Lab is the biggest project Bremer has worked on since starting to freelance. The article goes
into the most popular questions that people have asked on Google to better understand their cats and dogsI’d also always had. So from that
conference onwards, I started to
dedicate all of my spare time to learning
more about data viz: reading books to
learn about the best practices as well as
doing personal projects to get better
with d3.js.
I also told my boss that I wanted to
change from being a data scientist to a
data-viz designer. Thankfully he was
very understanding and helped wherever
he could, in terms of including me on
projects. And so, slowly, I started to
become a data-visualisation specialist.You went freelance in 2017. How has your
career developed since then?
Going into freelancing is one of the best
professional steps that I’ve ever taken;
I’ve not regretted it for a single day. I’ve
had the opportunity to work with clients
I thought I could only dream about, such
as Google, UNESCO and the New York
Times. I’m also enjoying the wide
breadth of clients and datasets I now get
to work with, from small startups to big
names and from magazines or non-
profits to corporates.
I generally try to have three to four
clients at the same time, so that I can
switch between them throughout the
week. I might be working on someI’ve been freelancing under the name
of Visual Cinnamon for almost three
years. I work mostly on pieces that are
used for marketing – either internally or
externally – used for articles (in print or
online), press releases or maybe even
some data art for in the office.
In my free time I love to travel and see
the world together with my boyfriend –
whom I’ve been with since I was 17 – and
read fantasy books, especially those by
Brandon Sanderson. I sometimes have a
manga phase again too, playing handheld
Zelda games and watching blockbuster
movies all cuddled up on our poofy couch.How was it that you originally got into
data visualisation?
My story starts with astronomy. I’ve
always loved the topic and studied it full-
time for five years at university. But I also
knew the next academic step wouldn’t
make me happy. So after I graduated in
2011, I decided to join the new analytics
department of Deloitte Consulting.
The word ‘data scientist’ hadn’t even
reached the Netherlands then, so I was a
‘consultant advanced analytics’. I was
doing analyses on people and companies
instead of stars. Different but still fun
and a lot more diverse actually. Yet after
about three years, I started to lose my
passion for data analysis. I didn’t enjoy
spending that extra hour making my
predictive model just a tiny bit better.
During a data-science conference in
2014, I saw the speaker Mike Freeman
calling himself a ‘data-visualisation
specialist’ on the first slide. And then it
hit me like lightning: ‘You can do data
viz as a separate thing?!’
I immediately knew that data viz was
where my passion lay. I did enjoy
spending extra hours after work trying to
make that chart look more professional
for the presentation the next day. I was
always trying to find ways to incorporate
d3.js-based interactive charts into the
projects. In retrospect, it’s strange I
didn’t realise it sooner.So what was it about data viz that struck
such a chord?
It combined the math/data/exact side
that I enjoyed with the creative side thatscience-related research, cultural
heritage and what people search online
for about cats and dogs, all in the same
week. I’m still keeping my fingers
crossed that maybe the WWF, Disney or
NASA/ESA will come knocking on my
door at some point.What are your go-to tools?
I use R to do my data preparation and
analysis at the start of a project. This
helps me build a mental model of what
the data is about, where the interesting
aspects to visualise are and how to
restructure the data in a way that I need
for the visual.
I design using either plain pen and
paper or with the Tayasui Sketches app
on my iPad Pro and Apple Pen, so the
designs really stay rough. From there, I
then go to Visual Studio Code to
program my data visualisation with
d3.js and JavaScript.
Depending on the amount of data to
show or the complexity of interactions,
I’ll create it with SVG for simpler things
or HTML5 Canvas for larger visuals. On
some rare occasions I might also create
the visual with three.js if the data is
really big.
Finally, if the finished piece is going to
be a static image, I’ll export the visual