Alter Everything

A podcast about data science and analytics culture.
Alteryx Community Team
Alteryx Community Team

Nadieh Bremer dials in from The Netherlands for a chat about her path to becoming a well known data visualization artist, and where she gets her inspiration.





Sydney Firmin - @SydneyF, LinkedIn
Nadieh Bremer - LinkedIn, Twitter, Visual Cinnamon


Community Picks






Episode Transcription

SYDNEY: 00:13 

[music] Welcome to Alter Everything, a podcast about data science and analytics culture. I'm Sydney Firmin, and I'll be your host. Today, Nadieh Bremer dials in from the Netherlands for a chat about her path to becoming a well-known data visualization artist and where she gets her inspiration. Let's get started. [music] Hi, Nadieh. Welcome to Alter Everything. 

NADIEH: 00:48 

Thanks. Thanks for inviting me. 

SYDNEY: 00:50 

We're so excited to have you here today. To start us off, could you please do the obligatory introduction of yourself to our audience and just tell us a little bit about your background? 

NADIEH: 01:02 

Yeah, sure. So I graduated as an astronomer, but I knew that I didn't want to continue in academia. I love doing the research, but I really, really did not like writing the papers. So I knew I wanted to do something a little bit more, I guess, also tangible and related to life as I noticed it around me. So I came into consulting as a data scientist. I started working at Deloitte in the newly formed analytics team. And I really enjoyed that, having a different client, a different data set, a different problem every few weeks or every few months, working [highly?] together with other people in my team. And because we were external, we often had to create slide decks with presentations about the things we'd found in their data sets and from the analyses. And these often contained some form of data visualization. It's not that much fun to actually show tables, so data visualization is kind of how I rolled into data viz there. And as the algorithms that I was doing were getting more complex, I guess, the visualizations that I found that would best show the results were also getting more complex, such as using networks to visualize results of an association analysis. 

NADIEH: 02:26 

And as I was doing that - and then we're jumping ahead like four years - I realized at some point that, "Wait. I actually enjoyed visualizing the data much more than still doing the analysis." So I didn't get the same amount of joy out of making my predictive model even better than I would get in making my data visualization look even better. So at that point when I realized that, I kind of consciously switched to wanting to specialize in data visualization. So I started reading all kinds of books, teaching myself new languages, like programming languages to do better data viz, and through that, eventually ended up at another company, where I was doing their data visualization dashboards based in D3. We'll get back to that probably later, but it's the program that I-- well, the language that I use. And after about a year of doing and creating dashboards, I figured out that that's not quite what I enjoyed doing as well. I was looking for a more creative approach to data visualization, I think. And then after a long thinking and having a really good first possible client, I took the leap and went into freelancing. So that's what I've been doing for the last two and a half years, freelancing as a data visualization designer. 

SYDNEY: 03:49 

And you love it? 

NADIEH: 03:51 


SYDNEY: 03:53 

Ah, good, good. 

NADIEH: 03:53 

[crosstalk] [a lot?]. 

SYDNEY: 03:55 

So if you're interested in knowing more about Nadieh's journey, there is a blog she wrote called My Journey Into Data Visualization, and it's really well written, and it's a lot about that story. And I was actually reading it, and I got the impression that R was your first love as far as programming languages go, and then you moved over to D3. Do you still use R, and what was that transition like? 

NADIEH: 04:23 

Yeah. So you're absolutely right. R was the first language after some arcane astronomy language that nobody uses outside of astronomy. In my first year as a data scientist, I remember counting at the time six new languages, from SAS to VBA to R. And R was the one that stuck with me as being the one I could most comfortably actually do the things that I wanted to do, analyze the data, and that makes visualizations using ggplot. And I still use it during basically every project that I do these days, although I'm not involved in data analysis or data cleaning. When I get the data from my clients, I always do some simple statistics and make lots of very simple bar charts and line charts, scatterplots, just to get a sense of the data that I've gotten from the client, and also to prepare my data into the format that I will need to make the visualization, which sometimes means making aggregations or structuring it in a different manner. So R is always part of my tool kit. 

NADIEH: 05:33 

And yeah, then I discovered the existence of D3 during a conference, and I was immediately-- I straight out fell in love with D3 with the way that you could do interactions and animations and make things move, which at the time, I think Shiny wasn't actually in R then, so you couldn't really do anything interactive in an easy way. So I have to admit, that was quite a transition because besides having to learn D3, it's based in JavaScript, HTML, and CSS, so it's like the web languages. So I had to learn all of those simultaneously while trying to create these visualizations with D3. And after getting back from that conference and being blown away by D3, it took me about a week to get a fully functional, built-from-scratch scatterplot working in D3 [laughter]. So it wasn't easy, but I was so enthusiastic and motivated by the possibilities that I could do with D3, that it really just kept me going. And there was so many examples available, as well, to easily get you started that it was always fun to just start with somebody else's example and then try and plug your own data set in there and see how things would change and how you would update that based on your data. So it was slow going, but it was fun. And it did take me, I think, about a year before I was good enough in D3 to actually start using it in projects at my work. 

SYDNEY: 07:06 

That's awesome. So most of the coding languages, it sounds like you've kind of had to teach yourself in a project-based way? 

NADIEH: 07:14 


SYDNEY: 07:14 

And another thing I noticed on your blog is that you've written a fair number of tutorials for D3. And I wanted to ask, how is teaching about and speaking about your work and the technical side of it helped improve your projects? 

NADIEH: 07:29 

So yeah, the tutorials, I started writing them basically because-- so they're not like beginner tutorials in D3, but very specific elements that I would have figured out during a project that I was doing. And there wasn't any blog available that could do it, such as making text appear on arcs or having a glow around something. So I wrote that up basically so I wouldn't forget it myself [laughter]. There are a few blog posts that I keep going back to every time because I just cannot remember the full thing that I have to do. So in that sense, it's helped me. Some of these things, they stick better if I've actually written a blog post about it and-- 

SYDNEY: 08:17 

The best way to learn is teach. 

NADIEH: 08:18 

Yes, exactly, exactly. So it's also an easy way for me to get back to code snippets that I can use on projects in the future. And yeah, I started speaking - but that was a bit later, a few years later - at conferences about D3-related things, data visualization. And I guess what I like about speaking is the-- somebody once told me this, and I thought it was great. It's like the best way in which an introvert can make connections at a conference, because you have like 40 minutes, and then people have to basically sit there and listen to you speak about something you like [laughter]. And afterwards, people come up to you then to talk about the thing that you like. So it's a good way to make new contacts, I find. And I really enjoy and have learned a lot about just general conversations that I've had with people. Although maybe direct client work hasn't come out of it as much as I'd maybe hoped, but making new contacts in the field or getting inspired by the ideas that people had or working on something, that's been a great help. 

SYDNEY: 09:29 

That's awesome. I'll have to keep that in mind. Speaking is a little intimidating for me, so that's a really good way to think about it. 

NADIEH: 09:38 

It was for me, too, by the way. So I get really nervous right before a talk. Yeah. But yeah, I do feel the rush of adrenaline and the joy of having finished the talk, and that keeps me going every time [laughter]. 

SYDNEY: 09:55 

"I'll do it again. It went okay this time." So you mentioned a little bit about inspiration. Your work is very striking, and it's also really whimsical. And we were wondering, where do you draw your inspiration from for a lot of your data visualization projects, and what is your creative process like? 

NADIEH: 10:16 

So I get inspiration from lots of different things. But what I try and do is that I always keep an open mind to things that I find and look beautiful to me or that I'm inspired by, and I curate lots of Pinterest boards. And if I see something that I like, I will put it on one of these boards in a specific bucket. So I have several ones specifically about data visualization. So these have the most obvious and direct inspirations that I might get, but I also have a few boards, one about space, one about [national?]-- what's it called again? Historical, nature drawings - I totally lost the word there - spirographs, one about generative art. So it's all over the place, just whatever I think looks beautiful. And then what I do to be inspired-- so when I start a project, I first need to know from the client what is the goal, what should people learn when they see this visualization, and what is the data that is available to answer that question or reach that goal? And then while keeping that in mind, I kind of pick a few of these Pinterest boards that I think would be relatable to what I'm working on then. And I just browse through them and look at the things that kind of stick, things that I think, "Oh. Maybe like that," or, "Maybe that could be something," or it could be a very small part of the image that I'm looking at that somehow I think could be useful. And I put all of these things that stick into a new hidden client mood board. And then I start designing while having that client mood board open and just kind of looking at it every once in a while, and it's like, "Oh, right. Yes. I had that idea as well. Maybe I could try sketching something out." So it's kind of how that goes, yeah. 

SYDNEY: 12:07 

That's super cool. I love Pinterest. It's cool to hear you using it in a really professional sort of setting. 

NADIEH: 12:15 

Yeah. I'm glad it exists. If you could pay for it, I would pay for it. 

SYDNEY: 12:19 

Yeah. Yeah. So do you feel like you draw a lot of influence from your background in astronomy? Is that something that is still a part of your work today? 

NADIEH: 12:30 

Oh, yes. Yeah, it is. So one of the things that brought me to astronomy, besides enjoying just math and physics in general, was the gorgeous images that you could take, like Hubble. Any Hubble images basically are beautiful to me. And I often get inspiration from what I remember from space. So there are several visualizations that I've made that have a very direct inspiration to space, such as I made a family tree of all the European royal houses. And I made it look like it was a night sky, with every person being like a little star. And then the connections, you could kind of see them as constellation lines. But even then, people keep telling me, even though they don't know, that somehow they say, "There seems to be a space theme in your design." 

SYDNEY: 13:23 

What [laughter]? 

NADIEH: 13:24 

Yeah. Yeah. So even for one client, I made a network about-- it was UNESCO, and it was a network about the Intangible Cultural Heritage and how these are connected through things they share, like multiple cultural heritage being about vocal music, for example. And I built this up, and I was like, "Oh, maybe I could do this, do that." And then I sent it to the client in one of the many emails that we send back and forth. And he's like, "Ah, this is reminding me of a constellation, so let's just call it the constellation." I'm like, "Right [laughter]. I made another thing space related." 

SYDNEY: 13:56 

It's intentional [laughter]. 

NADIEH: 14:01 

So yeah, I still get a lot of inspiration from it. And I even make visualizations about astronomy approximately once every year. Even though I don't intend to, it just ends up that way. 

SYDNEY: 14:13 

Yeah. Yeah. So my background's in geography, specifically with remote sensing, so satellites looking back at the Earth. And I saw you did a piece - it was about the Earth - the Green Lung, and it was remotely sensed data. I thought that was really neat. 

NADIEH: 14:31 

Right. Yeah. Even though that one's been done before, showing how green the Earth has been, but it was a really fun one to just try and see what you make of the same idea. Yeah. 

SYDNEY: 14:43 

Yeah. So sometimes I feel like there's a perceived dichotomy between art and science or creative versus technical skills. And there's a notion that even though it's limiting to only market yourself at one, it's kind of unavoidable. Is this something you've encountered? And as someone who's very talented with data visualization, do you ever feel pigeonholed into one over the other, the technical or creative? 

NADIEH: 15:11 

No. I think I've been quite lucky in that sense. So I try and market myself as data visualization but the creative side of data visualization, where besides people understanding what the visualization says, it's also important that there's a visual appeal to it. It will draw people in and make them want to look at it. So I guess from the technical side, people see me as the more creative type. And I guess from the creative side, people see me as the technical one. So that kind of evens out. But I guess, yeah, I don't think-- well, I believe that art and science and creative and technical can all blend together, especially what I see people doing in things such as generative art, where they work together with the computer and program things to make it look beautiful. I really love people doing that, and it's something that I love doing myself more as well. And now that this is more easily possible with the computers getting better and more people being able to do it on their own, I think it won't be long before creative and technical can easily go hand in hand. 

SYDNEY: 16:28 

Yeah. Yeah. Are you kind of tracking or a part of the AI and art scene, so artists using general adversarial networks to generate art? 

NADIEH: 16:40 

Oh, right. Yeah. I guess I'm not yet quite into who's doing things with AI. I know about Mario Klingemann, who's doing some interesting stuff. I think more in the generative art scene, I'm a big fan of Matt - and I will screw up his last name - DesLauriers. He's doing great things-- and Manoloide, all these names. Those two are doing some really wonderful things. And they're also sharing part of their projects, their underlying process, every now and then, which really helps, I think, other people to get started and how to think about doing these things themselves. 

SYDNEY: 17:20 

Yeah. What do you think the value in open sourcing processes and code like that is? 

NADIEH: 17:26 

I think it's really big. So I could have never gotten to the point where I am now if there weren't tons of examples or fully fledged projects or tutorials and blogs just really available online. I'm really reliant on just seeing how somebody else has done it and then sort of remixing that or starting from that or understanding that. When I want to get into a language, I do tend to first pick up a very introductory book. But after that, for me, it's always based on examples. So I might see something that I find interesting, and that gives me the drive to want to create it myself and, therefore, understand the code underlying it. Yeah. Without open source, I would still be a data scientist working at Deloitte, I think. Not that that's bad [laughter], but I guess I'm more happy where I am now, I think. 

SYDNEY: 18:25 

Well, that's what's most important. So where do you see the future of data visualization as a job? Do you think that the increased presence of automation seen in data engineering and data science will have an influence on data visualization as well? 

NADIEH: 18:44 

I think just in general that-- so the wave of everybody gathering data and companies as well and the big data has come, and the hype has passed. But then people are starting to realize that to understand the data, as humans, we need to make it visual. So I do see an uptick in companies looking for a full-time data visualization person, which was not the case at all just a few years ago. But now you don't have to look extremely far to find one of these at a job opening, which I think is a good thing that they are starting to see that a good data visualization is not a skill that necessarily belongs to a data analyst. I mean, it's fine if they do it, but it's also a specialty. And if you want to do it really well, you probably need a specialist. So I think that's a good thing. And yeah, I guess the automation goes then hand in hand with the creation of dashboards and keeping-- so making sure that your stakeholders are up to date on what's happening in the company. So this is something that I only worked at at the start of my career. But I'm now not really touching on anymore because I cannot pour the creative part of me into making a dashboard because it's more meant to be like easily-- you need to understand what the dashboard is saying within just a few seconds. And I like creative things that are a little bit more-- are bigger than that, more stories, more ways to dive into the main insights [of that?]. So I haven't touched that, but I think-- I mean, even just looking at some of the tools that you do more dashboarding things, the more automated things, just Tableau or Power BI, these communities have been growing like crazy over the past few years. So I think that's, yeah, also a very good sign. 

SYDNEY: 20:43 

Yeah. Well, so what do you think-- do you find you have to strike a balance between something really clear and instantly understood and communicable that way and doing something really aesthetic and more story driven? Did that make sense? 

NADIEH: 21:01 

Yeah. I know what you mean. It's always the tough thing, yes, the balance between making it so that everybody understands it within a fraction of a second versus making it look beautiful and drawing people in and making it memorable. And there are a few cases where both of these things are reached at their highest level. It's gorgeous. It's very intuitive. But typically, if you want to get more from the one, you have to deliver a little bit on the other. So in that sense, for me, I've come to understand that I always want to make it, like I said, more diverse and provide more context around the main insight that I want to convey, so I will have to deliver on that, making it extremely intuitive. However, I do try and balance that with-- so if people do give the effort to try and understand how to see the visual, and they give that 10 seconds and 20 seconds to figure it out, when they do, they have access to this visualization that they can dive into for 10 minutes, 30 minutes, and see all kinds of different stories and different insights. Because it's not just one line where you can see, "Oh, things are going upwards," or, "Things are going downwards." Now maybe it's 5,000 gold medal winners from the Olympic Games that you see trends happening in different sports. You see trends happening across time. You see other trends happening. So that's how I try and balance that to make it both visually interesting, providing a lot of context, but yeah, delivering a little bit on the making it intuitive. 

SYDNEY: 22:42 

Yeah. The Olympic Feathers visualization you referenced is really neat. And I think that's something really cool about your work is it does, in a sense, demand a layer of exploration and interaction. It causes people to engage with it. 

NADIEH: 22:57 

Yeah, exactly. 

SYDNEY: 22:59 

So you're speaking at the Eyeo Festival in Minnesota in June. Can you give us a little teaser on what you'll be talking about? 

NADIEH: 23:08 

Sure. Yeah. So it's in Minneapolis, and it's one of these sort of more creative technology conferences. I've been wanting to go there for three years, but due to different reasons, couldn't go. So I'm very happy that I'm now a speaker, which means that they fly you over [laughter]. So yeah, I noticed that when I-- I wanted to make a new talk, and I was thinking about what it should be about. And then I figured out that actually a lot of the projects that I've been doing since starting freelancing could be seen as revealing connections. And it could be connections such as family relationships or connections such as similarities between Intangible Cultural Heritage from UNESCO or connections in how a hierarchy in a system works from tasks, calling new tasks, which you can call new tasks, and so on. So I basically have a presentation now that kind of takes you through the design process of about six of these projects and shows you how, even though they are all revealing connections, because the data and the goals are different for each of them, the visual result is completely different for all of them, and how I went about actually choosing which direction to go, like which design direction to go into, and some of the technical difficulties that I came across, and how I did some of the more challenging things. 

SYDNEY: 24:42 

That's exciting. Have you been to Minnesota before, Minneapolis? 

NADIEH: 24:47 

Yeah-- no, I haven't. Wait. Is Minneapolis in Minnesota? I really actually don't know. 

SYDNEY: 24:54 

So I went to college outside of Minneapolis, about an hour south. And-- 

NADIEH: 25:00 

Ah, I see. 

SYDNEY: 25:01 

Yeah. It's -- 

NADIEH: 25:02 

Oh. I didn't know [laughter]. 

SYDNEY: 25:02 

It'll be warm and humid in June [laughter]. 

NADIEH: 25:07 

Well, I don't mind that. I always prefer the warm parts. 

SYDNEY: 25:12 

Yeah. Where are you right now, and what time is it [laughter]? 

NADIEH: 25:16 

Oh. So I'm right outside of Amsterdam in a really tiny city of 5,000 people, just as I'm looking out at cows and sheep and windmills, actually. And it is-- 

SYDNEY: 25:25 


NADIEH: 25:26 

--a little bit over, like it's half past 4:00. 

SYDNEY: 25:32 

Oh, okay. So afternoonish. We're not getting you too late. 

NADIEH: 25:35 

No, no, no. I always have to think about saying the time in English, though, because it's different in Dutch, the order that you say things [laughter]. So it's like-- 

SYDNEY: 25:45 

I did not know that. 

NADIEH: 25:47 

Yeah. Weird, small changes. Yeah. Sorry. Now I completely forgot the question though. 

SYDNEY: 25:54 

Oh, man. Me too [laughter]. 

NADIEH: 25:56 

I think it was still on Eyeo, but that's about [crosstalk]. 

SYDNEY: 25:59 

Yeah, yeah. Eyeo, yep, and the journey to Minnesota. Minneapolis, it's a really cool city. Minneapolis-St. Paul, it's a really neat area. I think that'll be awesome. 

NADIEH: 26:09 

Cool. Oh, right. Yes. I haven't been there before, and I'm looking forward to going there. I think that was [laughter] [crosstalk]. I think that was [crosstalk]. 

SYDNEY: 26:15 

Yeah, good. So what's the best piece of advice you've ever gotten? You talked about kind of the advice you got in terms of speaking at conferences. Is there a better piece of advice that you've gotten? 

NADIEH: 26:31 

So my favorite piece of advice is not new or anything. But it is the one that I heard pretty early on in my career and which I think has helped me get to the point where I am now. And that is even if you can do something, if you don't enjoy doing it, don't advertise that you can do it. So don't put it on your résumé. If there's a project that you did for a client, but you weren't very happy with doing it again, just don't put it on your portfolio or these kinds of things or tell your manager. I mean, if they ask if you can do blah-blah-blah, sure, you can be honest and say that you can do it. But I would not go everywhere and say that you can do all of these programming languages if you only enjoy doing a few of them. 

SYDNEY: 27:22 

Yeah. That makes a lot of sense. Have you found that freelancing, you get to focus pretty exclusively on the things you enjoy doing, or is it still kind of a mix? 

NADIEH: 27:32 

It's more that, I guess, it's easier for me to pick and choose a little bit. So if someone comes to me and asks me for a dashboard, I can say no. [inaudible] I can't say no to everything that seems even remotely not interesting, but at least I can say no to a few of them, which helps to make it a little bit more specific. But there are always projects that you think are going to be cool or interesting at the start. But once you start doing it, you realize that wasn't quite the case as it is, and you just need to finish it and then move on. But I guess one of the advantages that I also find from freelancing is that I enjoy having multiple clients at once, maybe two or three. So even if one is not as interesting, you can easily just make sure that you finish the work that you do for that one, and then you can move on to one that you are still excited about. So yeah, that's also, I guess, another benefit I've found in freelancing that I didn't even expect. 

SYDNEY: 28:38 

More diversity in the projects you get to do? 

NADIEH: 28:40 


SYDNEY: 28:41 

What have been your favorite projects recently? What's been the most enjoyable to work on or interesting? 

NADIEH: 28:48 

So I guess my favorite project recently has been the one I did for Google. That was released a little over a month ago. So they're very open to [inaudible]. They've been asking data visualization people for the last two years, I believe, to make like these websites, these articles that are data visualization based and revolve around some aspect of Google's data. And I decided to look into how people search on Google to better understand their pets. So if you ever have been on Google and you typed, "Why does my cat like boxes?" that's kind of what I investigated. So what is the most popular thing that people search for to understand cats and dogs? And that was a really great one because I'm an animal person, and I really like dogs, and I have a cat, really like cats. And so it was really great, just kind of looking through all of this data and seeing things that I've wondered. And then, apparently, it's very popular, so a lot of people have wondered the same thing or very specific things that you're like, "Wait. I never realized that, but that's an interesting question to ask," and then visualizing this as well. It's been a lot of fun. 

SYDNEY: 30:02 

Yeah. The Google tracker data is super fascinating. Have you read the book Everybody Lies? 

NADIEH: 30:09 

No. I haven't. No. 

SYDNEY: 30:10 

You've heard of it? 

NADIEH: 30:11 

I feel like I've heard of it though, yes. I mean, apart from immediately thinking of House, I think I've also heard of the book [laughter]. 

SYDNEY: 30:18 

It's based pretty completely on the Google tracking data, and it's kind of broad-level psychologies based on what people search. It's interesting, if you have time [laughter]. 

NADIEH: 30:32 

I will. I mean, I love that. I've also read Thinking, Fast and Slow and Algorithms to Live By. All these [economics?], all these kind of odd, human, and the real-world connections, I find that fascinating. 

SYDNEY: 30:46 

Yeah. No. It's really interesting. The different data sources we have now, as opposed to what was available not very long ago, to understand how people think is really fascinating. 

NADIEH: 31:00 

It is already on my list now [laughter]. 

SYDNEY: 31:04 

Are there any upcoming projects you're super excited about, or is that kind of secret stuff? 

NADIEH: 31:10 

It's a very early state. So I kind of took a month off from working with clients to do, I guess, projects that I typically have to do in my free time, but I just couldn't find the time to do it. For example, I'm writing a book, together with Shirley Wu, about the Data Sketches collaborations that we've been doing. 

SYDNEY: 31:31 


NADIEH: 31:32 

So that was really cool. So, as a little bit of background, Data Sketches has been a 12-project-worth - it used to be a 12-month but that became a little longer - so a 12-project-worth collaboration that I did. And we always would start-- both of us would start from the same topic, such as movies or nostalgia or nature, and then we would create the visualization around that. And because we're both different people and we have different histories and interests, that would end up being two completely different visualizations. And we wrote about the design process that went into making each of these. And we're basically turning all of those write-ups and the screenshots into a book together with lessons that we learned. So we're really excited about it, but we both couldn't find the time to really work on it. And last week, she was in the Netherlands, so she was staying over with me. So that full week we could use to work on the book, for example. But I also had this new presentation to create for Eyeo. And I had some blog ideas that I wanted to write, but they take me like two full days to properly prepare-- and all of these things. And I was like, "I'm just going to take a month off," so that's what I've been doing lately. And now I'm picking up some new clients again. And so it's very early-- there could be something really cool with the museum here in the Netherlands, but that's really early [laughter]. So yeah, it's nothing I would say that I feel like I can say before-- instead of jinxing it [instead?]. 

SYDNEY: 33:07 

What has been writing a book like? Has that been pretty intensive, especially with someone living so far away from you? 

NADIEH: 33:17 

It's been more work than I thought, but if I take into account these write-ups, they were already partly existing online. I just fleshed them out for the book. What has been easy, though, is that I'm basically writing about what I've done. So it's not that I have to do a lot of literature research to make sure that I'm saying the right things, because I'm just writing about my experience and the screenshots that I made and how I did things. That's been easier, but it still takes a lot of time to, I guess, do it properly and not just-- so for the website, it didn't matter. It was just a website. It was out there. But for the book, you kind of want to do it better [laughter]. So then you go over it three or five or ten times, and every time, you do it a little bit differently. And working together with Shirley, though, so she's in San Francisco, that has been not an issue because, I mean, we both have our own chapters. So we can write that, and we talk every day on Slack or Twitter, so we can discuss things. And we have weekly update calls, in general, to talk about what we've been working on, and then we also talk a little bit about the book. However, it has been extremely helpful that we've had a week together just last week, but also a year ago when we really started with the book, to flesh out the design that we were looking for and the general setup that we were looking for. That was just easier to do face to face and sketching, both of us, and just having that one-on-one talk. So I guess if we didn't have that, that would have made it more difficult, but still doable, I think. 

SYDNEY: 34:57 

Yeah. Do you find iteration and kind of editing is a big part of your process, both in writing and the generation of visualizations? I think you've mentioned a few times throughout the time we've been talking together just the back-and-forth emails and the tweaking. 

NADIEH: 35:15 

Oh, yeah. Yes. Iteration is really part of basically everything I do, especially in data visualization. So I realized recently that when I was still more doing data science, it was much clearer to me how-- if a client had an issue or some question that we had to find the answer to, it was pretty clear to me, like, "Okay. So we'll need that, that, and that data set. We need to figure out where to get that. We can talk to these people, and then we need to clean it. Then we need to do these analyses. And hopefully, then, we will have figured it out." But with a data visualization, I only have a vague idea of how-- after the design phase, I have a vague idea of how I want to put the data on the screen. Is it going to be circular? Is it going to be a line [inaudible] circles? And only once I get the data on my screen do I see if my initial idea and my design works with this particular and full data set. And sometimes it does, and sometimes it doesn't. But even if it does work, I really design with code. So I think about colors only when I have the data on my screen, about the smaller effects that make it look visually more interesting, or how to provide this context. All of these things is really iteration based, just trying out different ways. I can't really do it any other way because I work with a bigger data set. So I really need to program things out to see if it really works when you apply it to a few hundred or a few thousand data points. And there is no way to do that beforehand in something like Sketch, Illustrator, or even in R, which isn't as diverse in creating visuals. So yeah, these days, it's very much a back and forth between some thoughts in my head, sketches on my page, and programming it out. 

SYDNEY: 37:05 

Do you build smaller data subprototypes before using the full data set with the visualization, or is it kind of just all in right away? 

NADIEH: 37:14 

All in right away, yeah. So I know where all my outliers are, my quirks, the weird things. Yeah. If I do it on a small part, it might just be like a lucky, good sample. And then you put in everything, and it's like, "Oh, wait. This is so not going to work [laughter]." 

SYDNEY: 37:32 

Just kidding. Thank you so much for spending the time talking with me this afternoon, for you, morning, for me. 

NADIEH: 37:42 

No problem. 

SYDNEY: 37:43 

Yeah. Do you have any community picks for us today? 

NADIEH: 37:47 

Yes. Yes, I do. So they are totally focused on data visualization, of course. 

SYDNEY: 37:52 


NADIEH: 37:53 

So I think it's now two months ago that the Data Visualization Society started, which is a-- yeah, it's just a community. They primarily now live on Slack, but it's very broad. So they also have people from the data visualization community write Medium articles about maybe a historic database or discussions that have been going on in the Slack channel. They actually sort of write that out into full blog posts. There are competitions within the community about some data set. I don't think you can really win anything except for eternal glory [laughter], but it's just a fun way to-- 

SYDNEY: 38:35 

And the fame? 

NADIEH: 38:37 

Yeah [laughter]. Yeah. It's very good. And people are very-- so you spend a lot of time actually going through these things and giving you proper critique on your visualizations. It's also a place where you can just share stuff that you've made and then have people respond on that. So it's a great place, and I think it's a really good thing that this has come up. And this is for any kind of data visualization, so not just D3. There are lots of people there from Tableau, Power BI, R. It doesn't matter what you do as long as you're interested in data visualization. So I'd suggest looking that one up. But specifically for D3, so the thing that I fell in love with, I can definitely highly recommend the book called Interactive Visualization for the Web-- Interactive Data Visualization for the Web, sorry, by Scott Murray, which for me, as somebody who came from completely not knowing anything about any kind of web language like the HTML, CSS, JavaScript, this book assumes that you know nothing. And that really was the right kickstarter for me. 

SYDNEY: 39:42 

It's awesome. Thank you so much for these recommendations. And of course, I'd strongly recommend checking out your website. Visual Cinnamon is your company's name? 

NADIEH: 39:52 

Yes. Yeah, that's right. It's mostly because my first name is pretty difficult to spell, so I just wanted something easier [laughter]. And-- 

SYDNEY: 40:00 


NADIEH: 40:02 

--through a weird process, it became Visual Cinnamon. 

SYDNEY: 40:06 

It's a very cool name. 

NADIEH: 40:08 

Thanks. It's two camps. Some people totally don't understand it, and other people are like, "All right. I get it [laughter]." 

SYDNEY: 40:18 

Yeah. So Everybody Lies by Seth Stephens is an awesome read this week. The Data Science Blog on the Alteryx Community, I always have to put some shameless self-promotion in there. And then, yeah, I think that's it this week. Thank you so much, Nadieh. 

NADIEH: 40:39 

Oh, you're welcome. 

SYDNEY: 40:39 

We're super grateful for the time you spent with us. [music] 

S3: 40:54 

Thanks for tuning in to Alter Everything. To share your thoughts and ideas for future episodes, join us at, or reach us on Twitter using the hashtag #AlterEverythingPodcast. Have a unique story to tell? Send us an email at Catch you next time. [music] 

SYDNEY: 41:25 

Awesome [laughter]. Yeah. Everybody Lies, I do not remember the author's name. It's funny. That's just something someone at work gave me a few days ago. 

NADIEH: 41:38 

Seth Stephens-Davidowitz. I see why you didn't remember it [laughter]. 

SYDNEY: 41:43 

Long words [laughter]. 


This episode of Alter Everything was produced by Maddie Johannsen (@MaddieJ).