By: Alberto Cairo

Dear friends, Alberto here. Let’s begin with a few updates:

The Computation+Journalism Symposium is less than a month away (December 11-12). There’s still time to register to attend; and here’s the full agenda.

We’re working hard on getting the Open Visualization Academy ready for its launch on January 31st. I’ve watched some of the courses we’ll host, and they look fantastic. But I still need to record mine! Busy times…

A visualization project I funded and participated in, the Trans News Initiative (TNI), launched on Monday. It’s a collaboration between the University of Miami’s School of Communication, the Trans Journalists Association, and Polygraph, the folks behind a publication we all love, The Pudding. I’ll probably write about the TNI in the future; for now, you can check this piece by the Columbia Journalism Review and this one by the TJA, which goes into much more detail.

My latest book, The Art of Insight, is a series of in-depth, and sometimes deeply personal conversations with data designers, artists and journalists. I enjoyed working on it so much that since it was published (two years ago, time flies!) I’ve been thinking about a follow-up connected to the Open Visualization Academy (OVA).

The result is a new video series, titled Friends of the Open Visualization Academy. I’ll publish it in this newsletter for now, and it’ll become a section of the OVA’s website when we launch it, on January 31st, 2026.

Today I bring you the first episode, a conversation with German data journalist Nina Krug. We talked about her career and about one of her most popular projects, a visualization of the sentiments expressed in public speeches by candidates in the recent German federal election. The video and transcript are below.

Here’s Nina’s bio:

Nina Krug is an information and communication designer, working from Hamburg, Germany. With her work, she pursues her mission of making complex topics accessible and easy to understand. 

Since completing her bachelor's degree in 2021 (and subsequently her master's degree at the renowned Bauhaus University), Nina Krug has already worked on various projects for renowned clients: one notable example is the exhibition ‘Adobe Masters: Fernando Baptista’, which she developed together with infographic experts John Grimwade and Michael Stoll.

In summer 2022, Nina taught the class ‘Good News Flash – Information Design in an Editorial Context’ at the Bauhaus University Weimar. 

Since October 2023, she has been a permanent member of the Data & Visualizations department at the news magazine DER SPIEGEL, where she advocates for more creative forms of data visualization. 

You can find Nina on Linkedin, Instagram, her author’s page at DER SPIEGEL, and her own website.

ALBERTO: Nina Krug, welcome to Friends of the Open Visualization Academy. How are you?

NINA: Hi Alberto, thank you. I'm doing great today and such an honor to be here as your guest.

ALBERTO: Yeah, you're my first guest. You're the very first friend of the Open Visualization Academy, although it has many friends already. The reason I invited you, just for our listeners to know, is that we saw each other at a conference in Augsburg, Germany, the Visual Discovery Conference organized by Professor Michael Stoll.

And you gave a fantastic presentation over there. I was really impressed by your work. And so I invited you to be a guest in this conversation.

So why don't we tell us a little bit about who you are, what you do, where you came from. Let's begin with that. Yeah.

NINA: My name is Nina. I am a graphics editor at the German news magazine Der Spiegel. And as you already mentioned, we met at Michael Stoll's conference because I was a student of his back in 2017.

And since then... That was not so long ago, right before the pandemic. That's true. It feels like very long ago.

ALBERTO: It's like a time warp. Yeah.

NINA: And somehow I ended up here.

And then I graduated with my bachelor's in visual communications and communication design and also finished a master's degree. Yeah. And straight from the university, I ended up in the newsroom, kind of.

Yeah. Now I'm sitting in the office in beautiful Hamburg and yeah, doing odd things with graphics. That's amazing.

ALBERTO: You have also taught, right? I remember reading that in your website. Have you taught in universities? Right. Only one class at Bauhaus University in Weimar.

NINA: I was teaching a little class on creating creative infographics in more analog ways. Oh, nice. What do you mean by creative infographics? Because that connects with your talk in Augsburg.

NINA: We will get to that in just one minute. That's true. Creative infographics, I think everything that is not like the standard chart that you would see probably everywhere.

And yeah, I had students who actually used acrylic paintings and also did some paper collages. So it was really nice to see how that developed. And how did it go? How was the experience of teaching? Oh, it was very nice.

I enjoyed it very much. It was a small but very diverse group of students. And some of them never had anything to do with creative graphics or graphics at all.

So it was a bit challenging. But I think the results spoke for themselves. It was really nice to see.

ALBERTO: It was mostly analog, as you were saying, right? Like using objects and paintings and things like that. Yeah, I'm super interested in that type of trend. Tell us a little bit more about Der Spiegel. So what does the team look like? How big it is? What types of professional profiles we can find in the team?

NINA: A lot of different profiles. So we're around about 25 people working in the data and visualizations team. And we're like mainly three different branches of designers, like with graphic and communication designers like myself.

But we also have a lot of visual coding people who just make the visions that we have possible to click and to work through. And yeah, we also have some data journalists to find very good stories. And you work mostly as an independent unit within the magazine, meaning that you propose your own stories to the rest of the magazine? Or do you work mostly on demand? So reporters come to you and make requests from the team? We actually do both.

We've been doing a lot of this demand and requested graphics for a while. But since I joined the team, I felt like there's been a shift and we're mostly regarded as a very own unit in this magazine. And it's very nice because we can just think of great stories that we want to tell or that we think that have visual potential.

And yeah, we actually have, we get the time and we can make it possible. And it's very nice.

ALBERTO: Well, that's very nice because I don't know whether when you were studying, Michael Stoll perhaps shared what it felt like to work in a graphics department in the old times.

When I began my career in 1997, the landscape, the way to work was very different. So graphics departments were basically a service department. So we waited for reporters to come give us requests for graphics.

And we changed the culture little by little. It took a lot of effort and a lot of time to take internal cultures in newsrooms up to this day when graphics department as the one that you work for essentially work independently as a content production units within newsrooms, which in my opinion is a great development because it leads to excellent work like the work that you shared at Visual Discovery. So why don't we shift to that? Let's talk about some of your favorite work.

NINA: Sure. Is there something that you really would like to see or? I mean, you shared, there was, we talked about this during the conference about the, your project about the language of politicians that really caught my attention, but we could talk about any project that you want.

ALBERTO: You can share your screen and share it with the viewers.

NINA: I'll do that. I already have the campaign speeches project open because I also really like that. And it was, yeah, one of the first projects that we did that was so different from like the usual things that we had all around elections.

So yeah, I can just scroll through it. That's just a little intro, but you might already see what it's coming.

ALBERTO: I don't speak German. So maybe a little bit of context there can help.

NINA: Oh yeah. That's just two quotes by those politicians that you can see.

It's former Chancellor Olaf Scholz with the yellow text and now our Chancellor Friedrich Merz with the orange text. And we just wanted to give like a little bit of a more tangible intro to the story before we really start. So yeah.

ALBERTO: And what is the title there?

NINA: Oh, I actually have to think about how to translate that. It's something like Alice Weidel angry, Friedrich Merz casual and Olaf Scholz like a coach. So it's the emotions that were conveyed by the speeches, which is what the project is truly about, right? Yeah, right.

It's about emotions, but we are not there yet. So we'll get there.

ALBERTO: That was a spoiler. That was a spoiler on my part.

NINA: Only a little one. But I actually want to stop at this point to show you a great team that I work with.

And all of those people that Alice said, they did incredible work. And it was like we've sat long hours in the office. And so I'm very thankful for all of them.

ALBERTO: Are all of them a part of the graphics team or was this a collaboration with people from the rest of the newsroom?

NINA: So at this time that we made it, they were all part of our graphics department. And one of them was an intern at the time. So she's not no longer here with us.

But yeah, all of the others were part of the team. Yeah, so it's a lot of text still. But this is the first time that the graphics from the beginning repeat.

And it's another quote by Friedrich Merz, who is complaining about the state the country is in, like he did for a long time because he was in the opposition. And it was one sentence of campaign speech that, yeah, was in Neuhof-Fulda. And it's down here, the sentence that he spoke, we marked it in red.

And by scrolling, we just show all the other sentences about migration that he said in a speech. And also different topics. And yeah, then we also changed it to sort them by topic, all of the sentences he said.

And we also changed that he really loved to talk about economy, which is the yellow part. Yeah, but that was not everything we wanted. So we had the sentences transcribed from videos from the campaign speeches.

And we thought it was also very important to showcase how the people said what they said. And this is why we also had an AI trained to match the sentences to corresponding emotions. And we also hand checked that later.

And this is why we'll spend so, so many hours.

ALBERTO: I was about to ask, I would be a little bit nervous if I put these tasks in the virtual hands of an AI.

NINA: Yes, we were as well. And so we did not only let the AI check or match them once, but we also had to match it, like, I think, up to 10 times. And still, then there were some sentences where the AI was not really sure how to classify it. So we did that by hand.

And also inside of the team, we were also not entirely sure what we would, yeah, how we would classify the emotions or what would match it. So yeah, we then agreed on the most common answer that we gave. So yeah, it was also a little experiment.

And yeah, these are so different quotes that he said. And then we compared them, of course, to all the other candidates for the election, which are and Robert Habeck. And you can see there's a real difference in the topics that they touch on.

ALBERTO: So let me stop, let me stop you for a second, because this is one of the parts that got my attention the most, because these are essentially stacked bar graphs, right? So we can see the relative proportions, but the length of the speeches, were all the speeches the same length, the same duration, or do you normalize all the speeches, meaning are these percentages of time?

NINA: Yeah, these are percentages, it would have been too difficult to just have them all in different sizes and heights. And because it was already very difficult to just establish this color scheme. And then also the marked sentences.

And we just wanted to make like give an overview and percentages were Yeah, that's the best way for a reader to understand what we were showing. So yeah, and this is a different color that we had about the quote from Robert Habeck. And he was talking about the debate culture in the country.

And this was probably one of his favorite topics, because he spoke a lot about that. And we also, yeah, made this tiny bar chart kind of very short percentages for all the emotions and not just the Yeah, the topics that they touched on, but also how they spoke.

ALBERTO: And but the shapes also convey the emotion, we are going to get to that really excited about that.

NINA: We can get to that in a minute. Yeah, I'll just scroll through that. So then we combined them.

And then this cards that we have, we had the color for the topics and line work for the emotions. And in the end, we just made the whole thing interactive. So you could click on any line and read any sentence that we had and also click.

Yeah. To the next sentence, or yeah, to the following one. Right.

And you could also change the candidates just to see. Oh, and as you can see, far right candidate Alice Weidel was very angry at the time, because that is what we had like the zigzag line. Yeah, we should explain that.

ALBERTO: This is what got my attention the most, which are what I really like, because I found it really evocative, which is that even if I don't speak German, I take a look at these and see the spiky shapes and that guy was was angry or that person was angry. Whereas if I see the smoother shapes that it's sort of like it sort of like conveys a calming experience, I don't know exactly what the sentiment was. But can we go back to the to the legend and read it out loud? Yeah, because we have on the one hand, we have colors, which correspond to the topics.

But then we have the shapes. Right. And what is it that each shape, what emotion each shape communicates or represents? We have like the tiny squiggly line on top.

NINA: That's like the scary line for me. I tried to make it as scary as possible, but also not too scared. And then we have like the angry line, the line that zigzags.

And we have little slopes, which was a line of hope. We also had like the very neutral lines. And just if nothing really applied to the sentence that I said.

ALBERTO: Let's go back to the to the to the part where we see all the all the lines together. Yeah, this looks like it's an amazing, an amazing picture. Can you talk a little bit about the technologies that were used a to create this project?

NINA: Yeah, it was a lot of different.

Yeah, tools that we use. So the lines that I did, they were actually all hand drawn. And I think I have a picture of that somewhere prepared that for today.

Just open this one. So they were actually all hand drawn and I have pages in my notebook, where you can actually see that. Yeah, I was just trying out different lines.

I have also like a very chaotic line because we thought maybe we want to classify emotions as chaotic or as not leading anywhere. But yeah, this one did not make it to the final project. But that's also very nice to to show that to you.

ALBERTO: You also thought about using little dots over there as far as I see. What were you trying to with the dots?

NINA: They were not used at the end. But I think it was irritation.

I think there were notes. I will just look it up. I was thinking about using this as the bigger like the dots, just the contour dots were like enthusiasm, but I did not think that it conveyed as well as.

ALBERTO: Yeah, it's not it's not as intuitive. Yeah. This is such an interesting.

It was a very interesting project to me because it speaks about the fact that data in data visualization can be encoded in multiple ways, which are not limited to the traditional encodings, which are, you know, color, a position, area, things like that. But sometimes we forget that categorical encoding such as shape are also of and particularly shapes that try to suggest, in this case, an emotion can be extremely useful to create both an informative experience, but also an evocative experience. And perhaps that is the reason why this project got my attention so much when you when you presented it, but also the importance of including in legends, because as you were just saying, for example, a spiky shape may convey something to me anger, but it may be conveyed, it may be conveying something to you.

Or if you show me sort of like the an emotion represented by little dots, I may interpret which are not connected to each other, I may interpret that, for example, as a person who doesn't know exactly what they want to say. And therefore, they stop themselves to think what they want to see next. And therefore, the lack of continuity in the discourse, but it may convey something entirely different to you.

And that sometimes I think, conveys the importance of including an annotation layer in visualization to basically overcome those possible misinterpretations on the part of the reader, which I think that sometimes we journalists forget to include, because we think that our visualizations are intuitive, and visualizations are rarely intuitive, right?

NINA: Right. And so we also have like the tiny legends in the interactive piece, they're just right here, cannot write them down.

ALBERTO: Yeah, it's a lovely legend there.

NINA: It was very important to us to have it visible all the time. So also, if you click on a sentence, we also have it listed down here. Yeah, yeah.

ALBERTO: So there was a reminder on every sentence of what the emotion conveyed. Yes, yes. And you can also filter it by emotion.

NINA: So if you just want to see every sentence where I was very angry, you see, it's a lot.

ALBERTO: Yeah, yeah. So we were talking about the technology. So first of all, before I interrupted you, sorry about that. But I really like this sort of like this encoding part. But you were talking about, first of all, you use analog technologies, essentially to come up with the ideas. What do you do next?

NINA: So I had to digitalize them somehow. And because it was easy, the easiest for a coding colleague to just work with SVGs, I had to, yeah, have them drawn an illustrator after that. So the nice little hand drawn effects got a little bit lost.

But I think that there's still enough of it there. But yeah, so I just slice it. And then I handed it to our nice colleague who coded all of this together.

We had different kinds of layouts planned, and we tried out different things in Figma, actually. And yeah, this is what we decided on in the end. And also for the intro, that's actually not coded.

ALBERTO: That's like, I think it's a lot of technology. So it's a vector animation, but it's also based on PNGs. So it's actually picture animation, but it's not SVGs.

NINA: Okay. No, because we just then used what's far easier to load the tiny PNGs than the SVGs, because it was also easier to handle in After Effects, where I put all of this together. Because working from illustrator to After Effects is still not perfect.

It is not. It is not perfect indeed. There's a bridge missing somewhere, and it shows.

ALBERTO: I was about to say that, I mean, all this variety of technologies also speak about the ingenuity of designers, right, and of data journalists, which is something that I have experienced myself. We don't use a single technology, right, to create these types of pieces. When you see something like that, you may be led to think that everything is coded.

Everything here is JavaScript and D3 and whatever, and in part it could be, but then other things are essentially just handmade, handcrafted, and then manually animated.

NINA: Yeah, they actually have to be handmade sometimes, because the colleague who coded it was busy coding the huge interactive part, and she did not even see the possibility of doing the intro as well. We always try to get the job done as quickly as possible.

ALBERTO: Talking about that, how long it took to create this project from conception to publishing?

NINA: So, I think we started with the idea in late November last year, and the elections were at the end of February, so it's been a good amount of time. But we also did not work on it full time. We did other projects as well, so I was also doing the magazine graphics for print, and also other graphics for digital, and we all, yeah, came back to it whenever we had some time that we could spare, and the most important part was to get the classification done and the transcribing of the speeches, so I think we really started with the visualization in early January, and up to the last day, we were just still manually checking all of the classifications and the matching, so it was as, yeah, as correct as it could be.

ALBERTO: Yeah, yeah. Well, that is a great, I think, it's good information for our viewers who have never looked into the inner workings of a news organization, but that type of working schedule is very typical. On the one hand, you're working on your daily products, daily graphics, well, in your case, the weekly graphics, things that you need to do on a deadline very, very quickly, and for the print magazine, or for the print newspaper, like it was my case when I began my career, but then on the side, you always have these sort of, like, long-term projects that you have time to essentially evolve and work on and tweak and edit, and the result is something as fantastic, in my opinion, as these.

Thank you. Yeah. Nina, thank you so much for being with us here today.

NINA: It was a great pleasure, and it was a great… Thank you so much for having me. Yeah, it was great to see you in Augsburg, in Germany, also a few weeks ago. Thank you so much.