Digital typography: Suggesting, not dictating

Juli Sombat, Design Lead and Jason Pamental, Principal Designer

Chris Strahl: Hi, and welcome to the Design Systems podcast. This podcast is about the place where design and development overlap. We talk with experts to get their point of view about trends in design code and how it relates to the world around us. As always, this podcast is brought to you by Knapsack. Check us out at If you want to get in touch with the show, ask some questions, or generally tell us what you think. Go ahead and tweet us out @theDSpod. We'd love to hear from you. Hey everybody, this is Chris Strahl, the host of the Design Systems podcast. Today I'm here with Juli Sombat and Jason Pamental. Julie is the product design manager at the Encore Design System. Jason is the principal designer for CHIRP and a member of the web fonts working group. And today we're going to talk about type. Welcome guys. Thanks.

Jason Pamental: Hello.

Chris Strahl: So you all know more about typography than I probably will ever in my entire life. The sum collective knowledge of this, I get to do my favorite job in a podcast and I get to play the useful idiot that asked a lot of questions. And so I obviously know what typography is at fairly basic level. I know that it sort of represents the words that we see in our digital products, but you all think about this in a much more complex way, and I know much like how color systems are pretty foundational to an expression of a brand type systems are as well. And so maybe just give me a little primer on how you think about typography and design systems.

Jason Pamental: My favorite quote is from Beatrice Ward from the thirties, and she was the creative director editor of Monotype Recorder Magazine, and she said something along the lines of type is the clothes, our words wear. And I just has always been one of the most wonderful of what it means. It's the voice of our words. And for something about how to convey emotion, tone, just connectedness, humanness, there's so many things that type can do to kind of influence how you sort of hear what you read. It's just always been fascinating to me.

Juli Sombat: Yeah, I love that and I love the audio reference of course. I think what's interesting as well is that it just ties product and brand together so much because of what you were just saying and because you want the brand character, the feelings, the human-centered approach to really come through the product and type is a major way that we can do that.

Chris Strahl: Gotcha. So when you think about the expression of type in design systems, my head immediately goes to tokens. And so how do you think about those creation of those tokens? Because it's not as simple as like, here's one token that expresses my type, or here's a pair of tokens that expresses type. It's much more complicated than that. So when you think about that systematic expression of the close, our words wear across all of our different digital products, what do you think of when you construct a type system in something that is ultimately going to live in a lot of different places?

Jason Pamental: Well, I only have to worry about three different places things show up. I really want to hear what Julie has to contend with at Spotify because you have so many more places, things show up and you have to be so much more varied in how it's applied.

Juli Sombat: That's correct. We're dealing with 45 platforms. So it's not just web and mobile, it's a Samsung tv, it's everything in your car. We also have a car mode, mobile devices that it knows when you're driving and it's going to make the type bigger. We support 72 languages. We've got accessibility sizing. There's just so many different levers that can be pulled that will change the display of the typography in wherever you're looking at it. So the complexity, even thinking about tokens, which are a really trustworthy way to manage that complexity, it makes my head spin. The other thing that's really interesting is that there's just so many ways to control it in the browser. In the token itself, the information can live in multiple places. So there's a lot of decisions that can be made

Chris Strahl: Well, and presumably there's lots of different ways to accomplish the same thing. And so the implementation of type at that sort of last mile, there's probably a lot of different pathways to that that ultimately represent a similar or the same solution. So how do you wrap your head around that at scale?

Jason Pamental: So at Chewy on our design system, we're currently working with two typefaces on the web, different one in iOS, different one in Android, and with a goal of moving towards something that's more unified. But then we also have different sort of expressions of that across Chewy. And then we also have PET md, which is a content focused website, which has totally different branding. And so thinking through the token system, there was as much about how to be responsive in hierarchy and then how to be flexible in tone. So the tone being more about the typeface and the weight. The weight also plays into having more hierarchy in how loudly or softly you speak or how legible it is based on the size usage. So you might want to tweak the weight of a typeface if you're going to be using it smaller. So there's a lot of different considerations there.

We have a smaller number of them so far. We've been able to create a set of tokens that has things related to scale, like the font size and then a set that's related to weight and typeface itself. And then those kind of roll up into a style. We use Figma. So display five means different things on different platforms, but as a designer we can have one component that is mapped to that style and then it would work with a theme swap in any of those different expressions. As we're designing language is new, we actually just announced that we're going to move to Canada. We're going to be opening up Chewy in Canada later this year. And so that's the beginning of our exploration into internationalization. I'm just going to be coming back to replay this, so I'm going to let you take it.

Juli Sombat: Yeah, that's a hairy problem. I wonder what birds say in French. That could be the name. So we do something similar. We have different scales that define the hierarchy for different device classes. So we've got a small scale that mobile uses, a medium scale that desktop uses a large scale that smart TVs might use. So the same tokens across that line will differ in sizes to accommodate the devices. What I was thinking about when you first asked the question though was there are so many ways to get to the same solution. So a designer is going to present one solution. A web developer's going to think of it in a different way, and then the mobile developer's going to have a totally different point of view. So the way for us to align on this was to get everyone in the room and really have a cross-functional conversation. You can't come at it from a singular perspective. It has to be a team effort. And that's helped us understand the restrictions and the capabilities of each platform and get to a place where things can work relatively well across the scale.

Chris Strahl: So when you think about that and you having that person that does living room devices or auto devices next to that person that's doing an Android app and they're next to a person that's doing a web application, what does that conversation look like? Is that talking about implementation differences? Was that talking about fundamental structural differences in the way that you think about how typography is expressed?

Juli Sombat: That work lives in a phase of the process that we refer to as auditing, and it's owned by the design systems team. So we'll have developers and designers and sometimes product managers who all work on encore, go out and talk to the internal customers, talk to the other developers and understand the requirements and really get a sense of what those teams need to be able to use a token and have it do what they need it to do. So it helps us understand what we have to build.

Chris Strahl: So at the end of the day, you're trying to create some sort of harmonization at a very high level. So like, Hey, here's our brand. And then there's expressions of that that exist across platforms, but then there's also probably expressions of that that exist differently within teams. And so when you think about that levels of expression, that decision-making is then audited by that centralized team, is that what you're saying?

Juli Sombat: Yeah. We have to understand the functionality that they need in not only the tokens, but also the label components that we build because there's also information at that level. So that might be things like accessibility sizing or truncation logic that would define how things shorten. And then we've got the whole set of functionality that we need to build and we can start understanding how to do that.

Chris Strahl: So what about you, Jason? That's a lot, right? And so how does that look like in a little bit more contained structure?

Jason Pamental: I think we tend to start with the thing that we can prototype the most easily to convey what we're trying to get across. And so for that, for us it's starts with Figma and Web. We have a really tight partnership with the engineers that build the design system components. I actually have known Al Stefan is the design manager on that engineering team, and we've known each other from Drupal days going back to, I don't know, DrupalCon 2011 or something like that. And it was literally arguing over type in Drupal core in a bar in Chicago at two in the morning. I think that was the origin story of our working together. It's very on-brand, but I think it's easy for us to capture what we're trying to get across there and then look at how does this then scale out to iOS or Android. And I wanted to mention one of these things that was completely new to me when I first started working because I was much more web centric.

The whole thing about dynamic type and iOS is such a pain. It works so differently. I mean it's super important, but because it works and it's created so differently than the way we work on the web, the differences between there and the implementation of typography and iOS and on Android are all good in their own way, but it's like they were all done in a complete vacuum. So it's really hard to apply some of those same concepts. And so that's where that audit and those conversations are so critical to get across what you mean and let the developer who has the expertise in that platform and say, okay, I understand what you're getting at. This is what you want to have happen. This is how we would have to do it over here. And then we try and work that backwards into Figma so that we can have libraries that will approximate best how that will happen. And the design system team is really well equipped to create those things so that all the different product designers on the team have something to work with where they can swap a library and say like, okay, this is pretty much what it's going to look like on the web and how I can understand what the differences are going to be when this is going to travel over to iOS or Android.

Chris Strahl: Gotcha. So there's some fundamental core differences both in implementation and in structure for a lot of type in different platforms, and that makes it hard to sort of harmonize things. And the really intense work sounds like it's about how do you deal with multi-platform less how you deal with an individual implementation?

Jason Pamental: Well, I think it's like encoding. What do you mean when I say responsive typography? That means something to someone who works on the web that's very different from someone that develops for Android or iOS because we have tools in mind. Whatever the thing is you use to express it, that's kind of the vocabulary that you have to work with. And it's gotten better and better and better on the web. And now the differential is like, okay, how do we approximate this or do things in a somewhat similar manner where browser vendors have done an amazing job with the W three C to improve the type of graphic controls we have on the web. And it's not that you can't do these things in those apps, but is it rendered the same way a car interface is renders the type with a completely different rendering engine than a browser? So what do you do with that? That's a fascinating problem. That's really something that I am so fascinated by what you all have to work with at Spotify.

Juli Sombat: Even the vocabulary, the words that we use to describe something like adaptive sizing would be different with the web or with on Android. So yeah, it's a lot of the same problems that design systems solve over and over, but just bringing the disciplines together and giving them a shared language.

Chris Strahl: Can you help me with a concrete example of this? Because I think I understand conceptually what you're talking about, but maybe grounding this in the idea of this is a real life implementation problem. Do you have one of those that you can just pull out of a back pocket and be like, Hey, I was struggling with this the other day.

Juli Sombat: Yeah, I don't know if it'll exactly show you everything, but I have a good case study. So we've been dealing with an Arabic clipping bug. So when type is rendered in Arabic, there are a few characters in that script that reach way below the baseline. So they go really low in the line of text. And the way that we develop that on the web is okay, it can show what's below the baseline, but on mobile it gets cut off. So it looks like the letters are just completely cut off. You can't see the rest of the character.

Chris Strahl: It'd be like if you cut off the bottom of an L in English

Juli Sombat: Or like a J because it's going underneath. Gotcha.

Chris Strahl: Okay.

Juli Sombat: The way that we used to solve that would be just looking at those exact characters, just looking at the problem, the way that you would normally go in and diagnose a bug and whatever it would be, change the bounding box of the type so that it's fitting everything in the line. But now that we have a staff typographer on our team, they have the bandwidth and the expertise to be able to take a very holistic look at this and not only look at Arabic, not only look at Android, but really look at vertical metrics and how the sizes are working across all the scripts that we support and propose a solution that's much more holistic. So I'm extremely grateful that we're able to work in this way, but what she found was that we needed to adjust vertical metrics across all of our scripts in order to have a more unified look and feel. And so that fixed the bug, but also just improved everything across the board.

Chris Strahl: So that makes sense. So you have able to basically say like, all right, here's this specific problem, but then there's this systematic solution that is broadly beneficial across things way beyond that individual problem,

Juli Sombat: And that will set us up for adding a new language when we need to do that next.

Jason Pamental: Yeah, I mean that's something that I think the more I got into typography on the web, the more I got into being a part of a type I, being a part of the web fonts working group, you start to be exposed to all of these things that you were just talking about because Bianca and I are whom she's referring to are both on the web font working group. And as you sort of get into the internals of a font and you start to realize there's a whole bunch of information there that tells you about that bounding box and where the characters extend to. And the more you're aware of that, the more your application could actually do something useful with that information and then you can build it into a system. Because otherwise, like Julie said, if you just try and solve for those four glyphs, well, in Arabic typeface is like two and a half megs.

There's a whole lot of characters there. And to represent the language in an appropriate way, you want to respect the way it's written, that's a very calligraphic language. Other ones like Korean or Japanese or Devan, those languages in those scripts are actually pretty regimented in how far they extend above and below. But Arabic is based on a pen, and if you really want it to feel, if you want it to have a good voice, you have to be able to respect all of that. And they have one of my favorite people in the world working on that problem. Bianca's so smart, she's so good at this stuff, and the perfect person to be working on that had so much experience in the web as well.

Chris Strahl: I mean, Julie, I know this is a big passion area for you around localization, but when you think about localization just beyond, for example, the type structure of Arabic, what are those considerations that you're thinking about when you're dealing with different languages or different voices depending on the region or the person that is consuming your app?

Juli Sombat: So what I love about this is just how complex of a design problem it is and also how much it is a design problem. There's so many technical aspects, but it's really about how a person understands and feels and uses our product. So there's two things. One, Spotify has a lot of cases where you're mixing scripts. So a name of an album might have some English in it, and that would be two different typefaces, which is also a big problem to render properly and make it look not broken. But beyond that, we have a lot of mixed content on Spotify. You have Spotify owned and not operated playlists that need to look like our brand, and those mix with album covers people's individual playlists. There's a lot of content coming at you that you need to rapidly parse. So by having a very strong branded typeface, that's a major way that we as a company can take a stab in the ground and own those things visually. And that needs to translate across beyond Latin scripts. So even in things that are right to left, things that are the C J K fonts, there's just so many different ways that we still want to show that it's from Spotify.

Chris Strahl: It's interesting because I've got friends that do a lot of d d podcasts, and so they'll introduce unusual Norse glyphs into the titles of their podcasts. They want to make it look like medieval and cool and fantasy worldy, and so they want it to look like Tolkien's Elvis as their title. And I can't even imagine the monkey wrench that that must cause from time to time

Juli Sombat: That doesn't translate super well, so is not recommended, but the effect works.

Chris Strahl: Yeah, I think that I had a friend that was doing a Eldridge horror cyberpunk one that was using a bizarre font that sort of expanded way out in the middle of every word and then became thin at the ends again. I can't even imagine what that must look like for an implementation.

Juli Sombat: Yeah, that's the kind of challenge with user generated content and we get emojis thrown in there and all sorts of things. I'm more concerned about what that's like on a screen reader because it can get a little messy

Chris Strahl: For sure. And these people are like 20 listeners or something like that, so they don't think about stuff like that. But it is an interesting problem for you then because you have to think about stuff like that. What's been the biggest exciting moment in that localization challenge that has been something that you were like, oh, we solved that and that was really cool and benefited a ton of different people?

Juli Sombat: Yeah, I mean I think anytime we're investing in global typography and solving problems, it's exciting no matter what, because we want to enhance the user experience for people around the world. One major one that we did was we had a problem with Vietnamese where we weren't supporting the entire glyph set that they needed for whatever reason. So some things would be subbed out for other typefaces, so it looked like a ransom note as you were reading the words. And when we were able to diagnose that and improve that, we felt a lot better.

Jason Pamental: That was one of the driving forces behind the creation of Nodo Sans as a family of, well, not just the whole Nodo family, the San Serif, there's a serif version, but it's a script that is intended to cover the entire world. And then being able to layer those font files when it's not available in the first one, it's going to look in the next one and so forth. And there's some interesting strategies you can then come up with to help avoid that. But I think that's the most fascinating challenge with all of this is if as soon as you want to stray away from something that is resonant on the user's system, if you want to create that branded experience, eventually you're going to hit this problem where you get past the basic couple hundred characters that are part of what many people in any kind of western language reading it would encounter. And it's 16,000 in Japanese, I don't know how many thousands in Arabic. And then you've like every other script, it's like on and on and on. And if you are working on a platform that is used in every country in every time zone around the world, that's a big challenge. It's a lot of letters.

Juli Sombat: It has big performance implications too. If you have to force users to download those fonts or wait for them to load, it can really degrade things.

Chris Strahl: So Jason, you sort of have another passion area that is definitely related to this conversation, but a little bit different and that since I've known you, you've talked to everybody that's been willing to listen about variable fonts and you're the person that introduced me to this concept. And so I'd love your take on why you're so passionate about this and what excites you about it

Jason Pamental: As it happens. I actually just hit published on an article for Monotype on Friday working on a case study of what variable fonts can do for a brand and for user experience. And in a tiny nutshell, it is encapsulating all the different variations of a typeface, all the different weights, possibly different widths, different character shapes, all of those things in a single font file. So when you think about for what we do, whether it's on the web or packaged in an app, delivering that font information in a single file allows you to build all of it into your design. And with design systems in particular, that's where I think it really shines because you can do things for light and dark mode to adjust the weight of the text so that when it's in light text on a dark background, you probably want to make it a little bit lower weight than you would for dark text on a light background because there's sort of a light gain effect where it just kind of looks heavier.

And so if you can't make those adjustments, not a full step, but maybe because that would be like a hundred difference of 400 to 300, but 400 down to three 70 might be perfect. And the variable axis allows you to reach anywhere in that continuum or do things like fine tune the width of it based on the size of the screen. So if a typeface has a width axis, being able to on a small screen make it maybe 95% of the width, you fit a few more characters per line, you get a little bit better reading experience for people. There's another axis called the optical size axis, which if you imagine I'm going to borrow from the times New Roman that is on that no one else can see, but we can, when it's set really large, oftentimes the stroke contrast, the thick and thin can be really exaggerated.

So if you think of the cover of Vogue, I mean different typeface, but same idea. You have that really beautiful contrast of stroke weight when it's really small, that stroke weight would disappear. So an optical size axis will help attune that so that a typeface can be more legible at a small size, but a little bit more editorial at a really big one. And so those are always that you can kind of increase that vocal, that sort of dynamic range in your type without having to load another asset, especially in a dynamic system. So things that we work on at Chewy, things that I worked on at the state of Rhode Island or Georgia for their state web platforms, Spotify is a great example where you would love to be able to be a little bit more dynamic in the way you typeset specific bits of content.

Building that control becomes a lot more feasible, right into a content management system, for example. So that not only can your design system work for all content equally well, it could work for very specific content in a unique way without compromising the experience in either direction. And then all of that is still available to customize for if you had aspects of the font that are helpful for someone with dyslexia to modulate the shape of the characters a little bit, to be a little bit more legible to them. I'm just scratching the surface with all of these things, but variable fonts expose all of that to us as designers within the system and then also in that specific, and I just think it's fascinating and I think we're only just scratching the surface of what we could do with it from a design standpoint, from accessibility, and then it feeds into performance and all kinds of other aspects as well.

Chris Strahl: Yeah, I tend to use dyslexia friendly fonts. I've never been formally diagnosed, but it's more readable to me and it's very cumbersome to adjust that app by app by app or experience by experience. And the ability to have that as some sort of token that would follow me around my digital experiences would be really, really cool. I actually think that this is one of the interesting long-term ideas around design systems is this idea of what happens if we're able to actually have some sort of personalization token that represents what set of tokens we actually experience as a user that may vary from user to user to user, and hey, maybe we'll actually get web personalization at scale. But the interesting thing that you were talking about there, Jason, is much things like adaptive color. The challenge here is how do you represent that as a set of variables?

And I think that we're at a point now where that's maybe more well understood than it has been previously where you basically say, I have all those different things that you were talking about, the ability to change, the width of characters, the ability to change some of the optical treatments of them, the ability to change the weight of them dynamically as actual tokens in and of themselves. So type doesn't just become a typeface with a weight, it becomes a bunch of different tokens that all sort of group together to form a set that represents that experience and changes in the context of that set can also then edit those sort of token variables like you mentioned dark mode, right? And I think this fits very much with what Figma has been going for with the whole modes and variables thing inside of the recent config announcement. But I'm curious, is that how you see the implementation of this happening?

Jason Pamental: It's getting there for sure. Weirdly enough, I was at an event with Marson Wicky last weekend, and he is the driver of a lot of the typographic improvements in Figma, and he was responsible for helping ensure that they got variable font support in there. He's written a bunch of wonderful pieces there, and we were talking a little bit about this, and I think they feel like they're just scratching the surface with variables and how they can translate and implement and then go from Figma to code. There's so many different ways that it can get done. The basic things that you would think about in type size, weight, the typeface itself, the line height, the spacing before and after, any letter spacing you might add, magnify that with word spacing and other user configuration. To use Julie's phrase, the levers that you have to play with there, it's a massive number of them.

And as designers, our job is to pick the positions of those levers that work for the broadest spectrum of people in the broadest number of scenarios. But being able to expose modifiers to the end user I think is magic where you can allow someone to say like, okay, if I suffer from crowding, which is a condition that's a subset of people who have dyslexia suffer from, where if we can increase the spacing, the line spacing and the word spacing, their reading comprehension could be doubled. I'm kind of fudging the numbers a little bit, but it's a big difference. And the moment you put those controls in the hand of the reader, then they get something out of it that they could not before. And that's a really powerful thing. Tim Brown had a great quote in his book As Typographers now in a digital age, we don't dictate the typography, we suggest it, and we need to put out a good set of suggestions and a smart set of levers for the reader to make it useful for themselves. I love what you said, Julie, about the driving mode in Spotify, the ability to react to that context, allowing the user to customize that even further is just a wonderful extension of that sort of conversation that we have with the reader, the user of the app, the consumer of our pet items. That's a much stronger bond. It's a much more durable one. So I mean, I think it goes way beyond just being good. It also is really smart for us to have that kind of relationship with our customers, our users, and our readers.

Juli Sombat: And I think that's the reality of digital typography where we know that things aren't going to be what the designer builds and creates necessarily. There are plenty of ways for them to change things on their own in order to suit their needs. And that's such a mental shift for designers because we're coming from maybe a print background, we're learning things in design school about the best way to do things, and then it turns out that that might not be the best way. So it's still something that people are wrapping their heads around

Jason Pamental: Well, and what's the most common device that someone's using to access it in all these different countries, like the size of the screen, the density of the pixels, the quality of the text rendering, all of those things make it look beautiful or terrible, and we don't know. We can't know. So it has to be a more reactive kind of system, and you have to learn where that control is beyond you.

Chris Strahl: It's also interesting, right, because you can't just have people select every option because then you end up in a cockpit and nobody really likes a cockpit without years of training. It can be really intimidating. And so you have this one side of it that's like, how do you create reasonable constraints and sane defaults? The other side of it is how do you become context aware? How do you know somebody's actually driving? And then I think the last part is if users are able to change things, how do you keep it in runtime? And that's a lot of different, fairly substantial technical hurdles to what we're talking about. But I think that there is this really interesting idea about, as we think more about tokens and we think about the way that tokens impact our ability to not just express a brand like I was talking about very early in the show, but now actually personalize that brand to be more consumable accessible and a better experience for a user. I think we're going to start to see these problems get solved with interesting ways of changing these things on the fly without having to build our software in such a way that they're all available to everybody.

Juli Sombat: Yeah, there's such an opportunity with variable fonts to support the brand better than we do right now maybe, and create a axes that would allow for more brand expression or less brand expressions. You might have something that works for really creative marketing experiences that then scales back for a long form data dashboard.

Chris Strahl: So you all have had so much experience working with this and spending time on this and really thinking deep about this. When was the moment that you sort of caught this passion? When is it that you said like, Hey, this is something that's really complex and you sort of became aware of the scope and challenge of the problem?

Juli Sombat: So at Spotify, our design systems all want to look and feel like Spotify. So we all use circular. We all try to adhere to the main Spotify brand. But before that, I was working at Conde Nest where we were taking 20 different magazine brands and trying to build a design system umbrella that could unite all of those very different look and feels. So it was the opposite problem pretty much. And we had to do that by letting each brand identity host its own typefaces. And so it was a lot of performance concerns and conversations around, do you need 14 typefaces on a single website, which was a whole different can of worms.

Jason Pamental: So that was really kind of where you got your first deep look at the systems approaches to typography, or was there something that was even before that?

Juli Sombat: Well, I studied graphic design at risd, Rhode Island School of Design and learned just every deep detail about grids and the history of typography. And so that really shaped my point of view.

Jason Pamental: And what about you, Jason? Was there any crossover here that you guys run into each other prior to this? Give me a little sense of where you started to really get passionate about this. I studied graphic design at Rhode Island College, which is sort of across town from risd. And of course, this is like a West side story thing that's about to happen here. We felt like the poor cousins that lived down the road. But the cool thing was we shared a lot of part-time faculty with risd. So even though it was a small state school, the art department was really pretty wonderful and full of people who were teaching in both places. So we got really high quality faculty and then the people that were there, a lot of the same graphic design instructors are still there. And I've had the chance to go back and teach there a few times.

And I love the school because it makes a really high quality design education available to so many people in the state who would never otherwise have that opportunity. And that's kind of where I got my love of typography. But for the first 15 years of working on the web, there was nothing that I could do with it. So it was just kind of sitting there in the back of my mind. And when Typekit launched in 2009, I had a beta invite from a friend of mine, and it was just a reawakening of everything that I loved about design. And so I immediately started diving into learning about it. And as much as I've loved design and development, I think typography takes that blend of art and science to a whole other level. And so I immediately started getting into trying to deal with the things, just what Julie was mentioning.

Do you need that many typeface? Well, probably yes, but let's be reasonable. And figuring out how do you implement that in a way that doesn't kill the user experience? And I think the more I started to learn about it and write about it and speak about it, just got involved more and more and started finding more and more ways to manage the technical issues. And I think that was really, if there's anything that would set me apart from anybody else in the industry was having spent that much time both in traditional design as well as the technical implementation of it, and really trying to push the boundaries of those things. And the thing that was maybe one of my most satisfying quotes in this article that I just did for Monotype is really kind of a bit of a rant about the advice that good typography is using only two or three fonts.

And I think that that may be true, but it is nowhere near an absolute. It is probably better for the user experience because you're loading fewer font files, but if you take that out of the equation, it's about using the right typeface, the right font at the right weight, at the right time, and that answer is going to be different depending on what you have to say. And so if you're Roger Black designing the cover of Rolling Stone, he has bragged more than once to me about have he used 20 or 30 different fonts on a cover just on the cover of a magazine. Now, if you look at any great magazine, all the Conde Nast publications, they're gorgeous and they use a lot of different variation to convey meaning and hierarchy. And that's what using different fonts gets you. And so it is just another one of those levers that you can use in a really artful way.

And as the technical barriers fall away, then we want to start to bring that role of design back into the publishing process. I mean, I love design systems, but they're all about one size fits all. We can create a design system that will work for many, many things. I mean, just look at Spotify in every language across the planet, but to typeset for that album cover, that's a different concern. And so how can we bring those things together? And I feel like this is still the uncharted territory for us as digital designers to deal with the constraints so that we can bring that expressiveness in and then have another role for design to play in the specific instead of just in the abstract. And that's why I'm just still so excited to go to work every day to try and build things like that. Hey, I love my dogs. So especially in service of helping people find better dog toys, then yeah, I want to be super expressive. So we're working on that.

Juli Sombat: Yeah, looking back at the beginning of my career, it actually seems so logical and straightforward that a person would go from being a huge type nerd to working in design systems because a typeface is a design system. It has to be coherent, it has to look like a family. It has to feel like one thing, and then each character has to do its job and fit that specific need without stretching too far from the norm without leaving the family and looking like something totally separate. So yeah, the systems thinking is all the way there.

Jason Pamental: Yeah.

Chris Strahl: Well, I want to thank you both so much for being on the podcast. This has been super fun, very educational. I love having brilliant, passionate people that know way more about things than me. And so thank you for letting me be the useful idiot to drive the conversation today. And I just want to say, this has been awesome. So much, much appreciate you sharing your time, your expertise, and your knowledge. Thank you so much.

Juli Sombat: Yeah, it was a lot of fun. Thank you.

Jason Pamental: Yeah, it was great to meet you, Julie and Chris, you and all the team I've known so many of you for so long, it's great to be on the show.

Chris Strahl: Awesome. Well, thanks. We'll have to have another conversation again soon. Thank you all for listening. This has been the Design Systems podcast. I'm your host, Chris str. Catch you guys next week. That's all for today. This has been another episode of the Design Systems podcast. Thanks for listening. If you have any questions or a topic you'd like to know more about, find us on Twitter @theDSpod. We'd love to hear from you with show ideas, recommendations, questions, or comments. As always, this pod is brought to you by knapsack. You can check us out at Have a great day.

Get started

See how Knapsack helps you reach your design system goals.

Get started

See how Knapsack makes design system management easy.