Allison Horst | R Art Lessons | RStudio

Transcript#

This transcript was generated automatically and may contain errors.

Hi everyone, I'm Allison Horst. I'm an assistant teaching professor at the Bren School at UC Santa Barbara, and over the past year I've also been RStudio 's artist-in-residence. In this talk I'm really excited to share with you some of the motivation behind my R artwork and also some lessons I've learned from the R community along the way.

To get started, I want to ask you to think about what did it feel like for you the first time you started learning R? Maybe you were one of the lucky ones where all you felt was happiness and excitement and you knew this was going to be an awesome tool to learn, but for many people who are learning R or code for the first time, that's not the reaction. There can be some confusion at the idea of learning this new language. There can be some frustration. Maybe you were already happily working in another language and you're looking at R thinking like, what is a kakapo? I want my pandas back. Maybe you were feeling frustrated as you tried to get up this steep learning curve. Or maybe you weren't even expecting to have to learn how to code and you're looking at R just having a meltdown because it's something that's so completely new.

And all of these feelings are pretty common for learners. And this isn't unique to R. A recent Google search for why is coding so dot dot dot reveals that a lot of the top searches related to why is coding so are not totally positive. Why is coding so hard, so boring, complicated, frustrating, difficult, hard to learn, confusing, and stressful. And yes, mixed in here, there's some good ones like why is coding so important? And why is coding so fun? That might be an artifact of this being part of my search results. But the overall takeaway is that for a lot of learners, R and learning to code in general can feel a little bit intimidating and maybe not like a totally happy place.

And not only can it feel a little bit intimidating, but it's also a big energy and time investment. There are so many things to learn to get started in R and with coding in general. And there's always something new coming out. For those of us who are happily working in R, maybe we are this green monster on the right where we see new tutorials and new resources or packages coming out every day. And we are so excited knowing that it can be something useful for us to learn and use. But we can imagine that for learners who are just trying to get up this steep learning curve, seeing all of those new things coming out all the time could be a little bit daunting.

Art as a welcoming bridge

So what can we do when something is hard to learn and we're asking a learner to make an investment? One thing that we can do is we can put in a little bit of extra work to help something that originally doesn't look that appetizing feel a little bit more appetizing. For example, if we know that vegetables contain a lot of really great nutrients, but it's hard to convince somebody to eat them, maybe we can dress it up to look like a cute bug to make it more likely that they will invest in eating those vegetables.

And people are already doing this in code. For example, Julia Evans' zines that explain a lot of really important basics in CSS and Git and Bash are really useful teaching tools because they provide information for learners, but they also make it look very welcoming. These look like a thing that I would want to engage with.

And I found myself wondering when I was teaching courses in R to graduate students, most of who never coded before, do those resources exist in R? Are there nice, welcoming, engaging illustrations that I could include in my course materials that would be a welcoming entryway for graduate students to learn new functions and packages? Because what I found was that I would be up there in the front of a class so excited to introduce an amazing new function. And on the screen behind me would be like a screenshot of the documentation, which contains a lot of really important information that's going to be essential moving forward, but maybe was not the most inspiring first interaction that a student could have with a new function.

Which made me think, yes, there is a place to create engaging artwork that dresses up R a little bit to make it look more inviting and welcoming. And over the past year, that has been my entire goal as RStudio's artist-in-residence, has been to create useful and engaging illustrations that welcome users to explore, teach, and learn about new packages and functions in R. And they will still have to go on to eat their vegetables and learn those hard skills, but the first interaction could be something that is fun and positive and welcoming.

So for example, instead of their first view of Mutate being a screenshot of the documentation, maybe instead their first interaction with Mutate could be seeing these helpful X-Men monsters adding a new column to an existing data frame. And over the past year, I've not only been creating many illustrations as a welcoming interaction with different functions and packages in R, but I've also created illustrations showing general data science principles and concepts, and also a whole bunch of hex stickers.

Lesson one: cute and credible coexist

But what I want to talk about now is what I've learned from the R community over the past year. And these have been really important lessons for me as an artist and as a data science teacher. And the first thing that the R community has helped me to embrace is that cute and credible artwork coexist. Just like Bailey the Golden Retriever is sitting here looking totally credible and professional while he develops his R package, he also is looking very adorable. And that can happen with artwork too.

And even though I knew this, even though I knew that making and including artwork could be really useful for effective communication, making slides more engaging, connecting to learners and clarifying concepts, showing empathy, lowering learning barriers, I just really love doing it. And also including artwork into course materials can really help students understand that yes, creativity and artwork have a place in science. Even though I knew those things, there was always this little gremlin on my shoulder saying, okay, you can lean hard into the artwork for R and data science education, but just know that that might actually damage your credibility as a data scientist.

And there was a turning point when I realized that this was totally unfounded. And that was when Hadley Wickham asked me to create a series of illustrations for an upcoming talk he was giving called the Joy of Functional Programming. And I was so excited at this, I knew this was going to be like a big talk, a really awesome opportunity to share artwork. And I was expecting that because this was a highly professional, important, advanced talk that that this was going to be maybe some sleek designs I was creating to go along with the slides. And to my absolute delight, Hadley, when I asked him what style he was thinking about, responded that he was thinking about these monsters being included in this series. And I tried to play it cool and be like, oh yeah, of course the monsters would belong in there. But in my heart, I was just like, oh my gosh, the monsters have made it.

And, and this is like cute artwork being invited to be at the top level talks in R and data science. Not only were the monsters invited, but they were going to be baking cupcakes in this series. So really capturing for me, this idea that cute and credible are not at odds with each other, that including cute illustrations can be a really powerful teaching tool at any level.

cute and credible are not at odds with each other, that including cute illustrations can be a really powerful teaching tool at any level.

I think that one example that captures this so perfectly is the certified trainer badge from RStudio, which is one adorable, but two is the badge that's used when the certified trainers get their certification, which are some of the top educators in the R community. And this has been a theme that has been reinforced with pretty much every interaction I've had, both with RStudio and in the R community is that there's a refreshing lack of ego about what educational materials look like. And everyone is always just welcoming and encouraging more approachable educational resources.

What that means is that while I started with this little gremlin on my shoulder, questioning my credibility and bringing a lot of doubt into my mind, what I have now is the same gremlin, which is now encouraging me and always asking, can I actually make this more approachable or cuter somehow?

Lesson two: art as a teaching and learning tool

The second big thing that I've learned from the R community over the past year is that art truly can be a useful teaching and learning tool that helps learners engage with and remember new functions. An example of the type of artwork I've been creating is this illustration for the string squish function. The job of string squish is to remove leading trailing and repeated interior white space from strings. And here we see these four monsters who are working together to squish this string and in doing so are ejecting excess white space. This combination of having text and something verbally explained and also an illustration showing what that function is doing, I think is a powerful combination that can help with retention.

An example of this happening in the wild is this illustration that was trying to help people remember to use ungroup if they have used group by previously. We have in the forefront two monsters who are at a birthday party and you can see that one of them has received this amazing gift of group by. But more importantly for this illustration, we can see in the background this other little monster holding ungroup as a gift who is just so hoping they're going to be included. And the reaction to this illustration was exactly what I was hoping it would be, which is an emotional reaction ensuring that people will not forget ungroup.

For example, some of the responses were, I'm so sad for that little bub being left out. I'm always going to remember this when using ungroup. I'm never going to look down on ungroup again. And my favorite one is someone who said they were just able to help someone else solve a problem because they remembered this super cute and somewhat sad drawing. And I think this really reflects that artwork can be a way to engage people's emotions that helps them remember important skills.

artwork can be a way to engage people's emotions that helps them remember important skills.

Art can also be really useful in helping to clarify big or vague concepts. For example, I recently worked with Julia Lowndes on a series of illustrations to try to portray the benefits of working with tidy data and organizing data in tidy data structure. And what this illustration is showing is that if you have datasets that are all in tidy structure, that you can use the same tools in similar ways across datasets. But in contrast, if we have untidy data, what that means is we oftentimes are hacking together unique tools to deal with unique datasets in their own unique ways, which is challenging and tedious and makes it hard to collaborate with people. So I think that this series was really useful in taking some kind of big picture ideas about the benefits of tidy data and capturing them in a way that is more memorable visually.

Artwork can also make things that are kind of intimidating for learners feel a little bit more friendly and provide a more welcoming entryway. For example, for loops usually aren't the thing that learners are the most excited to learn, but they're especially intimidating when it's just like sequences of numbers and the code coming in and out that's a little bit daunting to look at. So what if instead there could be a vector that's actually a parade of monsters? And in going through that for loop, depending on the shape of the monster, they are either assigned to get sunglasses or a hat. And then what we can see is the outcome is the result of each of those monsters going through that for loop and either getting their sunglasses or hat.

Lesson three: art for community building

And the third thing that I've learned is that art can be an incredibly powerful tool for creating supportive communities. As a landscape painter originally, I did not understand how important this could be, but art can be a way to directly show support for learners.

It can show empathy. For example, here we see this little bud who is just needing to take a minute, so it can be really useful for showing empathy when learners are maybe struggling or feeling alone in the struggles that they're facing. It can be a way to engage the community by creating activities that include artwork. For example, with these coloring book illustrations of Our Session Aborted and everyone's favorite Objective Type Closure is Not Subsettable. And it can also just be a really cool way that different groups and communities can represent themselves and make other people feel included. The most common way that we see this is in the many hex stickers that exist that different groups, organizations, and even packages are used to represent themselves.

And I had no idea before joining the art community the power that a 2 inch by 1.73 inch hexagonal design could have. And this is something that I was so amazed by is the support and demand for hex stickers that I think really reflects how excited people are to have something that they can put on their computer or whatever to represent themselves in the groups that they're a part of.

What we can do

So in summary, what I've learned over the past year and that has been reinforced by so many people in the R community is that creative contributions can make coding and R a more welcoming or even just a more fun place for learners and users at all levels. And knowing that, what can we do to create and support more art in R?

And the first thing is we can find ways to support creative contributions. So at whatever level makes sense for you, how can you support artists and creative contributors? Just by watching this talk, you are supporting creative contributions. But you can also include artwork in your teaching resources. You can like or share artwork that's posted on Twitter. Whatever level makes sense for you, I guarantee that your support will be meaningful for an artist and will make them want to create more.

The second is you could think about ways to more formally collaborate with creative contributors. For example, can you contract an artist to add something to your resources? Can you write an artist into a grant that you're working on? Or can you create an artist in residence program where I guarantee you will have your work reflected in a way that's really engaging for people who might not have otherwise known about or engaged with it?

And finally, I would ask you to think about ways that you can be a creative contributor. So what do you do that you might be able to share with the community? We already have this amazing community of people and so many of them are creating and sharing incredible artwork and fabric and are generated designs, 3D renderings, amazing graphic designs made in ggplot and so many other things. And we should think about what other types of artwork or creative contributions would we want to invite? And the answer should be all of them. Baking, music and sculpture, dance. There is someone in the world who can do the perfect interpretive dance of pivot longer and pivot wider and I want to experience that. Memes, design, poetry, art and activities and so much more so that we can be creating artwork and creative contributions that engage and resonate with every different R user who we would want to feel included.

And sometimes I think there's this pressure if you're thinking about what can I contribute, there can sometimes be pressure to like feel like there has to be some learning outcome or somebody has to have a takeaway or a new skill. And I encourage you to remember that making someone feel a little bit happier while they're trying to learn something hard is an absolutely great reason to create and share your art. One experience that I had with this recently was I was having kind of a rough day, I was mope scrolling through Twitter and I happened to come across Jacqueline Nollis's picture of the R pancakes that she had created. And this didn't solve my coding problems, this didn't write my lecture for me, but what it did do is it made me exclaim, R pancakes! And then I just continued on my day a little bit happier because R pancakes had existed in the world.

making someone feel a little bit happier while they're trying to learn something hard is an absolutely great reason to create and share your art.

So to wrap up, no, art is not the solution to tech's problems, especially related to inclusion and welcoming. Unless we're talking about this whale tail sculpture that is literally saving a train that's gone off the end of the tracks, art is not going to save tech, which is why I'm so grateful for the many organizations, groups, individuals, and initiatives, just some of which are shown here, that have been doing this hard work to make R more welcoming for more users since before I even started using R. So while R is not the solution, I hope that it can be a supporting piece that helps those efforts by making the entire landscape of R, which is already so vibrant, even a little bit sunnier. I look forward to talking with many of you about all the different ways we can contribute more creative things to the R community. Thank you.

Allison Horst | R Art Lessons | RStudio

Transcript#

Art as a welcoming bridge

Lesson one: cute and credible coexist

Lesson two: art as a teaching and learning tool

Lesson three: art for community building

What we can do

Featured software#

rstudio