Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Dr. Cedric Scherer Teaches You Everything You Need to Know about R
328 Plays3 years ago
Dr. Cedric Scherer is a graduate computational ecologist with a passion for design. In 2020, he combined his expertise in analyzing and visualizing large data sets in R with his passion to become a freelance data visualization specialist. Cédric has...
The post Episode #212: Dr. Cedric Scherer appeared first on PolicyViz.
Recommended
Transcript
Introduction to Policy Vis Podcast
00:00:12
Speaker
Welcome back to the Policy Vis Podcast. I'm your host, John Schwabisch. On this week's episode of the show, I chat with Cedric Schara, who is, well, he's awesome at R. Let me just put it that way. He's awesome at R. He's awesome at ggplot. He's awesome at sharing how he creates his data visualizations. And so we talk about
Cedric's Early Journey with R
00:00:33
Speaker
Cedric's early use of R, he started with base R. So if you don't know base R, well, he's moved past it. We've talked about his early work in R. We talked about how he approaches sharing his work, how he approaches making things in R, and what he sees going forward.
Future of R in Data Visualization
00:00:50
Speaker
What are the technologies? What are the areas of R? Think markdown, think notebooks. What is the future of R look like for those of us who are working in the field of data visualization?
00:01:01
Speaker
Speaking of working in the field of data visualization, I am working every other week to bring you this podcast.
Supporting the Podcast
00:01:08
Speaker
So if you would like to support the show, please consider heading over to my Patreon page for just a couple of bucks a month, like a cup of coffee. Well, a cup of coffee is a little bit more than a couple of bucks a month these days. But if you would like to support the show with just a few bucks a month, head over to my Patreon page.
00:01:25
Speaker
where you can get some goodies and you can help support the show to support all the sound editing and transcription and web support that's needed to bring the show to you. Or if you'd rather just use a one-time payment, you can head over to my PayPal account. But more generally, feel free and please do share the show with your friends, your families, your colleagues, your networks, whoever you think might benefit from learning more about data and data visualization. All right, having said all that, let me bring you my conversation with Cedric Sherry.
Cedric's Background and Transition to Visualization
00:01:56
Speaker
Hey Cedric, how are you? Welcome to the podcast. How are things? Hi John, everything fine here. How are you? I mean, you know, hanging in there. It's a sunny but cold day here in Virginia. So it's, it's all good. Um, you're in, you're in wear again.
00:02:11
Speaker
I'm in Berlin, Germany, and it's already dark here now. It's getting late. Oh, really? Yes. That's right. That's right. I got to get you up before bedtime. So thanks for coming on. I'm excited to talk about your work on R. We've got, I think a lot of things to kind of talk about. I want to get your story of like how you got into R, maybe some of your other tools that you use.
00:02:31
Speaker
maybe we could talk about like, did you use tools before
From Base R to Tidyverse and ggplot2
00:02:34
Speaker
you got into R? And then as we talk about a little bit more, maybe some of the other tools that you use in and around R. I know a lot of people like, they'll make something in R, they'll pipe it out as like a PDF and they'll bring it into Illustrator to clean it up. And I'm curious about your process on that.
00:02:49
Speaker
So maybe we could just start with how did you get to this point of like being like, I don't know, I kind of like now like you, he was like the date of his guy in R, which I'm sure there are other people out there who are mad at me about saying that, but like, you know, what's your sort of R journey to get to where you are now?
00:03:07
Speaker
Yeah, it's a pretty long journey. Good questions. So I'm not a graphics person for everyone who doesn't know me. So I'm an ecologist by training or biologist first. So when I started looking for my courses, I was thinking about graphics design or biology. And I thought like, okay, biology will give me jobs. Graphics design will not. So I ended up doing biology.
00:03:27
Speaker
Yeah, fun story now.
R's Role in Academic Journey
00:03:31
Speaker
I mean, I still have a biologist job, so that's fine. Yeah, so yeah, I already had a passion kind of for design, not so much for coding actually. And then it was the first contact with I was already in my first bachelor semester, 2008 already, but really like just clicking running code for statistical analysis based plots.
00:03:51
Speaker
So yeah, ours very common or became very common around that time for colleges and biologists to use for analysis. So this was more like really the geeky old R times. Yeah. Like base R times. Yeah. Like linear models all time and base R graphics and people were using it for statistics, mostly statistics. So nowadays R is very different. You can do all kinds of things like you already
00:04:17
Speaker
crossed it, covered it. It's kind of like plots, tables, web pages. You can write books. You can do really fancy things with it nowadays. Back then it was kind of like the, yeah, the tool to use to do statistics in the proper way. So I also learned about SPSS and Excel. So this was basically after that first semester, I quickly switched back because yeah, I didn't learn much in R and most people were not using R with my, in my group. So we were just using Excel.
00:04:44
Speaker
And didn't really like the extra graphs. And then during my bachelor, I was still using Excel. And then during the master, it was more on our courses. Then it became an ecology course studies. And during the ecology field, there's mainly the people driving also many of the R packages and R developments.
Discovering Tidyverse and ggplot2
00:05:03
Speaker
And then I started my master's and I got into it more and more. And then towards the end of my master, beginning of my PhD, I found out about tidy routes and ggplot2.
00:05:13
Speaker
Okay. So, and then it was really, I fell in love and during my PhD, I just realized that I spent much more time on designing than on writing. So actually we had talked to too many people and many people are like, yeah, you're so perfectionist about so many details. You need to drop it. If you want to be a successful scientist, you can't spend three days on the graphic. And I was like, I don't want to drop it. I know I'm a perfectionist, but I want to
00:05:36
Speaker
keep that and make it a good thing and not a bad thing. So yeah, so kind of like morphed after my PhD, I morphed into some more design person. So I'm currently doing both things being academic, but also being a freelance designer or consultant, or I mean, there's so many rights for people. So when was that, that you like, that was like right when Hadley had sort of come up with tidy verse and GG plot too. So that was like right at the very beginning.
Mastering ggplot2
00:06:02
Speaker
I can't really recall. I think ggbot2 is really common since 2015, maybe. Yeah, something like that. Yeah. So, um, I should know actually. Um, anyway, so, uh, yeah, it was, I think 2016 end of 2016. So maybe a bit late. So I wasn't really following the packages. So I met someone and he was completely coding an R everything. He had his Google page and blog posts and.
00:06:32
Speaker
I was just looking at it and was like, oh man, I just know base R. I can't compete. And then we really became friends and we're doing that together. And so he pushed me. And yeah, what, I mean, the thing why I was so interested about ggbot actually, I wanted to do a small multiple, um, so trellis plot and which is called cassettes and ggbot. And so easy to do in base R for everyone who knows base R was so hard to remove although
00:06:56
Speaker
labelings and putting it together in the right format. And that's actually one of the best showcases for GG, but I think the small multiples.
00:07:04
Speaker
Yeah, I totally agree. I mean, whenever I want to do small multiples, I'm going to R to do it because it's like one extra line of code. So which kind of brings me to a related question. So like, is there something about, I mean, I know why I enjoy, and I mean intro baby R coder, but like, is there something particular about R? Because it sounds like you've coded in a couple of other languages too. Is there something about R?
00:07:31
Speaker
In particular, ggplot, is it the philosophy of it that you'd like? Is it just the ease of use? Like what is it about our versus any other language that like really attracted you to using that particular tool?
00:07:43
Speaker
Yeah. So even though I'm programming every day, I don't see myself really like a programming person, like IT programming person. I'm not really, really a computer nerd. I don't know much about hardware. I know much about bash stuff. So it was really also a hard start, but at some point, yeah, I also did a lot of my research or I do a lot of my research just on computational data. So
00:08:06
Speaker
simulation data. So I also picked up a few other programming language, but most of them not for visualization. So a bit of Python, but this is really like, yeah, it's R what we are using. So everyone was using R. So there was basically no choice or I didn't even think about using something else.
00:08:21
Speaker
I was very happy, actually. So I learned a bit of C++, actually, which I found super complicated for me as a non-IT person and dealing with compilers and through repugging. And I had to build my own functions to just get simple jobs done, which was really annoying me. And I think with ours, the perfect combination. I mean, that's why many true programmers, I call them now.
00:08:45
Speaker
don't like R maybe because it's something in between. It's easier to read for non-programmers, I would say, and easier to learn maybe than some of the other programming languages. Yeah, I think that's right. I mean, yeah, it's definitely like there's stuff going on in the hardware. It's back there, but I'm not going to worry about it.
00:09:05
Speaker
Yeah, I think for statistics, let me add something for a statistic. I think it's the, it's the programming language at least was back then to do statistics in a reproducible coding way. So I think also I'm not sure about the state of Python, but C plus plus you don't do statistics. Well, at least I've rarely seen someone using C++ for statistics.
00:09:24
Speaker
Right. I mean, and if you are in like a language, I mean, I tried to learn C plus plus back in college and you know, that's too long ago, but like in Fortran, for example, like if you want to do statistics, you need to code the entire thing. Like it's not like you, you know, it's not like you load a package or type regression Y X and it runs a regression. Like you need to actually.
00:09:46
Speaker
invert the matrix and do the whole thing, which is valuable as a learning process. I remember coding in MATLAB as a statistics class at one point, because you had to invert the matrix. And that's super useful to understand statistics, but on a day-to-day, doing a job is like, I just want to get this done. I need to move on.
00:10:08
Speaker
So what does your workflow look like? Is it simply just bring the dataset into R, do everything in R, export the visualization as a PNG, JPEG, and you're done? Or do you have a whole bunch of things surrounding your workflow? Yeah, it evolved a bit. So during my PhD, I was mostly programming in R, so I wasn't really touching any other tools. First of all, because it should be reproducible,
00:10:36
Speaker
So really clicking the code and then the images returned. So now for my design work, it changed a bit. So I think for these two charts, like academic, scientific figures, and some maps and stuff, you really can go the whole route in R and ggplot. You could also do it for more complex things. I mean, we have shown that with Tye Tuesday, you also have Thomas Mock.
00:10:58
Speaker
in your podcast, so people maybe
Creative Process in R Visualizations
00:11:01
Speaker
know about How You Tuesday. So this is where I really started getting crazy with ggplot, kind of like really tricking the system, like finding ways to manipulate the code to get whatever I want. But it can be tedious at some point.
00:11:16
Speaker
I still do most of the work in R. So it really depends also on the client or on the use case, what we have as a final product. So the really nice thing about R still is that it's reproducible. So if you have something, some project where the data is likely to get updated.
00:11:29
Speaker
I definitely use R. If I have a client who wants to produce thousands, hundreds or thousands of plots every day, every week with updating data, then we need to find a clever algorithm to place the labels nicely. Luckily, there are also many packages for that. We do it definitely in R. I also like the notebooks, like the reporting style that you can directly include it.
00:11:49
Speaker
But if it's really about more artsy, more complex data visualizations, so I don't really bother around if I have a one static graphic and I need to play some annotations, I do it usually in Figma because I don't have really other tools I know. So Figma was now what I started with. So it's really, I'm not coming from a design world, so I'm just learning the
00:12:10
Speaker
I say it's simple. So I think it's not so simple for me because I need to find the point where I stop in R and then move on. And yeah, usually I use vector graphics anyway. So yeah, so I guess 90 to 100%. So it's not really like I've seen Nadi using R and others and they really
00:12:31
Speaker
I like spent maybe 5%, 10% of the time an hour, and then they move on. And for me, it's definitely much more. So the final thing, it's really like already defined the colors, almost defined the colors. Yeah. So you mentioned the notebooks. Can you talk a little bit about that? Because I would suspect
00:12:50
Speaker
that many R users are familiar with ggplot, they're familiar with R Markdown, but I'm curious about like how you think of notebooks and sort of maybe also your view of like what does the future of R look like and maybe that's a broader question of the future of coding.
00:13:10
Speaker
Yeah. Can you sort of walk me through a little bit like when, how you use notebooks in R and in particular, like your particular, your specific process, especially when you're working with a client and maybe you're sharing things back and forth.
00:13:22
Speaker
Yes, the notebooks in R also work with R Markdown. It's basically R Markdown reports. The nice thing about it is that you really can use the R Markdown language or the Markdown language to write text surrounding it. There are pretty neat themings and styles to create your reports. You can also write your own CSS and customize it as much as you want.
00:13:47
Speaker
You can also script it, but I always end up using a markdown notebook usually. It really depends on what you want to do. So even if I'm just drafting one figure for some challenge or some personal project, I'm using these.
00:14:02
Speaker
because it just became my workflow. I also like that I have kind of like a HTML or you could also knit it or render to a PDF, but I really also like that I have like a copy of my code. So if you don't get lost, I still have this HTML template or also you can add the session info to the end. So I know which package versions were used and so on, which is a bit more difficult. If you do it for an R script, for example, then you might want to use some images or so, but it's
00:14:29
Speaker
It's not super easy then to really, years later, to come back to the same setup, but at least I would have the option to get back to a package version which made it possible. I mean, some people write full scientific publications in R.
00:14:46
Speaker
I tried it. I'm not really got lucky with it, but also because, I mean, you need the co-authors on board or some people then kind of like go to word and go back to R, but this kind of like got a bit fuzzy. And also it depends really on data. If you have very big data sets, I still find it a bit of pain if you need to knit and it takes a long, long time.
00:15:04
Speaker
Um, but yeah, I think it's, it's pretty neat and you asked for about clients. So it's the same basically. So you can hide the code and just show the report. You can also have buttons where you can show the code if you want to. So that's mostly my setup. So I'm not showing the code on pre and on default. And then if people want to have a look, they can do it. If the client knows a bit about our, um, but these are very simple to, to share. Yeah. That's the, that's the nice thing.
00:15:27
Speaker
What is your sense, you may not know the answer to this, but what is your sense of people who are writing in our scripts versus people who are writing in Markdown? Do you know what the split is? Or maybe a better question is the folks that you work with, what's the split of people using those two approaches?
00:15:46
Speaker
I think that, well, that's a bit hard to estimate for me. So I basically have two groups. So the scientists are now, those that I know mostly using the notebooks.
00:15:58
Speaker
Um, for like in the, our community on Twitter. And so I see both, but also mostly mostly our markdown, which might be because I'm a tidyverse fan. So, I mean, you may know that there's kind of like this people who use the tidyverse and people who don't use it. And I think people who use the tidyverse are also more likely to, to use our studio and everything. Yeah.
00:16:20
Speaker
provides. So I think even though non diverse user might use the RStudio, they might not use our market on just like the subjective feeling maybe. Right. So I like I like what RStudio is doing. I like these ideas of having notebooks. So I definitely using it. And yeah, I'm also suggesting it to others. So when we have new PhD students starting, I mean, also the younger generations, they anyway learn that now. Yeah, I think it's more about the older ones who just know they're old routines and they don't
00:16:49
Speaker
Yeah, absolutely. Yeah. I mean, it kind of makes sense. I mean, if it's from a collaborative position, that makes sense that maybe scientists are more likely to use a notebooks because they are multi-person teams and maybe it's a little bit easier to collaborate as opposed to like sharing a script back and forth. Even if you're like doing a Git version control, maybe the notebooks are a little bit easier to do that collaboration.
00:17:12
Speaker
Yeah, it's also a collection of your output, right? If I get an academic context, I'm creating also an inclined context. I'm creating like, I don't know, 20, 50, 100 plots, generative things, different designs, depending on the current stage. So it's easy also to collect them and one document can just scroll through and you don't have to open it and find the file again. Yeah.
00:17:31
Speaker
And also in terms of talking about ggplot2 or about visualization with R, the nice thing about these notebooks or about R mark-ons is that you have kind of like settings, how high and wide these images should be. And then you see it the same way because if you are an RStudio user, you know that the plot window where the visualization shows up,
00:17:50
Speaker
It's not what you then really see in the end when you save it. So that's a way to fix the aspect ratio of the graph. So to really see how they look when you save them. Right. So I want to shift gears a little bit and talk about your site, the tutorials on the site, and your book plans.
00:18:10
Speaker
So you've got a ton of our tutorials on your site and what I've noticed you're doing lately on Twitter is I think kind of ingenious is like you seem to like ask a question like you ask a question like what is your biggest challenge of like adding labels in
00:18:28
Speaker
on our plot and you sort of give people like four or five options and then you basically like write a blog post about like the best way to do it. So can you tell folks a little bit about the site and about the tutorials and maybe your kind of philosophy of how you figure out what you want to write about what you want to talk about and like the challenges you're trying to solve for people.
00:18:51
Speaker
Oh wow, big question. You know me, I try to go for the big things. I'm not going to let you off the hook here. So for the one big tutorial I have on my homepage, this was my learning and at some point I wanted to share it. So this was based on a tutorial by someone called Seth Ross and he's also producing ggbot tutorials for free.
00:19:13
Speaker
And as mentioned, I found gtplot thanks to the faceting option and I just jumped in. And as many of us, we kind of just copy paste this code from some random pages, mostly Stack Overflow.
00:19:24
Speaker
And at some point I was like, okay, why isn't this working? Typical problems. And I was like, okay, just give myself one, two days, go, go through the hopeful tutorial and kind of create my own version out of it. And then I updated it. And I mean, as I mentioned, I'm a perfectionist, so I wasn't happy about the default look. So I was also polishing the plus, which Zef already provided.
00:19:45
Speaker
At some point, I gave it a major update working on it every weekend for a few months. And yeah, now it contains, I don't have the exact number, but I think 120 different ggplots. Most of them scatterplots nevertheless, because it's really about like, okay, how do I color the access text on the X and X axis, but also about a bit about plot types. So there's no mapping yet in there. So there are many, many more things to cover. So this is how we ended up or I ended up. Yeah.
00:20:12
Speaker
Thinking about writing a book and people approached me. Um, so the origin idea was we need to, this is kind of like a how to tutorial. So how do I do ABC? Yeah. The idea was to turn that into a book. It now merged a bit with also having a broader context of how to do graphics design with gg pub two kind of like doing these more complex things with all in R.
00:20:34
Speaker
without any post-processing with any of the other tools, Illustrator, Scape, Figma, maybe also a combination of both, how to work with that, but more like not so much on the technical details, but more about the creativity when you're working with code as well, because you also need to be creative to achieve with code what you want to having on your final plot. So it's not only about the creativity in terms of color and chart type and all these kinds of things and story, it's also about the creativity. How can I force ggplot to do what I want?
00:21:07
Speaker
No, I was just gonna say like, so how do you think about a book? I mean, my always concerned about writing like tools books is like the tools change so, especially now they just change so quickly and like packages are always being updated and like today what you might have to sort of make a workaround for tomorrow, there'll be a package where it'll just be like, hey, load this package and you're good to go. How do you think about that as you work through the book? Is it just,
00:21:25
Speaker
Right, right.
00:21:35
Speaker
This is just where we are right now. And, you know, follow along with like the website or this other thing. And, you know, you'll see some updates. Like, how do you, how do you think about that? Yeah, definitely something we discussed a lot. So I'm relying on a lot of extension packages. So, but only those I need. So let's maybe first kind of like the philosophy of how I use extension packages. So I've, for example, I'm not using a package, which allows me to do a dumbbell charts or a lollipop chart because I can do it with gg.2.
00:22:01
Speaker
But for example, if it's allowing me to create a fancy Zenke bump chart, there's a package out there, I will definitely use it because I don't want to code it on my own. What you mentioned also about Matlab, for example, so if it's doable for me and I now have quite some ggplot experience, so I know how to code many of these things without the other packages, so I'm trying to replace them.
00:22:21
Speaker
But there are definitely some packages which are, they are there to be used. Some of them are not even really officially hosted on this thing called CRAN or the CRAN, however you pronounce it. Yes. There's another good question. We'll see. We'll see what the split is on that one. CRAN versus CRAN. I learned it's officially CRAN, but most people call it CRAN. So I'm always saying both, ending up saying both, which is kind of like plain stupid.
00:22:48
Speaker
I'm saying the correct name, no one understands me and I'm saying the wrong name and then everyone's confused. Just showing how nerdy I am.
00:23:01
Speaker
We're talking about the packages in the books. Something we definitely discussed and I kind of made a list which packages I definitely need. We will have an online version. It will be on host on GitHub. So I will kind of like really try to keep it up to date. That's another thing just beside the book. Also, if you're kind of working with ggplotworld in any programming language, things change so fast. So you really need to
00:23:25
Speaker
kind of like, yeah, be up to date. And I feel like it's super interesting to see every few days, something new and to find something new. Yeah. At the same time, it can be also very stressing. So I'm getting older. So I kind of like thinking about what happens when I don't want to scroll through Twitter every day, write new packages. Right.
00:23:45
Speaker
No, but it's a very smart way to go about it, right? Because then it doesn't matter. It's not just like, oh, I know how to make the reader. I know how to make a dot plot now because I follow Cedric's like step by step. I don't need any other packages. I can do it. But I'm sure just learning that helps me do a bunch of other things that I don't also need packages for.
00:24:05
Speaker
Yeah, it's more this idea. I mean, I get many questions. I find it very difficult to build a workshop on that, but I often get questions like, okay, how do you come up with this exact idea? So from the design perspective, I mean, you maybe also get these questions. And then, I mean, of course there are inspirations, but also some things you've just learned and you cannot even maybe say what exactly it is that you see things differently than others.
00:24:28
Speaker
Right. Especially when it comes to colors and font choice sometimes as well, but colors are super hard. I think maybe Lisa helps us now to kind of just say like, buy her Brooke and you will know. And you're all set. And in general, yeah, but also with the coding. So I think as mentioned, I'm not a super coder, so I'm not writing many new functions. So I'm really trying to trick the system just today. I was super happy about some neat trick I did. I needed to have two colors, but I already had
00:24:52
Speaker
the fill and the other color reserved for something else and used kind of like transparency as a third level. And these things make me just happy. Yeah. And this after, I don't know, after six years now into Gigi, but I still kind of like every few days I find something new or a new trick or something. So it's more a collection of kind of like these approaches. How can you approach these things and how could you maybe come up
00:25:15
Speaker
It's something, yeah, which is not really in these usual teaching books, things like, okay, the next step is adding a layer. The next thing is adding a coordinate system. It's more thinking out of the box, maybe. Right. Going down. And so also, I mean, there are many gg-plot books and I also got these questions. I mean, I proposed the book and then also got feedback, something like, yeah, but there's already the cookbook, for example, and will not be like, okay, how do I build a box plot with
00:25:40
Speaker
Um, some jitter and a label with the sample size, it will be more like, okay. Yeah. The way I think GG plot, let's put like this. Yeah. Yeah. More of the philosophy. Yeah. Um, well, it sounds great. Um, I will look forward to it. Um, we'll see if I can give you some time. Yeah. I'll give you some time. Yeah.
00:26:01
Speaker
So before I let you go, I wanted to back way out and ask you whether you think, this is like a philosophical question, because there's always a conversation about this, but I'm curious whether you think everyone should learn to code.
00:26:17
Speaker
No, definitely not. Um, I mean, everyone should anyway, should I, I don't like these kinds of hard, hard rules or hard, I guess, no decisions. Right. Yeah. Should, should anyway. So everyone may benefit from coding, but I also don't think everyone would, but, um, let's, let's focus on our, on our kind of like bubbles. We have like the design bubbles or the analysts and.
00:26:41
Speaker
I think designers could really learn a lot or kind of use it a lot. And I mean, we see it with ggbot now also with d3 and all these other libraries we see now emerging or some Python, there are some interesting ones. I think there's a benefit. I mean, I heard about people who really kind of like.
00:26:58
Speaker
Yeah. Type in the numbers in their design tool. And then they, if they need to update it, they just type it again. Um, so I think of automating some of these things and yeah, I'm coming from the complete coding perspective. So I think at one point I'm missing like, what can I do outside of coding? So because I feel always challenged to do it in code with code at the same side. Yeah. I mean, you do what you feel comfortable with, but I think also many, many things can be more efficient.
00:27:23
Speaker
And I think Martin, Martin Lumberish mentioned it to me in the very beginning, like, why are you doing everything with ggplot2? Yeah. For the tidy Tuesday challenges, for example, this is kind of like the rule, right? There are also people not doing everything in ggplot2, but for me, it was kind of like, yeah, for sure. I code everything in ggplot2. That's the task. So sometimes I was sitting in the night, just three hours to place my labels based on coordinates.
00:27:45
Speaker
Which is then, at some point, I got really tired about that, for example. So I still do it, but if I see someone doing these crazy kind of like, slice a bite in German, like work where you just need to pay lots of time to kind of be perfectionist about it. Yeah, so I think a combination is good, but I think coding, I mean, in a world full of kind of being digital and full of computers and smartphones, I think it's anyway a good idea to maybe deal with these things that
00:28:12
Speaker
are around us anyway. And then in terms of efficiency, I think, and in terms also of honesty, I mean, if I type in my numbers into some design tool, it's pretty hard to check. It's not only pretty hard to update, but it's also pretty hard to check. I mean, if it's more really about reporting, I think this is important to be transparent, to be honest about what's the data,
00:28:34
Speaker
And yeah, we see it also that it becomes more and more standard. I think not only in the scientific community, but also in the business and also maybe in the data design world.
00:28:44
Speaker
So I think that's definitely a good thing. And also, I mean, I love to share and I mean, I wouldn't be here with all the people doing great stuff for the R community and also the database community helped me a lot. So I love to share. I think I can share my codes. It's hard to share something like I clicked here. I clicked there. I need to write a blog post, but here I just share my code. People can pick it up. People can get inspired, reuse it. And I learned from them again and they learned from me. And I think it's,
00:29:10
Speaker
also very neat exchange. Yeah. Yeah. I think all that, all that's spot on. Although, although I fear there may be someone listening to this who is all excited about going to your site and learning R and then heard you just say you spent three hours one night trying to get the labels in the right spot. You're like, nah. Yeah. Those were many labels.
00:29:27
Speaker
And I mean, I'm not talking about labels on the axis, just to be clear, I'm talking about annotations now, right? Right, right, right, right. Okay, so hopefully people aren't too scared off and they'll give a shot. They should check out your site. And then the book when it comes out and there'll be a GitHub site and an online companion probably. So that's all. Cedric, thanks so much for coming on the show. Always good to chat with you. And yeah, take care. Thank you. Goodbye, everyone.
00:29:54
Speaker
Thanks everyone for tuning into this week's episode of the show. I hope you learned a lot about using R, and I hope you'll check out the episode notes to the show. I put in a lot of the links to Cedric's work. His website is a treasure trove of free downloadable information and code that you can use to improve how you create visualizations in the R programming language. Okay, until next time, this has been the policy of this podcast. Thanks so much for listening.
00:30:22
Speaker
A number of people help bring you the policy of this podcast. Music is provided by the NRIs. Audio editing is provided by Ken Skaggs. Design and promotion is created with assistance from Sharon Satsuki-Ramirez. And each episode is transcribed by Jenny Transcription Services. If you'd like to help support the podcast, please share it and review it on iTunes, Stitcher, Spotify, YouTube, or wherever you get your podcasts.
00:30:43
Speaker
The PolicyViz podcast is ad-free and supported by listeners. If you'd like to help support the show financially, please visit our PayPal page or our Patreon page at patreon.com slash PolicyViz.