Resources
Jun 12, 2025 |
54 min
|
877 views
|

Posit Conf 2025 Keynote Previews | Kieran Healy & Jonathan McPherson | Data Science Hangout

To join future data science hangouts, add it to your calendar here: https://pos.it/dsh - All are welcome! We'd love to see you! Thursdays at 12PM US Eastern ✨

We were recently joined by upcoming Posit Conf 2025 keynote speakers Kieran Healy, Professor of Sociology at Duke University, and Jonathan McPherson, Software Architect at Posit PBC, to chat about how and why open-source IDEs like RStudio and Positron get made, how to do data visualization for discovery and explanation, what their keynotes are going to be about, and what’s next for Posit’s IDE development, including AI integration.

In this Hangout, Kieran talked about the trustworthy data visualization. He highlighted that while data visualization is a powerful way to condense and present information, often creating compelling and authoritative artifacts, phrases like "visual storytelling" can be problematic if they encourage presenting a predetermined narrative not fully supported by data. He emphasized that the trustworthiness of visualizations does not come solely from the techniques used or the software, but from a "web of social processes and individual commitments" that cannot be easily automated.

Jonathan talked about the future of Positron and its relationship with RStudio, addressing whether Positron is intended to replace RStudio. He clarified that the long-term goal for Positron is to make it the best Integrated Development Environment (IDE) for working with data in any language. He explained that Positron is built with an extensibility layer, allowing anyone to write plugins for new languages or capabilities, making it a robust and evolving data science workbench. It does not have all of RStudio's features and makes different design trade-offs. RStudio, having evolved over decades, is highly optimized for specific R-based workflows and remains the best at what it does for those use cases.

Resources mentioned in the video and zoom chat:
πŸ”— Posit Conference 2025 Registration β†’ https://posit.co/conference/
πŸ”— Kieran Healy's Website β†’ https://kieranhealy.org
πŸ”— Kieran Healy's book "The Ordinal Society" β†’ https://theordinalsociety.com/
πŸ”— Kieran Healy's book "Data Visualization: A Practical Introduction" β†’ https://socviz.co/
πŸ”— Jonathan McPherson's LinkedIn β†’ https://www.linkedin.com/in/jonathanmcpherson
πŸ”— Joe Cheng’s AI Talk on Harnessing LLMs for Data Analysis β†’ https://youtu.be/owDd1CJ17uQ?feature=shared
πŸ”— TidyTuesday GitHub β†’ https://github.com/rfordatascience/tidytuesday
πŸ”— Positron IDE β†’ https://positron.posit.co/
πŸ”— Will R Chase's talk on making clear plots β†’ https://www.youtube.com/watch?v=h5cTacaWE6I

If you didn’t join live, one great discussion you missed from the zoom chat was about the ongoing debate and practical tips for moving from presenting tables of numbers to visualizations. Community members shared various strategies, including using color-mapped tables as an intermediate step, providing both tables and visuals, and ensuring accessibility and interpretability for diverse audiences. Are you team tables or team graphs?

β–Ί Subscribe to Our Channel Here: https://bit.ly/2TzgcOu

Follow Us Here:
Website: https://www.posit.co
Hangout: https://pos.it/dsh
LinkedIn: https://www.linkedin.com/company/posit-software
Bluesky: https://bsky.app/profile/posit.co

Thanks for hanging out with us! πŸ’›

image: thumbnail.jpg

Transcript#

This transcript was generated automatically and may contain errors.

Hey there, welcome to the Posit Data Science Hangout. I'm Libby Heron, and this is a recording of our weekly community call that happens every Thursday at 12pm US Eastern Time. If you are not joining us live, you miss out on the amazing chat that's going on. So find the link in the description where you can add our call to your calendar and come hang out with the most supportive, friendly, and funny data community you'll ever experience.

Can't wait to see you there. I want to go ahead and introduce our featured leaders today because we have two and they are both keynote at Posit Conf, which is really exciting. So we have Kieran Healy, Professor of Sociology at Duke University, and we also have Jonathan McPherson, Software Architect at Posit. Say hi to everybody. I am going to ask Kieran to go ahead and introduce yourself first. A little bit of background about Kieran and something you would like to do for fun outside of work.

Hi everyone, and thanks for coming. There's a big crowd here. My name is Kieran Healy. As Libby said, I teach Sociology at Duke here in North Carolina. I'm originally from Ireland and I've been using, as I like to say, I've been using R since it was a different letter, since it was S. Back as a graduate student, I was introduced to it by one of my teachers and since then it's been sort of the main way that I do data analysis and teach it and draw pictures with it. And kind of over the years, having just being a user of it, I've become somebody who also writes about it and tries to teach people how to use it specifically with data visualization. So that's my background with R.

Anything for fun? Being Irish, I'm just naturally averse to those kinds of questions. I restore old computers. How about that? I build those things that you can see in the background. I rehabilitate in my basement vintage computers and then occasionally threaten my students with having them run their their analyses on them whenever they complain about running out of memory or something like that.

Does it does it pain you to call them vintage? Because it pains me. It does, yeah, because that means I'm vintage. Right. Yeah. I mean, I'm like, well, these are the things I had as a kid. They can't possibly be vintage.

Wonderful. I'm going to have Jonathan do the same thing. Jonathan, could you introduce yourself? Tell us a little bit about you, how you got here and something you like to do for fun.

Yeah. Hey, everybody. I'm Jonathan. I'm coming at you live from the Seattle area of Washington, where it is still morning. I still need coffee. I'm a software engineer and architect at Posit. I'm not a data scientist, to be clear, but I do make tools for data scientists. So I was fortunate enough to be one of the early employees at RStudio, now Posit, and have worked for most of my time here on the RStudio IDE.

If you have not used RStudio, it is one of the first and, in my opinion, still one of the best IDEs for the R language. A couple of years ago, I switched gears and helped create a new IDE called Positron, which kind of takes a lot of the ideas from RStudio and develops them into a bigger, more ambitious multi-language IDE. So I'd love to take questions about RStudio, Positron, data science tools, software engineering and architecture, and so on.

For fun, I am a dad. I have three kids, so I spend a lot of time with them. I'm into music. I play the piano. I read a lot of books. I go through at least a few books a month. Fiction, nonfiction, everything in between. And I also ride my bicycle a lot, so that's me.

I go through a lot of books every month, too, and then I tell myself, I don't have time for anything. I don't have time to read a lot of books.

Keynote previews

All right. Well, I wanted to give us a chance to sort of preview both of your conf talks. This is putting pressure on you because you might not even have your conf talks done. I have given a talk at conf and I was not ready months before, but I thought maybe we could do a pseudo abstract for each of you just so that it gives us all a little bit more context. Kieran, would you like to go first and talk a little bit about what you plan to talk about in your keynote?

Sure. Yes. And my talk is certainly not done at this stage. My talk will be about trustworthy data visualization. As I said a minute ago, I'm interested in data visualization because it's such a powerful way to condense and present information, often tremendous amounts of information, into sort of a compelling and seemingly authoritative, often beautiful artifact that travels really well, that you can show to people and convince people of things with. And now, in a way, all of statistics, all of the things that we produce with data science models and tables and all the rest of it is like that, too. But visualizations are especially like that, I think, because they strive to be accessible and seem immediately interpretable. I mean, that's what we want when we make visualizations.

But it's that power that also makes me kind of suspicious, a little bit suspicious anyway, of the whole rhetoric that has grown up around them, of their popularity, even though I do what I can to encourage that popularity myself. And as I say, the whole kind of language around data visualization from sort of old stuff like lying with graphs all the way through to more contemporary phrases like visual storytelling and things, I'm just a little bit suspicious of that. We want our data visualizations to be compelling and we want them to be convincing. But we also want them to be trustworthy or we should want that. And so we can ask, how can we do that? How can we make trustworthy data visualizations? And where does our trust in them come from? How is it sustained?

And I think what I want to argue in the talk is that what we find is that the answer is not is not really the techniques used in the visualizations themselves or even in the software used to make them, even though that's made such tremendous strides over the last 20 or even five or six years, these are very useful and important, but they're not sufficient. And instead, the trustworthiness that we want depends on a kind of web of social processes and individual commitments that can't easily be automated and that we inevitably rely on. And if those processes break down or those commitments are weak, then it won't matter how good looking our graphs are or how compelling the stories we seem to tell with them can be. So that's what the talk is going to be about.

the trustworthiness that we want depends on a kind of web of social processes and individual commitments that can't easily be automated and that we inevitably rely on.

It's a very juicy topic for me. I love it, especially because I do Tidy Tuesday every week and having 30 minutes to create a graph, I am frequently going, hmm, well, I got something done, but is it is this trustworthy? Is this good? Especially if the information that I'm conveying is something about like health care, you know, things that are very, very important and not frivolous like Pokemon or D&D monsters, which it is this week.

All right, everybody, rev up your questions for Kieran about data visualization and other things that Kieran has mentioned. Jonathan, how about you? What is your conf talk going to be about?

So my talk is going to be a little bit of history and a little bit of today and then a little bit of tomorrow. So kind of one of those time spanning talks. So like I mentioned in my intro a few minutes ago, I worked at RStudio. I've been at Posit for quite some time, and so I'm going to reflect a little bit on a decade of work on RStudio, kind of the principles that have led it to become the standard data science environment for R. I'm going to talk a bit about how those same principles developed the basis for Positron. I'll talk about Positron for quite a bit, and then I'll probably spend a little bit of time kind of looking to the future and talking about some of the ideas we have for the future of data science tooling.

Audience Q&A

Wonderful. I know that we're probably going to have lots of Positron questions. Well, I want to hop right into Slido questions. We've got a few here already because Noor asked one that is the exact question that I wanted to ask. Noor, not a naive question, as you said in Slido. This is great. Are you available to unmute and ask yours?

Hello. Yes, I am not fit for visual consumption, but my voice is here. This is more directed towards Jonathan, and I think it's a naive question, even though Libby doesn't, but what exactly does a data architect do from your perspective? And what is your day-to-day like?

Yeah, so what is my day-to-day like? So I have worked on Positron since really day one. And so what an architect does in my role is quite a lot of coding and also quite a lot of fitting pieces together. So a system as complicated as an IDE has a whole bunch of subsystems. So if you think about the pieces that compose a complicated data science IDE like Positron, you have a user interface that displays the console and the graphs and charts and data grids and your variables viewer and all that stuff. You have language execution engines like R and Python that are actually performing the computations and then emitting results. You have other pieces that are going on. You have terminals. You have connections to other systems. These systems will often run in a web browser or in a desktop IDE. And so an architect's job in many cases is to simply design the way that all those processes work together to form a cohesive whole.

So in the early days of Positron, I spent quite a lot of time just kind of slotting pieces in and out to figure out how to build a solid, reliable, robust system that does all the things we need it to do. And as the product has matured, my job is to be spending a lot more time actually writing a lot of middleware and glue code that keeps everything talking to each other, as well as writing a bunch of stuff that does individual features on the front end and the back end. So building UI, building capabilities that connect to R and Python, and then writing the systems that make those things talk to each other and work well together.

Noor, I hope that that answered your question. I thought that was a great question. I say it's not naive because I don't know instinctively what a data architect does. And I've worked with data architects who do very different things. I feel like you might have the same title in another company with different functions and do something totally different.

All righty, well, let's move on to Arsenis. Arsenis, you had a great question in Slido. I would love it if you could unmute and ask that. Yeah, sure. Thanks, Libby. Hey. Hey, so thank you, Kieran and Jonathan, for joining us today and chatting with us. My question is more for Kieran on this because you mentioned having used S when R was S, and I like the history of these languages quite a lot. So my question is, what are some of the developments in R that most stand out to you versus kind of the original S that maybe a lot of people might not be aware of or might've missed, and what are some of the developments that you hope to see in the language in the future?

Boy, yeah, there's a lot. I mean, it's like 30 years and more of development kind of that is behind that. One of the things that changed is that you used to be able to use the underscore character as the assignment operator, which made for really terrible readability for code, and thankfully that went away. I think that it's in the character of R as a language to be constantly trying out new things within a sort of more or less structured overall community of the core developers and the wider world. And so in effect, what you have with R always is this kind of like these stratigraphic layers of previous efforts to develop aspects of the language, not all of which hang together very well.

And so that was the main reason, I guess, that sort of Tidyverse and the whole place that Posit eventually can be traced back to, that was a real revolution in the usability of R for kind of people coming to data science broadly construed or data analysis, statistical data analysis for the first time. Not that there weren't sort of strictly speaking, there weren't that many things that you literally couldn't do before Tidyverse tools came along. But the harmonization that it brought really was a big step forward and sort of helped streamline things a little bit with all of the usual kind of trade-offs and caveats that come with that kind of development.

And then the other big development, I think, going back to what Jonathan was talking about in his role, has kind of been in many ways the tooling, right, whether it's IDEs specifically, but then also kind of the ease with which R can talk to other aspects of general computing has really improved a great deal. So I'm not a software developer, although I do write R packages from time to time. And so I don't have extremely strong views on where the language should go or where the tooling should go. That's kind of not my area. But I do feel like the language as it exists now and the tooling as it exists now is just so much better than what we had in the past. And we'd like to see that sort of continue both in terms of its sort of flexibility while maintaining its core strengths, which have always been sort of it's the language that's built for data analysis at its core and builds into it kind of abstractions like the humble data frame that a lot of other languages kind of have to work a lot harder to integrate into how they work.

And so between that and if we can continue on that path and maintain the sort of really vibrant kind of community that R has always had, then we'll be in good shape, I think.

Something that Kieran said when I first talked to him that really stood out to me was something about like, you know, everyone's kind of worried about Positron, like, oh, it changes scary. Right. But when you've seen things progress over a large number of years from S to R, you've seen all of these changes happen. You're a lot less afraid of change because you know that it happens. Change happens. Stuff gets better over time. And it was such a positive outlook.

Yeah, I mean, he's been around long enough. There's after a while you do you do feel like the sort of negative version of that is saying you don't want to turn into sort of a cranky old man yells at cloud. Anytime some new thing comes along, I'm real mad about it.

In almost all of these cases, you know, something comes along and changes. It offers some advantages. Some people will seize on that and take advantage of it. Other people will be like, well, things work fine the way they are for me. So if it's not broke, don't fix it. But there's room within the overall kind of space of the language and of languages and tools generally for exactly that sort of range of reactions. It doesn't mean that kind of, you know, people are going to be sort of forced down one road rather than another. There's generally kind of a lot of heterogeneity that can persist for a very long time, even in the context of kind of innovation and change. Right. And good things are coming always. It doesn't change, doesn't have to be terrible.

What do S and R stand for?

Well, there was a little discussion going on in the chat that I wanted to call out because Zach asked to the chat, like, hey, what do S&R even stand for? Like is S for statistics? I have always thought, yes, S is like loosely for statistics. And for anybody who is curious, my understanding of R, which is reflected in the comments here, is that Ross and Robert were the two guys who were working on this. And so they both start with R. And then, of course, it was S before. So let's call it another single letter. Great question, Zach, for the chat. And also S by analogy to C. Yeah, there we go. Because they come out of Bell Labs. They both come out of Bell Labs. It's C for computing and S for statistics.

Data science informing software development

OK, awesome. If you have any questions for the chat, by the way, that are not for Kieran or Jonathan, do exactly what Zach did, put it in the chat and say, hey, question for chat. All right. Great, great question so far. Let's see, what have we got over in Slido? Alan, you had a great question. Would you like to unmute and ask that?

Oh, OK, I can't, OK, I can't. Oh, there he is. Am I, am I? You're here, you're a robot for a second, but you're better. Good. Yeah, my question for Jonathan. But I also have to say, hi, Kieran, it's been a long time. It's really fun to see. We should catch up.

Jonathan, I'm, I'm curious. We talk a whole lot and Kieran just mentioned tooling. We talk a whole lot about software development, software engineering, informing data science. And we see that in practices and, you know, deployment and methods and tools and stuff. I'm curious if on the software development side, if you see data science practices, methods like informing, making software better. I've started thinking a lot about, like, what are we what are we pushing back into software development that might be interesting that that that that we don't see from the from the data side or from that from the data science side? I'm curious if being at RStudio, Posit, a long time, if that's informing the way you think about software development.

So I'll kind of answer in three parts here, so, so for the first two, I think that there has been limited flow in the other direction, probably less than less than you might hope. But there are a variety of things that are valuable when doing data science that have leaked back in the software engineering world, especially for people who work on software that does data science.

But I will say that like having been part of both communities, like a lot of the back flow that I'm seeing is around like number one around iteration. Right. So like the when you when you develop software, like it is a very iterative process, right? Like, you know, like almost nobody, even in the age of LLMs, can one shot an application. Right. And so one thing you notice when you watch data scientists work is how iterative that work is, like especially when they're doing exploratory data analysis, like that process of like iterating, like testing, like there's a there's a loop here, right, that you're doing as as you iterate on your and explore your data. And that mechanism or like the tools that you use to explore that space are similar to the tools that you use at a philosophical level when you're exploring the software space. It's like if you watch software engineers kind of explore a new space, if they've been informed by data science, you can watch them do kind of the same iterative approach. The second is in reproducibility.

Right. So one of the things that matters a lot to data scientists is making sure that the work they do is fully reproducible. Like this is one of the original goals of doing code first data science is that it generates an artifact, you know, that anybody can use to verify your analysis. You know, and I'm very fortunate to work in the world of software engineering in open source software, where one of the core values we have at Posit is we want to make our software reproducible in the same way that the data science are doing with the software is reproducible. Right. So, you know, we publish the source code to pretty much everything that we do. I can make sure that anybody can kind of build and verify copies of our software. We make sure that the environment in which we build the software is documented and available to people. So those are those are kind of two examples of practices that have kind of crossed the boundary or kind of co-evolved in these two separate communities.

You know, and I'm very fortunate to work in the world of software engineering in open source software, where one of the core values we have at Posit is we want to make our software reproducible in the same way that the data science are doing with the software is reproducible. Right.

And then the third one is simply that, like, as we make software for data science, you know, and I'll probably talk about this a bit more in my keynote, but that that software is informed by the data science community and is built differently than people who are writing software for software engineers. These two communities have some similar needs, but also some that are some that are quite different. So, yeah, not a whole lot of crossover, but a lot of co-evolution and certainly some ideas have leaked over.

Wonderful. Thank you for that great question, Alan.

Moving industry from tables to visualizations

All righty. My next question that I wanted to ask, oh, all my Slido questions just moved one one minute here. All right, Lisa, you had a question for Kieran. Would you like to unmute and ask that? Sure. Lots of you hear me, too. Yeah. Yeah. Hi, Kieran. Great to sort of meet you. I think I've read a book or two of yours maybe on DataVis.

So I'm super curious how to successfully move industry folks from tables of numbers to visualizations. And by the way, I'm not anti-tables. I just think, you know, let's say a 20 by 100 table of numbers. I sort of question what anyone is getting from that. So is the is the problem kind of just habits, you think, or are or is it sort of perceived differences in the information that's being conveyed that that's behind that resistance?

Yeah, that's a good question. I was kind of hoping maybe your sociologist brain would have the answer. Why are they so resistant? I think people keep disagreeing with me that, yeah, I think that there is, well, you know.

There's a lot of. Habit makes a big difference to how organizations work, and established practice is very hard to shift, even for good reasons. And and I will say that, like, you know, one of the so-called, you know, one of the things that I've learned is that, you know, data visualization as a as a technique is is great. And it has, you know, I'm all in favor of it. And I think in a world where we're giving sort of talks with sides all the time, it's much more effective in conveying what it is that you that you want to show than presenting a slide full of numbers, for example, can be.

I I'll say two things about about sort of resistance. One is that because if people have been doing it for a long time, it doesn't just mean that they're entrenched. It also, you know, for sort of negative reasons, it also means that their skills are oriented towards interpreting that artifact like a table, that thing. I remember very, very strongly one of my kind of early experiences being trained just as a social scientist was encountering people, advisors and teachers who could just look at a table like of models or statistical data or summaries and see things kind of immediately in the table because they've been looking at them for years and they would just kind of put their finger like, look, that that coefficient there, what something is wrong with that, you know, that standard error or whatever it is like they would they just had that ability. And it was because they'd looked at like thousands of these things, hundreds of thousands of these things over a very long time. And so then if you're asking people to shift from that habit to sort of some new skill, then that can be that that can be tricky.

And they too were raised in a you know, those people were trained in a world where people weren't presenting to audiences in a it was it was instead, you know, here's here's the here's the paper and a handout maybe that you could sort of look at for a while while the talk was being given as opposed to a world of slides. And so so then tables of data translate quite poorly to slides for that kind of person.

One of the things that I. Emphasize a lot, and it's also something that people don't grasp all that straightforwardly when they're learning data visualization, is that an awful lot of visualization that that we do, in effect. Is structured the way that a table is structured, this is true of kind of small multiples, but also even simple things like like a huge amount of time, we have kind of some sort of we have some sort of continuous measure, a number and some sort of category broken out by some sort of set of categories, whether it's, you know, scores by country or by state or divisions of an organization or sales categories, whatever it is, and then some quantity associated with those. And one of the simplest things you can do that's extremely underrated and very effective for visualizations is to try and turn your visualization into something more like a table that people can read. Often, that's as simple as put the categorical stuff on the Y axis and not the X axis, so people don't have to turn their heads or, you know, to waste your time sort of figuring out how to put the labels at 45 degrees so that they're that they're readable. And instead, put the continuous quantity on the X axis and put the put the categorical quantity on the Y axis. And then suddenly it's just a table again, except that now there are dots or lengths which are easy to to grasp. And so the I think the I think the similarities between tabular layouts and and graphs are easy to underestimate and that one way to get people who are used to looking at tables of numbers to over to the visualization stuff is to help them realize that effective visualizations are often just tables of visual quantities rather than just tables of numbers and are easy to easier to read for that reason.

And so the I think the I think the similarities between tabular layouts and and graphs are easy to underestimate and that one way to get people who are used to looking at tables of numbers to over to the visualization stuff is to help them realize that effective visualizations are often just tables of visual quantities rather than just tables of numbers and are easy to easier to read for that reason.

I love it. There's so much good conversation happening in the chat, if you are watching this or listening to this later and you're not here live in the chat, you're missing so much because we have a community full of really smart and talented individuals who are giving us all of their their takes. Some of my takes are if you have somebody who can't let go of a table, you could do color mapped tables where bigger values have a different color or or something like that. I it's a go between at least. Right. So you can see visually as your values diverge. All right. Awesome question, Lisa. Thank you so much.

Positron's long-term goals

Let's go to Zach's question. And Zach has an asterisk in his. That means Libby, ask it for me. So this is for Jonathan, I think. What? Oh, Jonathan, this is the question you've been dreading. Are you ready? What are your long term goals with Positron and do you think it might eventually replace RStudio?

I did know someone was going to ask this. Thanks, Zach. It's a great question. So I'll answer the first part. So our long term goals for Positron is to really make it the best IDE for working with data in any language. So Positron is built in such a way that the R and Python subsystems that we added to Positron are just extensions. Right. So if you think about the again, I'm an architect. Think about the overall way the Positron is composed. We built it so that anybody can write a plug in for like a new language or new capabilities to Positron. So like Positron as a core is going to have the really like first class, like hopefully best in the world at being like a data science workbench that will be extensible to new languages and new capabilities over time and continue to evolve as the industry does.

So for the second part of that question, is this eventually going to replace RStudio? The answer is probably not. So a couple of things to keep in mind here. So the first is that Positron is not and will never be a superset of RStudio. So it does not have all of RStudio's features. It won't ever have all the features that are currently in RStudio. And it makes a different set of tradeoffs. So like I want you to think about things that are more general and then more specific, like you can get a lot more power and capability if you're willing to make your tool fit for a very specific use. You can also get a lot more breadth for your tool at the expense of making it like a little bit awkward for more specific use cases. And this is one of the things we're really wrestling with as we develop Positron is,

there are people that are coming from RStudio or from VS Code or PyCharm or Dataspot or whatever, and people who are using their tools in a very deep way, it's difficult to make one tool, one generic data workbench that serves all these very specific needs without turning into a very complicated piece of software.

So RStudio is very, very fit. It's evolved almost like a river rock, just smoothed out by decades of work to be very, very good at specific R-based workflows. And honestly, even as someone who has worked on Positron for years now, I don't think Positron will ever be as good as RStudio for those specific use cases, simply again, because you can't take something so generic and make it work as well as something that is fit for a more specific purpose.

So we think if you work for R and Python together or just Python, you'll be a lot happier in Positron because it's so much better at multi-language stuff. It has much deeper ID capabilities because it's built on VS Code. A couple of people in the chat have pointed that out.

And also, we are continuing to invest in RStudio with a lot of the ideas that we have germinated in Positron. So if, for example, if you use a really recent build of RStudio, you'll notice that it has much better treatment for errors and warnings and messages in the console, which is basically derived and pioneered from work that we did in Positron.

Same thing, for example, we're building some tools for R that run inside of Positron, and we're pulling those tools back into RStudio to improve the R experience there. One of those tools, for example, is called AIR. It's an incredibly fast code formatter and server for the R language. And it's native in Positron, but also you can now hook it up to RStudio in a recent release to have it format your R code beautifully every time you save it.

So we're seeing a lot of cross-pollination between the IDEs, which we think is great.

So RStudio is very, very fit. It's evolved almost like a river rock, just smoothed out by decades of work to be very, very good at specific R-based workflows. And honestly, even as someone who has worked on Positron for years now, I don't think Positron will ever be as good as RStudio for those specific use cases, simply again, because you can't take something so generic and make it work as well as something that is fit for a more specific purpose.

Audience Q&A: Starting with answers and visualization pitfalls

Well, it's Williams iPad, I believe. Okay, he's there. I see him with his awesome shark background as usual. Okay. I did just ask to unmute Bill. He's working on it. He's on zoom on the iPad, which is a little bit trickier. We have to ask. Oh, there we go. Sorry about that.

And Karen don't feel bad about us. I used APL which was done in the 60s. So, yeah, it's okay.

My question really revolves around the way people approach things in terms of the visualizations and the data that they're doing in that, you know, if we did it the way we should characteristically as a data science group, we let the data tell us what it is, but what I've seen recently and recently being the last few years people are starting with answers. And typically, well, that's not an evil thing to do. It slants how everything goes particularly the visualizations. And are you seeing that with all the work that you've been doing?

I think that well first I have I have used APL a tiny bit because one of my former colleagues Ron Breger used to use to use it. Imagine a language for those who don't know it consisting entirely of operators. It's amazing, amazingly compact, but it's one of those write once read never languages.

I think the thing you identify is a real issue for data analysis generally. I think it's kind of a chronic issue in the sense that there's no sort of sin you can commit with graphs or with data visualizations that you can't also commit with tables or models if sinning is what you want to do.

And I say, you know, and it's not something that like it's a feature, but it is something that's kind of can tend to be supercharged by the power that data visualization gives, although in much the same way again that it can be supercharged by the immense sort of power of the data analysis tools that we have generally.

I think there let me let me pick out sort of two reasons that that sort of thing happens across both visualization and statistical modeling generally.

One is that the most general one is that when we're looking at data, we're looking for patterns and humans are pattern recognizing creatures. We're very good at finding patterns and we're often motivated in data visualization to find patterns. We want them to be there whether that's because we're an academic wanting to write a paper that has an interesting finding or that you're in industry and you're looking for looking for something that you want to sort of pitch or explain or like you want to find that pattern and and half of the reason that statistical techniques exist is to sort of slow your role a little bit and give you criteria that are not just kind of I think this looks like it's real and and so we really do need to that.

That's why when I said at the beginning that I was a little suspicious or it tended to make me a little bit queasy that phrases like storytelling with data kind of, you know, bring me out in hives just a little bit is because they tend if you misunderstand the intent behind them, they can encourage you to just kind of well, I've got a yarn. I want to tell and I'm just going to look at you know, for the things that help me tell that yarn and a compelling way and a graph or a table is something that you know, or a fancy moss fancy statistical model is something that can help you tell that yarn even if it's not supported by the by what but by the data in there and so data in that sense never tells you what it is all by itself. It our jobs would be a lot easier if the data just transparently represented itself to us in a truthful manner.

So we're committed to finding things and to look to look for things but you know, we have to sort of we have to be constrained in some way and then the second thing I would say is that one of the things both again both in data visualization and I think in the whole business of the whole world of modeling data statistically on the whole it that we see at IC again and again is out in the world and in classrooms with students and wherever else is that is that there's a again understandable but real tendency for people to say to themselves to begin by saying like what's the most complicated thing I know how to do and and and start there like I'm going to do that whether that's some fantastically complicated, you know multidimensional graph that that they can write ggplot code to produce because it has the capacity to do it or it's some fantastically complex statistical model that maybe you're not quite sure how to interpret but you know that that's the thing that you're supposed to in some sense supposed to do people often go immediately for that like that for I want I want to find a pattern and I want it sort of validated by the most complicated or technically sophisticated thing that I know how to do and very often the advice that you are warning that I end up giving people a lot of the time get people sort of want to resist it but because they say look I want to do the fancy thing I want to do the hard thing why are you telling me just look at the simplest thing first and then see where I should go next and so it's those kind of you know it's that sort of advice that you end up trying to trying to get across to people that often simpler is better or you know don't don't overextend yourself too early when you're looking at data and that and that's true for any kind of analysis no matter how you're presenting it but I think it is true that it's something that you know complex or beautiful visualizations can and can encourage and and can make it out the door because they're because they look so so great visualization is maybe sort of more prone to those kinds of problems than other approaches but but it's not unique to visualization by any means.

So we're committed to finding things and to look to look for things but you know, we have to sort of we have to be constrained in some way and it's that sort of advice that you end up trying to trying to get across to people that often simpler is better or you know don't don't overextend yourself too early when you're looking at data.

awesome thank you Karen all right we have less than 15 minutes left because somehow 45 minutes has flown by so let's move on to another question and Mike you had a great question would you like to unmute and ask that

AI integration in Positron and balancing power users with beginners

yep so I hope people can hear me this is question for Jonathan I wanted to first of all say thank you for all the developments in our studio IDE over the years and you know things like our markdown notebooks have really transformed many people's ways of working and so what if I'm allowed a kind of like have two-part question first part is what are you most excited about that's coming in either our studio IDE or positron and also the second part is how do you balance features for power users and ease of use for beginners

those are the two great questions so I'll talk first about what we're most excited about and I almost hate to say this because everyone's excited about this right now but we are excited about AI integration quite a lot one of the things I've been working on it more than anything in the past couple of months is integrating a an AI assistant into Positron so we have not really shared or demo much about that yet but you'll be you'll be hearing it quite a lot about it in the months to come there are there are so many places in an IDE where an AI can provide like really good feedback if you if you've ever used like you know github copilot or cursor or like windserver into these in these IDEs as like as these tools are primarily used today by software engineers, but the really cool thing about using them in an IDE like Positron is that they have that is that you get they have access not only to the like that the state of your project in terms of code, but also the state of your data as I can as I can in memory object that can be manipulated with the result that you can get some like really like really quick robust analysis going by kind of describing what you want in natural language and kind of working with the AI kind of hand in hand to to arrive at a result.

I don't want to get too off into the weeds there, but there's just like quite a lot of super exciting innovation happening today in the at the nexus of artificial intelligence and the IDE like not not like we're not that interested in like replacing, you know human activity like we don't you know, we're not going to create like some like an agent that you just like run and like let it go, but we are quite interested in you know, AI assisted operations in the IDE and there's a lot of a lot of places where that can be really helpful things like like asking the AI to explain errors like asking it to like, you know, modify your charts when you forget the ggplot syntax, which I do all the time.

And like asking it to like, you know, help you find outliers asking it to you

know, like help you generate analyses like reports that there's lots of great stuff.

So second part of the question is about making an IDE that works well for both power users and beginners and I'm going to say that there's actually you know, this is actually very hard to do.

And at the end of the day, we basically we kind of compromise, you know, that you can't make an IDE that's perfect for power users and beginners. So the best thing that we can do is to try to like keep the complexity in a place where you know, it doesn't show up unless you need it, you know, I've in these systems, there's usually what I call like a conservation of complexity, which is to say that like the amount of complexity is constant. It's sort of like the amount of energy in the universe. Like there's there's just going to be complexity and you know, you have to decide, you know, where that complexity goes and you know, how it's accessed and to like to the best of our capability. We take that complexity upon ourselves and try to push it like into the IDE as opposed to exposing users to it.

We take that complexity upon ourselves and try to push it like into the IDE as opposed to exposing users to it.

You know, I think, you know, I think VS Code like does a like the trade-offs that they've made with respect to this are largely gated toward power users, right? So they pretty much hide everything, you know, like they don't have a lot of menus or toolbars, like it's a very Spartan aesthetic and you access most tools using the command palette, you know, when people come over for something like RStudio, which is a little bit more geared for like beginner use, people who are not accustomed to software development tools, it's got lots like menus and buttons and you know, other affordances for getting your work done.

So if you look at the design of an IDE like Positron, you'll find that it lands somewhere in the middle. You know, we've added a lot more affordances for things that beginners are accustomed to while still like preserving like the full power of the command palette of like a rich set of commands for advanced users that maybe don't necessarily have all the buttons that beginners are drawn to. So it's a trade-off. It's not perfect for everyone, but we try to make sure that that complexity lands in the right place.

Awesome. I was just chatting with Mike about this in the chat where he was saying like, could you have an RStudio IDE or Positron skin that sort of like hides complexity for newbies?

And I was just thinking, oh yeah, Positron was really overwhelming when I first started using it because I was not a VS Code user before that. I had never really liked VS Code. I'm like, I don't want VS Code. I want something different. Why did we do this? Why is it still VS Code? This is the worst.

But then I remembered that RStudio was also really overwhelming when I first started using it. It's got a billion panes. There's a bunch of buttons everywhere. Every pane has like 50 tabs on it. I'm like, it was so much.

So just remembering that everything is hard when you're bad at it. Everything is tough when you're new to it and it gets better over time. You just have to work hard at it and lean on people who can support you.

Fantastic question, Mike. Thank you for asking as always.

Visualizing your way into data analysis

And then I saw a question. I know we only have five minutes left, but I did like this question so much. It's from Simon. Simon, if you are available to unmute and ask this, that'd be great. It was a question for Kieran. Sure. Yeah, thanks Libby.

So Kieran, I read your book, Data Visualization, a few years ago. It was handed to me by a coworker who we've been trying to onboard folks to using R or Python scripting languages to build data visualizations and open that up as like the broader world of data analysis. And so I liked your approach as like, hey, let's start people with plotting so they can get to see these results. And so I wonder if you still like that approach for getting people into the world of analysis.

Thank you. Yeah, I do. Yeah, I'm in the middle of writing a second edition of the book that's trying to sort of do the same thing again. One of the reasons, this is kind of right at the center of kind of things Jonathan was just talking about it and Libby too and the questions.

There's, if you're learning how to do this stuff with data science broadly construed, then you have to learn how to do statistics. You have to learn the idea of representing stuff for public consumption in tables or graphs. You have to learn a computer language, programming language to do that in and you have to learn some sort of IDE to do it with and that's just tremendously overwhelming for newcomers. And it's very hard.

The flip side to what Libby was saying about how everything is hard when you're bad at it is for experts, it's what sort of the psychologist called the curse of knowledge, right? That once you know something, it's very hard to imagine not knowing it. And that makes things like sort of getting people to, onboarding people to new tools for people who already know how they work. It's very difficult to teach them, to teach new people because you forget yourself that, you know, what it's like not to know something.

Data visualization is really handy as a way in because it can help you get past this feeling that you have to know everything before you can do anything. In coding. And so I really do feel like the combination, in R specifically too, the combination of pipeline approaches to coding, first this and then and then and then and then, which leads naturally to sort of thinking in terms of functional programming, but you don't know that that's what it is first, combined with the idea of kind of doing a series of minor small analyses that emit a graph you can see immediately and understand and interpret is a really powerful way, right? You know, to get people, you know, up the steepest part of the learning curve by because then you can revisit. Okay. What is this? What does this analysis mean? Or what is this? What can we do? That's more complicated and so on, but all in the context of, you know, just giving people the reward of saying, oh, look, I produced this thing that you can actually show to somebody else and it looks good and it's understandable and it makes sense.

I think I think for that reason, data visualization is a very handy way to introduce people, not just to, you know, the business of data analysis, but to the whole ecosystem that's required now to do, you know, the whole enchilada of IDs, code, software, and the file system and all the rest of it.

I think I think for that reason, data visualization is a very handy way to introduce people, not just to, you know, the business of data analysis, but to the whole ecosystem that's required now to do, you know, the whole enchilada of IDs, code, software, and the file system and all the rest of it.

Yeah. Yeah, as Mina says, let them eat the cake first, right? Like let them get the beautiful thing first. Well, thank you so much, Karen. That was an amazing answer. I totally agree. I think that teaching plotting is like the first step to everybody understanding even what the concept of tidy data is.

Lightning round: career advice

I have two minutes left on the clock. I want to ask each of you a lightning round question, which is one piece of career advice that you have found valuable or that you think other people should know and do it as quick as possible because we have two minutes on the clock. Jonathan, hot seat.

Do you have anything on your mind? For a career advice specifically, career advice, I think, you know, the just being humble and curious is the best advice I ever got, you know, that's it's very simple, but like being humble and curious and just like nice to people that you work with, it sounds so simple, but it's very powerful.

Fully agree. Couldn't agree more. Thanks, Jonathan. Karen, how about you?

I would just I would just say that an elaboration of the curse of knowledge thing that I just said, once you know something, it's real hard to imagine not knowing it. So don't forget that people who don't know things don't know it yet. And when you're trying to explain something to somebody or learn it yourself. Or learn it yourself. Yep, be kind to everybody and yourself.

And I think that this translates to if you built it, maybe don't write the docs for it because you know too much about it and your docs are going to be Swiss cheese and full of holes.

All right, everybody. Thank you so much to Karen and Jonathan. This was so much fun having you both. I cannot wait to hear your keynote talks at PositConf and everybody. I'm so excited to see you today. Thank you for being in the chat. If you would like to save the chat and you're still here. There are three dots in the top of your chat window where you can click that and then click save chat that will let you save all of the resources that everybody has shared and we hope that you stay with us next week to next week. We're going to have Toby Hall EVPNCIO at Delta Dental of Michigan. Thanks Karen and Jonathan.