posit::conf(2024) agenda is now available!

TBD

TBD

TBD

TBD

For the last 20+ years, I’ve been reading data into R. It all started with the humble scan() function. Then, I used fancy new-fangled file formats, such as parquet and arrow, before progressing onto trendy databases, such as duckdb, for analytics. Besides the fun you can have by messing around with new technologies, when should consider the above formats? In this talk, I’ll cover a variety of methods for importing data and highlight the good, the bad and the annoying.

Posit provide an amazing set of products to support data science, and we will learn about many great packages and approaches from both Posit and the wider community at posit::conf. But underlying all this are a number of open source tools, notably R and Python. How can we contribute to sustaining these open source projects, so that we can continue to use and build on them?

In this talk I will address this question in the context of the R project. I will give an overview of the ways we can contribute as individuals or companies/organizations, both financially and in kind. Together we can build a more sustainable future for R!

We hear a lot about creating R packages, but R packages don’t last forever on their own. I describe my experience resurrecting rvertnet, an abandoned ropensci project that had become stale on CRAN. I talk about how I found out the package needed a new maintainer, how I took ownership of the package, and how I decided what needed fixing. I discuss several examples of package repairs I implemented, including fixing outdated CI, removing unnecessary files and dependencies, writing workarounds for deprecated functions, and fixing building of a vignette. Finally, I’ll describe my positive experiences communicating with the old maintainer and submitting a package to CRAN for the first time.

The blessing and the curse of open-source software is that it lacks the infrastructure of a corporation. It can often be difficult to ensure that projects have stability and longevity.

In this talk, I will discuss ongoing work on an NSF Pathways in Open-Source Ecosystems grant focused on the {data.table} package. Like many R packages, {data.table} has incredible functionality and thousands of users - but no cohesive community or governance structure to support it long-term. We are working to build this ecosystem.

I will provide my advice and insight for key aspects of a sustainable open-source project: Engaging casual users, supporting developers, generating content, emphasizing education, and creating a home base for the community.

As a global non-profit established in 2016, R-Ladies has more than 100k members from 233 chapters in 63 countries to support the mission of increasing gender diversity in the R community. Empowering local chapters is challenging as accessibility and awareness of communication methods, software choices, social platforms, and support avenues varies internationally. Join us for insights into our journey of developing a global technical and social infrastructure while fostering collaboration and growth and granting chapters the freedom to tailor their activities to local contexts. Walk away with practical technical and social strategies to empower and diversify your own data science communities based on learning from continuous feedback.

Everyone will have their own recipe for a bRewing a great collaboration, but we wanted to share ours. Ingredients: equal parts learner and teacher, 90 kg of supportive management, 1 whole database, complete or incomplete, a dash of creativity, 60 hours of time (recipe included in main presentation), fun to taste. First, make sure your ingredients are organized and prep area is tidy. Sift data into central database and simmer and stir into separate r scripts. In large cauldron, combine scripts and narrative into one giant Rmarkdown. Lubridate your pan and knit into desired format. We want to share the rest of our recipe to make a delicious report that builds confidence in the learner, new and strong friendships, and lifelong skills.

Building on my experience writing books with Quarto for various audiences (R learners, statistics learners, and Quarto learners), for various venues (self-published and publisher-published), in various formats (HTML books hosted online and PDF books printed), in this talk I will share best practices and tips and tricks for authoring reproducible, dynamic, and elegant books with Quarto. I will also highlight a few features from the recently-released Quarto 1.4 that pertain to books (e.g., flexible and custom cross-references, embedding computations from notebooks, and inline code in multiple languages) as well as share examples of how to make your web-hosted books more interactive with tools like webR and shinylive.

Quarto is a game-changer for creating reproducible, parameterized documents. But the beauty of Quarto—that it has so many different use cases with various output formats—can lead to disarray with numerous .qmd files floating around an organization and too much copy-paste when creating something new. Quarto templates are perfect for easing the burden of developing a report and instead standardizing the structure, style, and initial content of your projects, no matter the output format. We’ll discuss tips and tricks for implementing enough html and css to create beautiful documents that match your organization’s branding and also explore how easy it can be to deploy those Quarto templates with a single function within an internal R package.

Scrollytelling is a style of web design that transitions graphics and text as a user scrolls, allowing stories to progress naturally. Despite its power, scrollytelling typically requires specialist web dev skills beyond the reach of many data scientists.

Closeread is a Quarto extension that makes a wide range of scrollytelling techniques available to authors without traditional web dev experience, with support for cross-fading plots, graphics and other chunk output alongside narrative content. You can zoom in on poems, prose and images, as well as highlighting important phrases of text.

Finally, Closeread allows authors with experience in Observable JS to write their own animated graphics that update smoothly as scrolling progresses.

In this presentation, I’ll confront the traditional dependence on dashboards for business intelligence, pointing out their shortcomings in delivering prompt insights to business professionals. He proposes a shift in strategy that employs Python and R to generate dynamic, customized emails, utilizing Quarto and Posit Connect for seamless automation. This technique guarantees direct and effective delivery of actionable insights to users’ inboxes, enhancing informed decision-making and boosting engagement. Sean’s recommendation not only redefines the method of data delivery for optimal impact but also prompts a fundamental change in mindset among data practitioners, urging them towards a more engaged and individualized form of data narration.

Tired of WordPress’s endless updates and security headaches? Want to spend less time on server administration and more time with friends and family? I found freedom by switching to Quarto with R’s help! I’ll show how I used R to automate the transformation of complex WordPress sites—custom JavaScript, styles, and content—into clean Quarto markdown. Additionally, I’ll demonstrate how I enhanced my site using Quarto, and how switching has vastly improved my publishing workflow. This talk will show attendees how this decision can streamline and improve your blog or website: enhancing speed, improving security, and minimizing site management.

In clinical trials we work in interdisciplinary teams where the discussion of outputs is often facilitated using static documents.

We wanted to bring the advantages of modern tools (R, markdown, git) and software development practices to the production of company documents.

We used an object-oriented approach to create classes for report items with a suite of tests. Finally, the report is rendered programmatically in docx format using a company template. This enables our statisticians to work in a truly end to end fashion within a GxP environment with the end product in a format suitable for interdisciplinary collaboration.

We are currently piloting this package with our internally before we release in the open-source community.

Traditional medical research dissemination is slow and cumbersome, often culminating in a diverse array of outputs: reports for our sponsors and regulators, manuscripts for peer-reviewed journals, summaries for online platforms, and presentations for conferences. However, it takes a great deal of time and effort to organize all these outputs so that our findings can enter the patient setting. Quarto can change that. It’s a tool that lets us efficiently create various polished formats from a single source, while meeting diverse submission requirements. This talk will showcase how Quarto can revolutionize our communication, making research more impactful and speeding up the delivery of treatments to our patients.

In my work I am often asked to develop interactive data explorers for public-use data sets, with an emphasis on making the tools engaging, easy to use, and understandable for a general audience. I’d like to talk about the work my team does to develop user-friendly Shiny applications that look and feel like full websites and share some of the tools we use. This includes things like designing landing pages, creating detailed About pages, letting users share links to specific charts or download static versions, adding social media sharing links, site meta tags, and sub-URLs, and so much more. After attending this talk, I hope others are excited about leveraging tools to make their users say Wait, that’s Shiny?

The Shiny team has been working to help apps’ UI and UX scale more thoughtfully and elegantly. We are currently working on Shiny Templates, which are opinionated boilerplate code that bring our refreshed component library and layouts together to help users create small, simple apps or large, complicated, multi-page dashboards that just look right.

What data processing in your Shiny app is redundant or must happen within the app at all? What makes Shiny beautiful is how it blends data visualization into a compact bundle of code. That said, there are challenges to overcome to get from a developer’s console to users’ screens. Tools like Posit Connect help with this process, but as an app matures, developers and their users may encounter different performance issues. To address this, sometimes this means evolving and introducing separate data pipelines. In this presentation, we will overview some different types of scaling issues for a Shiny app. Additionally, we will introduce the new package {matte}, which provides support for adding data pipelines to your app that live outside Shiny.

This talk explores the creation of an advanced portfolio analysis system using Shiny and Plumber API. Focused on delivering real-time insights and interactive tools, the system transforms financial analysis with user-centric design and reusable Shiny modules. The talk will delve into how complex financial data is made dynamic and interactive via an internal R package integrating with an ag-grid javascript library to enhance user engagement and decision-making efficiency. A highlight is the Plumber API’s dual role: powering the current system and hosting other enterprise applications in other languages (python), demonstrating remarkable cross-platform integration. This system exemplifies the innovative potential of R in financial analytics.

Shiny Express is the easiest way to get started with Shiny. It’s a new syntax for writing Shiny apps, one that trades structure for minimalism. It’s designed to make Shiny dramatically easier to learn and faster to write, yet is still suitable for writing everything from throwaway prototypes, to realtime dashboards, to cutting edge model demos, to production-quality business workflow apps.

In this talk, I’ll introduce the syntax of Shiny Express and compare and contrast it to the traditional way of writing Shiny apps (which we now refer to as Shiny Core).

I hope to inspire any R or Python data scientist who’s been putting off learning Shiny, to finally give it a try!

Integrating editable data frames into Py-Shiny and Shinylive applications streamlines data scientists’ workflows by allowing real-time data manipulation directly within interactive web applications. This new feature enables users to edit, copy, and paste cells within the data frame output, facilitating immediate analysis and visualization feedback. It simplifies the process of data exploration and hypothesis testing, as changes to the data set can be instantly reflected in the application’s outputs without the requirement to update the original data, keeping data-scientists scientists, not data-janitors.

Building ML and AI apps with Shiny for Python

You probably already know that you can use Shiny to build interactive web applications like data dashboards and data analytics tools. Did you know that Shiny is also a great platform for building interactive machine learning and AI tools?

In this talk I’ll show how Shiny can be used to build the following kinds of applications:

• Interactive model training
• Displaying model inference results
• Chatbots

We’ll see how to build these applications quickly, easily, and with a minimum of fuss.

Most Python packages that provide interactive web-based visualizations (e.g., altair, plotly, bokeh, ipyleaflet, etc) can render in Jupyter notebooks via the ipywidgets standard. The shinywidgets package brings that ipywidgets standard to Shiny, enabling the use of 100s of Jupyter Widgets as Shiny outputs. In this talk, you’ll not only learn how to render Jupyter Widgets in Shiny to get interactive output, but also how to leverage user interaction with widgets to create delightful and bespoke experiences.

Tables are great, and we’ve been doing a lot on both the R and Python sides to make it possible to generate aesthetically pleasing tables. The gt package for R has been under continuous development for six years and there is still so many things we can do to make it better. Great Tables, our new Python package, brings beautiful tables to Python users and provides an API that’s in tune with that ecosystem.

While we have had made great strides and unlocked new table-making possibilities for our users, our ambitions our huge! So, we’d like to show you the state of things on this front and we also where we intend to go with our collective table efforts.

The quality of data science insights is predicated on the practitioner’s understanding of the data. Data documentation is the key to unlocking this understanding; with minimal effort, this documentation can be natively embedded in R data frames via variable labels. Variable labels seamlessly provide valuable data context that reduces human error, fosters collaboration, and ultimately elevates the overall data analysis experience. As an avid, daily user of variable labels, I am excited to help you discover new workflows to create and leverage variable labels in R!

The gtsummary R package empowers researchers and analysts to create publication-ready summary tables efficiently. Developed at Memorial Sloan Kettering Cancer Center, it quickly gained traction and has become the most downloaded package for summary tables on CRAN.

2024 marked a significant expansion for gtsummary. A comprehensive codebase update enhanced performance and introduced new features. Further, the adoption of CDISC’s Analysis Results Data standard enables compliance with emerging FDA submissions standards, maintaining relevance for various research and regulatory needs.

gtsummary offers a robust solution for generating clear, informative tables, saving time and ensuring quality for researchers and analysts across diverse fields.

In the world of crochet, the Woobles kit simplifies yarn, hooks, and stitches for beginners, alleviating decision fatigue and fostering early success. This model unexpectedly extends to the domain of R. Newcomers, especially in industries less familiar to open-source, may find mastering new tools daunting. We grappled with this while developing {tfrmt}, a table-making package for pharma. This talk will draw parallels with crochet to explore strategies for engaging and retaining new users. Attendees will grasp the role of a starter kit for easing the learning curve and the value of nurturing experts with transferable skills. They’ll glean insights to support their own audiences, whether in creating an R package or crafting a cuddly unicorn.

The rise of generative AI is fundamentally changing how we learn to code. At Dow, we’ve had nearly 200 learners participate in Posit Academy to learn R or python and apply it to their work. As coders embrace these new tools, we are witnessing a before and after moment. This talk will share real-world examples of how researchers at Dow are learning by using code generation, highlighting the most effective tools including like copilots and chat agents, to grapple with the challenges and opportunities of learning to code in this transformative era.

Machine Learning involves a lot of math, and a lot of code. But it can also involve LEGO, coloring sheets, and 3D Printed gradients! Hands on, kinesthetic activities help people learn complex technical concepts in an intuitive way, and to be honest, provide a welcome break from all the formulas and function definitions. These activities not only increase engagement, but make the topic more accessible to a wider range of people. Come learn how to use and design these activities for yourself, or for a class!

In this data-centric era, the demand for responsible data science practitioners is more crucial than ever. However many data science education programs don’t adequately emphasize data ethics. To address this need, my colleagues, Ava Hoffman, Michael Rosenblum, and I have developed a course at Johns Hopkins, offering students hands-on experiences collaborating with community-based organizations on diverse data science projects. We’ve partnered with organizations championing various causes, including youth leadership, voting rights, transportation advocacy, and community tool banks. We’ve gained valuable insights about hands-on data ethics education and demonstrated that even data science education itself can support social good.

I recently undertook to build a robot to play a video game online. Using reinforcement learning, a custom computer vision model, and browser automation –all implemented in R/Python– I was able to create an AI which played the game to perfection. In this presentation I will share the lessons learned as I went through this process, and some hints to avoid the pitfalls I tackled. I will present some real-world business cases to answer the obvious why-question. For colleagues who teach Data Science and AI, I will show how an activity such as this can provide the entry-point and basis for discussion for more than half a dozen topics, ranging from formal logic, game theory, empirical inference, all the way to shiny and quarto.

In recent years, high-profile analyses have called attention to many contexts where the use of machine learning deepened inequities in our communities. After a year of research and design, the tidymodels team is excited to share a set of tools to help data scientists develop fair machine learning models and communicate about them effectively. This talk will introduce the research field of machine learning fairness and demonstrate a fairness-oriented analysis of a machine learning model with tidymodels.

Survival analysis is a part of statistical modeling (and machine learning) specifically for time-to-event data. This is common in medical research but has broad applications across industries, for example for analyzing customer churn.

The tidymodels framework is a collection of R packages for safe, performant, and expressive predictive modeling. We added models for survival analysis a while ago. Now we are back with the rest, including performance metrics specifically for these types of models. I’d like to show attendees how they can now leverage the entire framework for survival analysis: for all steps of the modeling process, from data prep to tuning. We are so excited to show you all of this!

Censoring in data can frequently occur when we have a time-to-event. For example, if we order a pizza that has not yet arrived after 5 minutes, it is censored; we don’t know the final delivery time, but we know it is at least 5 minutes.

Censored values can appear in clinical trials, customer churn analysis, pet adoption statistics, or anywhere a duration of time is used.

I’ll describe different ways to assess models for censored data and focus on metrics requiring an evaluation time (i.e., how well does the model work at 5 minutes?). I’ll also describe how you can use tidymodel’s expanded features for these data to tell if your model fits the data well.

This talk is designed to be paired with the other tidymodels talk by Hannah Frick

Tidypredict is one of my favorite packages. Being able to turn a fitted model object into an equation is very powerful! However in tidymodels, we use recipes more and more to do preprocessing. So far, tidypredict didn’t have support for recipes, which severely limited its uses.

This talk is about how I fixed that issue. Spending a couple of years thinking about this problem, I finally found a way!

Being able to turn a tidymodels workflow into a series of equations for prediction is super powerful. For some uses, being able to turn a model to predict inside SQL, spark or duckdb allows us to handle some problems with more ease.

When gathering data from groups (e.g., surveys), where does it go, and who does it help? How do we consider the privacy of respondents and power dynamics in who can access and benefit from data?

In this talk, I describe the creation of tools to let respondents get personalized feedback from the data they provide. This shifts the balance of power, allowing everyone to benefit directly, rather than providing information to only top decision-makers. I examine how Quarto and Shiny enable the creation of such tools, and describe takeaways from implementing them with two Fortune 500 companies.

This talk teaches how personalized tools can make data accessible to all, and how to alter the power dynamics of how organizations gather and use data.

Leveraging Posit Connect, our company transforms client engagement by providing direct support, extensive documentation (built with Quarto), and no-code applications for data exploration and analysis of real-world oncology data. This strategy provides us the greatest flexibility for subject matter experts to deliver client value, provide client assistance, enhance self-service learning, and to lower the technical barrier for data insights. Our commitment to client success and innovation is evidenced by our use of Posit Connect, providing tools for a competitive edge and data-driven culture.

Platform owners often ask questions like ‘how quickly do users migrate to new versions of R’, ‘what programming languages are used’, and ‘how are internal packages, dashboards and outputs consumed’?

The answer to these questions and many more lives within the operational logs collected by systems like Posit Connect, Github, and AWS.

I’ll share examples of how we use this data at Roche to shape our product roadmap. I’ll also share some ideas we are exploring to use this data to help empower our data scientists to understand the hidden consequence of how they work through feeding back personal cost and environmental impact - enabling informed decisions, e.g. what it means to schedule a pin to update daily, or request 2 vs 8 cores.

Every day, we face decisions, such as when to purchase a flight ticket to Seattle for posit::conf when prices change dynamically over time. As a decision scientist, I aim to understand these choices and the cognitive processes underlying them. In my talk, I’ll delve into how I leverage both R and Python to decode human decision making. I’ll focus on optimal stopping problems, a common predicament we all encounter, in which a decision maker must determine the right moment to stop exploring options and make a choice based on their accumulated knowledge. Attendees will be introduced to the field of decision science and learn how R and Python can assist in advancing the study of the human mind.

How can we level up the R skills of a team with varied backgrounds and levels of experience in R? At PicnicHealth, a healthtech company that collects and abstracts patient medical record for use in research, we’ve come up with a number of strategies to share. We’ll discuss building internal packages that assist with common tasks and distributing them with Posit Package Manager, hosting dashboards on Posit Connect and integrating them with other internal apps, maintaining a wiki, and holding regular Office HouRs to give folks a place to ask for advice. We’ll share examples and show some projects that have benefited from this approach. This talk is suitable for anyone who has at least one coworker using R.

Do you know R helps reduce tailpipe emissions like Carbon, NOX and other emissions. Am Saumiitha Leelakrishnan, mom of 3 kids who cares for our environment, Technical Specialist leading Diagnostics and Emissions Data Analysis projects in Cummins - A 100+ year Engine Manufacturing company. In this talk, I will be sharing how R helps meet Global Product Compliance and Deliver Solutions that lead to a cleaner environment. You will learn the transition from MATLAB to R and Python, how I utilized R’s seamless integration, statistical capabilities, advanced modelling techniques, Quarto, ML Algorithms to develop and maintain web applications in posit connect. This talk will benefit Data Science Community with example of harnessing the power of R.

Lee Durbin’s journey at Auckland Council transitions from Excel reliance to R, highlighting data analysis evolution in the public sector. Emphasising proficiency and sustainability, he leverages R’s rich ecosystem and community support to foster a mature data culture. Despite initial inexperience, resources like R For Data Science and TidyTuesday rapidly enhanced his skill set. Addressing the sustainability challenge, he implemented the RAP framework and {renv} and {targets} packages, emphasising collaboration and continuous improvement. This talk outlines transforming data practices through R, ensuring organisational resilience even when responding to natural disasters.

Have you ever considered using the power and flexibility of Quarto for note-taking and knowledge management? I did, and now I use Quarto websites to track my PhD progress, document insights from conferences, manage collaborative research projects and more. Let me show you how easy it is to implement standard knowledge management system features, such as cross-referencing, search indexing and custom navigation. But what if you want more advanced features like glossaries, document listings and summaries of datasets? Well, with some creative use of Quarto’s many features and extensions, almost anything is possible. Whether you’re new to Quarto or a seasoned expert, consider adding Quarto to your note-taking toolkit.

In this talk, we will walk through the process of converting an idea into a creative visualization in R and ggplot2, from finding inspiration to writing the code. We’ll look at handy tips to make the creative and coding process smoother, how to create more personal plots, as well as the importance (and fun!) of sharing your work with a great community.

Quarto makes web graphics accessible to data scientists, letting them write Observable JavaScript (OJS) right alongside the languages they already use, like R and Python.

OJS is powerful, but making graphics that animate and transition can be a challenge. In this talk I’ll demonstrate ways to use Quarto and OJS with graphics libraries to make them react and animate according to your data.

We’ll even look at making bespoke, reactive graphics with Svelte and D3.js using Sverto, a Quarto extension designed to help you on your web graphics journey.

While many visualization framework focus on facilitating the creation of a single plot, combining multiple plots into a single coherent figure is often the end goal when creating a visualization. There is no shortage of packages in R for doing this but they often lack in flexibility or are cumbersome to use. Because of this, the patchwork package has become the tool of choice for many, with it’s clear API and flexible customizations. This talk will guide the audience through the core concepts of the patchwork package, starting with simple compositional tasks and moving all the way up to advanced nested layouts and insets, preparing them to use the package in their day-to-day work when they come home.

Imagine a world where crafting a stunning, insightful data visualization is not just about the end product, but the journey. A world where every decision, every tweak, every step in your creative process is not just a fleeting moment, but a valuable artifact.

{camcorder} is an innovative R package that revolutionizes the way you create and share your data visualizations. It not only allows you to preview your visualizations exactly as they will be saved, but also records every plot you create, turning your creative process into a compelling narrative.

Discover the inspiration behind {camcorder}’s creation, its use, and explore the ways you can leverage it to tell captivating stories to your stakeholders. Are you ready to press play?

My talk is about how our team took the past year to take a local R workflow into production. This updated process uses a mix of R, Python, SQL, and Tableau dashboard, all of which involves multiple teams and stakeholders.

The project started as a manual monthly process to measure the effect of ESPN’s marketing to get consumers to watch more of the same sport.

But then we not only needed to automate this process, but also to scale it to measure the effect of marketing to engage consumers with other ESPN products.

I wish to share a few lessons our team learned about in coordinating people and data at a high level to deliver a complex data product. This talk is for everyone.

When a highly regulated enterprise organization wants to explore new worlds and build an open-source R Statistics Package it just needs to assemble R and statistics experts, right? GSK’s recent development of a Bayesian Stats Package found that was not the case and emphasized the need for an interdisciplinary team. Using The Expanse TV show as an analogy, I will discuss our journey from an idea to enterprise adoption. I will talk about how we integrated varied technical skill sets, learning each other’s unique language and technical worlds, while dodging metaphorical asteroids that could derail package development. Join me to learn about building a domain specific R package and fostering a deeper connection with the open-source community.

COTA’s team of oncologists and data scientists curate real world data used by life science companies and healthcare partners to inform drug development and patient care. Over time, we have received many of the same questions from our data users, which indicated a dire need for translating our internal clinical guidance and data model knowledge into a tool for successfully navigating our data. We developed rwnavigator, an R package that helps users easily prepare COTA data for analysis with time-to-event packages. As first-time package developers, we ran into many challenges as we created, tested, and deployed rwnavigator. We will share with the R community our motivations for developing rwnavigator and best practices we learned.

Researchers have been and are facing a replication crisis. At the same time, research teams have been growing, allowing for more complex analyses. Our redistricting research team approaches these difficulties by building on the existing R package infrastructure, like devtools and testthat. Specifically, we design a template-based approach which sits within an R package. New analyses are initialized via a function in that package. Each analysis gets a readable, unique ID and every file within is assigned a number. This structure allows us to then validate each analysis and share the results of long analyses for code reviews. We use this template approach to design academic research projects which are easily reproduced and hopefully correct.

AI tools like GitHub Copilot have drastically accelerated the pace at which data scientists can get things done, but they can also undercut the learning process if not integrated with care and intention. We’ll discuss emerging best practices for use of coding assistants like Copilot for those attempting to build long-term productivity and wisdom while still getting things done.

rainbowR is a friendly community for LGBTQ+ folks who code in R. We run monthly online meet-ups where participants chat and share their R and Quarto work in a supportive environment. We also organise a buddy scheme, which randomly pairs members of the community, to encourage people to meet and connect. We have exciting plans for the future!

You’ll learn about what rainbowR does and how you can get involved, whether as a member of the LGBTQ+ community or as an ally, and hopefully forge new connections at the conference and beyond. We believe the whole R community benefits when that community is diverse and inclusive.

Packages extend the power of a programming language and working with source code makes workflows repeatable. However, the number of existing packages can be overwhelming to the point that the purpose of individual packages or their role in projects becomes unclear. This talk will discuss the importance of working with source code, and introduce ways to enhance package load calls. First, by building annotations about loaded packages with the annotater R package and also by learning from existing comments in public code from GitHub. The general value of recording information such as package source, version, title, or even which functions or data are used in a script will be discussed in both data science and teaching environments.

Documentation stands at the heart of software development, ensuring clarity, usability, and continuity. Acknowledging the rich diversity of documentation tools in the Python and R ecosystems, this talk introduces JSquarto, a tool inspired by these communities to fulfil JavaScript documentation needs. JSquarto uses Quarto, to generate API reference docs & tutorial for JavaScript tools. Through the integration with Quarto, JSquarto extends beyond the capabilities of JSDoc by offering multilingual support. This talk will unfold the journey of JSquarto, from its conceptualization motivated by the R and Python communities to its development as a tool that addresses the specific challenges of JS docs for Open Science Community Saudi Arabia.

Findable, accessible, interoperable, and reusable (FAIR) data sharing is a key component of open science, but presents a challenge for researchers, especially those with limited technical expertise or resources. If datasets are shared, it’s most often as static files, restricting the FAIRness of the data. We use Quarto and Observable JS to develop DataPages, tools and templates that bridge this gap. DataPages enables researchers and other data distributors to easily share versioned datasets along with interactive visualizations, rich documentation, and user-friendly access functionality. We’ll present a gallery of DataPages that we’ve developed for a diverse array of datasets, and demonstrate how to create a DataPage for your own datasets.

Posit Connect is powerful because it allows data scientists to become tool builders. Connect enables rapid prototyping and workflow/process automation which helps streamline data science communication. The delivery of workflow orchestration integrations helps data science teams work with their data engineering counterparts to do more sophisticated automation and data reporting. This talk will showcase a new Python SDK for Connect which will make it feasible for workflows running outside of Connect to trigger events in Connect.

Disinformation campaigns use coordinated networks of inauthentic accounts. This talk shows how to leverage R to uncover these operations on social media. We use previously shared datasets from Twitter (now x) , then wrangle with dplyr/tidyr. We check for coordinated behaviors like brigading, astroturfing, content/timing similarities, and common posting schedules. Network analysis including community detection, centrality measures, and ggraph visualization maps relationships and identifies clusters. Anomaly detection via clustering and sentiment analysis tracks amplified spread. Together, these R techniques combining large-scale statistics with investigative methods can reliably identify coordinated influence campaigns.

In my talk, I want to make a case for becoming more language agnostic. This doesn’t mean that I want to discard all notions of I like this or that language better. Instead I want to foster a mindset that is more about problems to be solved rather than languages to be used.

As part of my talk, I’ll illustrate how the R skills I’ve developed are broadly transferable to other languages I have to use at my new job. And to do so, I’ll provide specific examples using R packages that are available in other languages or programming paradigms that are shared across languages.

In the end, my talk should equip R users with a list of low-hanging fruits that make it easy to jump into other languages.

As we work in open data science spaces, we frequently peer-teach coding and collaboration skills. The setup work is often grossly underestimated and unseen. I’ll share how Openscapes automates setup with the Kyber R package that uses googlesheets4 and creates RMarkdown documents that become collaborative Google Doc agendas, and sets up repositories and organizes people on GitHub. Kyber replaces manual steps with R functions while maintaining the ability to edit outputs so we’re not constrained by the automation. It has enabled us to teach workshops repeatedly in less time – in 2022 we led 4 concurrent learning cohorts with 160 government scientists! Kyber is openly available to fork, reuse, and extend, and other groups are doing just that.

While there are numerous resources to learn R online, many professionals tasked with instructing a live R course feel under prepared, because of lack of guidance or time. Better resources for experts-turned-instructors are part of the solution to meeting the growing demand for programming skills education. In this talk, we distill our experience teaching R to hundreds of public health professionals, graduate students, and undergraduate interns who lack a computer science background. We offer 10 key takeaways for teaching introductory R so you can get up and running teaching R quickly. We believe these rules will help you prepare and inspire learners for the next steps of their R learning journey.