R and Python together at last, the AI takeover, and Quarto ups its game.

Posit::Conf, Posit's annual conference, was held last week in Seattle, WA. While the name might be unfamiliar, many of you might recognize Posit's previous iteration, RStudio. Posit has produced many products R users know and love, including RStudio, Shiny and the Tidyverse. After a whirlwind three days, here are some takeaways from the conference!

A New All-in-One IDE

Togetherness was a key theme, with Posit rolling out the beta version of Positron, an IDE that combines elements of RStudio, VS Code, and introduces its own unique features. While switching IDEs is generally not at the top of any coder's to-do list, Positron is multilingual, enabling users to code in R, Python, and Julia within a single project. It also introduces new ways to interact with your data, such as readily available summary statistics, the ability to filter or sort data by multiple fields while retaining the active query in the window at all times, and the option to resize graphs and figures in the plots pane through a simple user interface. Positron also leverages many new and existing VS Code extensions, offering a wealth of customization and additional capabilities. Rstudio is by no means leaving the picture– Positron is still in beta, and Rstudio will continue to be supported for a good, long time. However, if you (like me) crave a unified interface for your R and Python coding, Positron may be worth trying out.

Helping R Users Learn Python

For Py-curious R coders, a session on “Python Rgonomics” suggested some packages to make the transition easier for those of us spoiled by the tidyverse.

• Polars for exceptionally fast data wrangling and dplyr-like syntax

• Plotnine and seaborn for ggplot-like syntax when making graphs and figures

• Great tables for producing functional, readable tables (also available in R as the gt package).

• Pyenv for environment management.

• Pins (for R and Python!) publishes objects to “boards” that allows users (or multiple users) to access them across projects. Boards can include shared/networked folders, like DropBox or Google Drive.

AI for All

Melissa Van Bussel provided practical tips for using generative AI. She highlighted some new capabilities of ChatGPT 4o, including the ability to transcribe handwritten notes and tables, preserving colors and formatting. GPT can even convert these into HTML or Quarto formats.

She shared insights on prompt engineering (i.e., how you ask questions to AI engines), noting, “Writing effective prompts goes hand in hand with your existing expertise.” Achieving correct and effective output requires providing specific prompts and making corrections when errors occur. She recommended structuring prompts in a way that mirrors coding practices. For example, to generate a graph, start by specifying the data, then define how to map each axis and assign colors. Next, specify the graph or chart type and indicate any grouping by other variables. Finish with aesthetic (eg palette, theme, title, legend) preferences.

While presenters were enthusiastic about the possibilities of generative AI, a recurring theme was the necessity for users to provide clear direction and verify the results. One presenter compared AI to hiring a new human assistant—AI can assist with tasks effectively when given proper guidance but will make mistakes and requires careful supervision. Generative AI performs best when used to quickly accomplish tasks that users already have the knowledge and skills to handle themselves.

One of the most prevalent use cases for generative AI was in combination with Shiny. Joe Cheng's presentation demonstrated integrating AI into Shiny apps, specifically into Shiny dashboards. Users could request modifications to the data displayed on the dashboard using plain language, which the AI translated into SQL queries to adjust the output based on the request. This is particularly noteworthy as the AI accessed only the schema, not the actual data, to apply the filters.

Winston Chang developed an AI assistant to help people build Shiny apps and did a live demonstration. Although still experimental, Shiny for R is widely used, and the assistant showed promising outputs.

Quarto Expands its Horizons

Quarto is a relatively new version of R Markdown (a publishing tool) that allows users to knit together code into documents, dashboards, web pages, PDFs, and even eBooks. An added advantage of Quarto is its multilingual capability– it, like Positron, supports both R and Python. Some new capabilities were highlighted at the conference:

• Dashboards: Easily build dashboards using the Quarto extension in RStudio or Positron. Each visualization (graphs, maps, tables, and even just text boxes) can be arranged like tiles or cards. Dashboards can also include sidebars and toolbars, and can support interactivity, including cards that use  jupyter widgets, leaflet, and shiny.

• PDFs: Quarto now uses typst instead of LaTeX, enabling users to create customized PDF outputs with a more intuitive language.

• HTML (Websites): Quarto (like R Markdown) can produce HTML outputs. However, now it also supports more flexible HTML code chunks, and allows for HTML/CSS/JavaScript integration.

• Quarto live: A quarto extension that allows users to embed code blocks and exercises for R and Python into Quarto documents. This has lots of teaching applications, and can be used to generate exercises similar to those you find in DataCamp and other online coding courses.

• Closeread: A scrollytelling extension for quarto that enables interactive storytelling similar to that seen in fancy New York Times articles or Esri Storymaps. The gallery has some example outputs, while the guide can walk you through the process of creating your own scrollytelling page.

Finally, one of my favorite quality-of-life take homes was simply that it is possible to include emoji in your R or Python code, either using unicode or simply pasting them in. While the demonstrated use case was to make specific messages stand out in your log or printed statements, sometimes a picture can convey what 1000 characters cannot and help you enjoy coding just a little bit more.

If you love (or begrudgingly engage in) data science, I encourage you to check it out next year– virtual registration for educators and academics in 2024 was free, and hopefully will be next year too!

At a time when GeoAI, machine learning, and big data analytics remain the most popular topics in data science, ESRI maintains its commitment towards classical statistics and spatial statistics. While these sessions did not draw few hundred people as some of the sessions related to deep learning and GeoAI had, it certainly lured niche audiences working in the realm of data analysis. Day 4 of the ESRI User Conference had several sessions focused on novel approaches and tools to perform spatial and temporal analysis.

Neighborhood Explorer:

If you have done hot spot analysis in ArcGIS Pro, you may have manually hovered over a cluster and explored values at that location and at its neighboring locations. Or you may have used Select by Attribute tool to explore values at all locations within a neighborhood. Gone are the days when you must do such tasks manually. ESRI has introduced Neighborhood Explorer tool that will spit out an attribute table when you select a polygon in the cluster. Let's say you select a census tract in one of the hot spot analysis clusters. The software will show the attribute values that just includes all the features in the neighborhood of the selected census tract.

Spatial Weights with Network Data:

Traditionally, creating spatial weights had been limited to contiguity, adjacency, or distance-based methods. Now, users can include a barrier layer such as a freeway or street data. The weights matrix file that is created will account for whether these features are divided by a freeway or not. Since the type of neighborhood that is one side of the freeway can be different from the opposite side of the freeway, ESRI has created tools that can account for real world conditions.

Hot Spot Comparison:

How many times did you run hot spot analysis on one variable and were tempted to run on another variable and compare the patterns in hot spots between the two variables? ESRI has heard its user requests and developed a new tool called Hot Spot Comparison tool that allows a user to compare hot spots in two layers.

• Do hot spots of asthma overlap with hot spots of wildfire?
• Are hot spots of rent in 2020 in Los Angeles the same as hot spots of rent in 2024?

These are some of the questions that the tool allows you to answer by creating a standardized approach to compare hot spots between two different variables or between different time points of the same variable.

Time Series Forecasting:

Outside of space-time cubes which were introduced a few years ago to examine spatial and temporal data, ESRI now offers new tools from classical time series models. The cross-correlation function that is used to compare two time series and examine whether one variable leads the other or lags the other can be assessed in ArcGIS Pro.

• How does energy consumption increase as temperature increases?
• What is the relation between covid-19 reported cases and deaths?

These are some of the questions that can be answered using the time series analysis tools to understand the lag between one variable and the other.

Compare Geostatistical Layer:

This tool allows users to compare the results based on output of different kriging methods. Let's say a user ran the tool with simple kriging and then re-ran the tool with ordinary kriging then this tool provides various parameters that would allow users to identify the best method.

3D Interpolation & Nearest Neighbor 3D:

While tools for performing Inverse Distance Weighting, Kriging, and Empirical Bayesian Kriging have been existing since quite a long time, ESRI has expanded some of these tools to include 3D models. This includes Empirical Bayesian Kriging with 3D and Nearest Neighbor 3D.

Causal Inference:

While this was the last session towards the end of the conference, right before the party at Balboa Park, it still did not deter the audience from staying for the entire duration of the session.  The session did an excellent job in explaining to the audience the principles of randomized control trials and causal inference, the tools in ArcGIS, and how to assess uncertainty in the model.

•  Does social capital increase life expectancy after controlling for confounding variables?
•  Does living in close proximity to a toxic release facility increase respiratory-illness related    hospital visits after controlling for median household income?
• How does changing the amount of fertilizer increase crop yield after controlling for soil type? 

 These are some of the questions that researchers can now explore and visualize the results spatially.

The third day of the ESRI User Conference was an exhilarating deep dive into the world of Artificial Intelligence. The AI Summit, the first of its kind hosted by ESRI, focused on the integration of AI into the ESRI ArcGIS Suite of software and tools. The excitement in the air was palpable as attendees explored the transformative potential of AI in GIS.

AI Summit Highlights

The AI Summit highlighted the strategic incorporation of AI into two broad categories within the ArcGIS ecosystem: GeoAI and AI Assistants.

• GeoAI: This category encompasses advanced AI models and algorithms specifically designed forgeospatial analysis. Attendees learned about howGeoAI can enhance spatial data interpretation, improve predictive modeling, and support more accurate decision-making. The sessions showcased real-world applications, demonstrating howGeoAI is being used to tackle complex challenges.  These tools included:
  • Deep Learning Models and object detection and pixel classification
  • Unstructured text searching and geocoding
  • and many others 
• AI Assistants: These tools are designed to augment the user experience within theArcGIS platform, making it more intuitive and efficient. AI Assistants leverage generative AI to provide intelligent recommendations, automate routine tasks, and enhance data visualization. The demonstrations highlighted how these assistants can streamlineworkflows, improve productivity, and empower users to extract deeper insights from their data.  AI assistants are being included in:
  • Help documentation, online and within ArcGIS Pro
  • Webmapping - Smart mapping
  • Survey 123 webform design
  • ArcGIS AI Hub Assistant
  • and the list goes on

Innovative Applications and Future Directions

Throughout the day, experts and innovators shared their experiences and insights on the cutting-edge applications of AI in GIS. From machine learning models that classify historic and current aerial photos to AI-driven tools that optimize road condition extraction from full motion video capture, the presentations underscored the vast potential of AI to revolutionize the field.

Exciting Outcomes

One significant outcome of the day was my decision to request access to ArcGIS Hub Premium and its AI Assistant. This powerful tool will enable the querying of geospatial data created and maintained by IGIS and UCANR using AI using a natural language model. This will open up major opportunities for those needing information about UCANR and data that is important to Californians, making it easier to access, analyze, and leverage critical geospatial information.

Looking Ahead

The integration of AI into the ArcGIS Suite represents a significant leap forward, promising to unlock new possibilities for geospatial analysis and decision-making. As the conference progresses, attendees are eager to see how these advancements will continue to evolve and shape the future of GIS.

Perspective on AI

In response to concerns that AI might replace GIS professionals, the panelists provided some important perspectives. They emphasized that GIS professionals may potentially lose their jobs if they don't embrace AI. Just like if a GIS Professional was still using AML or Avenue programming laguages instead of modern languages and tools, they would eventually be replaced by someone who was using new languages and tools.  The was reinforced a comment by one panelist who added, "If you lose your job because of AI, it will be to someone who is using AI."

Stay tuned for more updates and insights from the 2024 ESRI User Conference! The journey into the future of GIS, powered by AI, has just begun.

Where is the nearest oak tree close to 1600 Pennsylvania Avenue NW, Washington? How many trees are in Washington DC? When is the trash pick-up date for 1600 Pennsylvania Avenue NW? These are some of the questions that residents can answer using AI feature in the Open Data Portal.

ESRI's team along with its partners is testing models through which AI can be trained to search through and query data sets within the data portals of state and local government as well as within ArcGIS Hubs.

To ensure data are easy to discover, share, and can be reproduced, ESRI devoted a session on ensuring users are maintaining metadata and preparing it in a way it is ready for AI. How to ensure that AI assistance search is limited to data within the hub is something ESRI is still working on.

ESRI is working on a suite of tools to increase collaboration among teams and within community.  One such feature is the Discussion Board in ArcGIS Hub. For example, when a small team working for the parks and recreation has to decide some good spots to set up picnic tables, a user can invite other members in the team to share their comments.  Along similar lines, ESRI also gave considerable attention to its product Pipeline. Similar to Model Builder which works in ArcGIS Pro, Pipeline connects datasets and workflows in ArcGIS Online.  ESRI's collaborative tools are also aimed at including users and databases that are not spatial in nature such as Knowledge Graphs. Knowledge graphs harness the knowledge of graph theory to bring connections between people and places and what happens when one of the nodes in the chain gets affected by an externality. Clearly, with products like knowledge graph ESRI is reaching new territories as it brings people who are working with non-spatial data also into using ArcGIS products.

ESRI's partnership with Microsoft remained in the spotlight with sessions focused on Introduction to Microsoft Fabric, ArcGIS tools available in Office 365 with Excel, SharePoint, Power BI, and Teams.

After a series of technical sessions, the crowd was eagerly waiting to watch John Nelson's and Kenneth Field's session on Map Wizardry. Even 45 minutes before the session started, participants had already started queuing up outside the ballroom. With over 500 participants, this probably was the most attended session after the plenary. Both Kenneth and John kept the audience entertained and engaged with their cutting-edge techniques on 2D cartography while Nathan Shephard amazed the crowd with his 3D maps.

With several technical sessions and user presentations, ESRI continues to keep the voracious appetite of the conference attendees to come back for Day 3 of the conference.