July 2022 Summaries

A Front-End Developer for the San Antonio Spurs focuses on converting NBA player data into visualizations, such as shot charts, which map shooting locations on the court to help the team's front office and coaches make strategic decisions. This role involves analyzing spatial patterns of player performance and fouls, drawing from a background in cartography and GIS. The developer's journey began with tracking human rights abuses in Burma using mapping tools, which later evolved into creating interactive web maps and developing skills in JavaScript. A pivotal moment came when the developer was introduced to Kirk Goldsberry and discovered the potential of data visualization in basketball. This led to a successful application to the Spurs, as the developer leveraged a public data set from Second Spectrum to create visualizations that demonstrated their expertise in both front-end development and basketball analytics. Motivated by a passion for basketball, the developer saw the opportunity to apply analytical skills to a beloved subject matter as a key driving force.

A Front-End Developer for the San Antonio Spurs focuses on transforming data related to NBA players into visualizations like sparklines, bar charts, and shot charts, which map points attempted from locations on the court. Shot charts, popularized by Kirk Goldsberry, provide insights into spatial patterns, helping the front office with player acquisition and coaches with player development. The developer's journey into data visualization began while tracking human rights abuses in Burma, which led to a GIS Masters from Penn State. Their skills in cartography and GIS, honed at Axis Maps and through interaction with Goldsberry, facilitated entry into the role with the Spurs, especially after creating visualizations from data provided by Second Spectrum's video tracking system. The position was driven by a passion for basketball and the opportunity to apply analytical skills to the sport.

TypeScript developers often encounter situations where they need to determine the specific type of an object within a union of possible types, a process known as "narrowing." This article explores various approaches to type narrowing, including inline checks, type predicates, and discriminated unions. Inline checks are quick to implement but can become messy and are not reusable, while type predicates encapsulate logic but can be error-prone and require careful maintenance. Discriminated unions, which involve adding a property to each type to indicate its kind, offer advantages in maintainability and clarity by making explicit what combinations of properties are possible. They require more upfront modeling but lead to more robust and maintainable code by leveraging TypeScript's capabilities to handle changes across the codebase. The article concludes by advocating for discriminated unions over other methods due to their ability to simplify type checks and improve code reliability, especially in complex scenarios like modeling asynchronous requests in React components.

The article explores type narrowing in TypeScript, a process crucial for determining a specific type from a union of possible types, using examples such as async request states and pet objects. It compares three methods for type narrowing: inline checks, type predicates, and discriminated unions, highlighting their respective advantages and disadvantages. While inline checks are quick to implement, they can become complex and are not reusable, and type predicates, though encapsulating logic, require careful management to avoid errors. Discriminated unions are presented as the most robust solution, as they simplify checks, enforce comprehensive handling of all union cases, and enhance code maintainability by leveraging TypeScript's type-checking capabilities. The article argues that discriminated unions, despite requiring more initial modeling effort, provide clearer, more reliable, and maintainable code by explicitly defining possible type combinations and reducing the need for manual updates across the codebase.

Felt, a mapping company, explores the challenge of delivering an intuitive and efficient user experience by managing complex data layers and ensuring swift interaction capabilities. This involves making subjective decisions on how to represent data in maps, which are purpose-built to effectively communicate ideas. The company emphasizes the importance of heuristics over universal solutions for determining data layer extents, particularly in their "Zoom to fit" feature. Instead of burdening the frontend with complex calculations, Felt leverages PostGIS to handle bounding box calculations during data upload, using functions like ST_Extent and ST_EstimatedExtent to determine the most efficient approach. While ST_Extent provides accurate bounding boxes, ST_EstimatedExtent offers a quicker, pre-calculated alternative, as demonstrated in a test on a large table of multipolygon geometries. This approach allows Felt to balance accuracy and performance in map visualization.

Felt, a mapping company, focuses on delivering an intuitive data interaction experience by making complex data layers easily accessible through features like fast tile delivery and interactive geometric attributes. The challenge lies in the subjective nature of creating effective maps and the difficulty of implementing universal solutions. To address this, Felt engineers embrace heuristics when determining the extent of data layers, using PostGIS to calculate bounding boxes for large datasets. They leverage PostGIS functions like ST_Extent and ST_EstimatedExtent to optimize performance, with ST_EstimatedExtent providing a significantly faster, though slightly less precise, estimation of spatial data extents. This approach allows Felt to effectively manage large data layers while maintaining efficient frontend performance.

Felt, a mapping platform, has introduced ten new data layers in June, including Bureau of Land Management Trails and Campgrounds, the US Drought Monitor for June 2022, and various political and geographical boundaries like Congressional Districts, State Senate and House Districts, and ZIP Codes. These data layers are designed to streamline workflows by consolidating information from multiple sources into a single view, addressing user frustrations of toggling between mapping websites. The platform's offerings also include global features such as a network of lines representing meridians and parallels, a global drought risk assessment, a comprehensive listing of lighthouses, and worldwide libraries sourced from OpenStreetMap. Users are encouraged to duplicate maps and explore these new layers, and they can request additional layers for future updates.

Felt, a platform that integrates various data layers onto a single map interface, added ten new layers in June, catering to user requests for more streamlined workflows. These new layers include Bureau of Land Management Trails & Campgrounds, the US Drought Monitor for June 2022, a graticule for mapping, global drought risk indicators, lighthouses, Congressional and State legislative districts, global libraries, and ZIP Codes. Each layer is designed to provide users with specific information, such as tracking drought conditions over time, exploring public lands, or understanding electoral districts. The platform encourages users to duplicate maps with their chosen layers and invites suggestions for future data layer additions to further enhance its offerings.