January 2020 Summaries

GitHub Desktop 2.3 introduces new features designed to streamline the user experience by addressing common obstacles encountered during repository contributions. The update includes a solution for the confusion between cloning and forking repositories by automatically creating a fork when a user tries to push to a cloned repository, eliminating the need for complex "Git magic" to salvage work. Additionally, the update provides a warning when a user attempts to make changes on a protected branch without the necessary permissions, prompting them to create a new branch and continue their work seamlessly. These enhancements aim to prevent common frustrations and empower both new and seasoned developers to contribute to open source projects more effectively, while reducing the need for extensive tutorials and workarounds.

GitHub is using machine learning to enhance its "good first issues" feature, designed to help new contributors find beginner-friendly tasks in open-source projects. Initially launched in 2019, this feature relied on manually curated labels to identify suitable issues, but only covered about 40% of recommended repositories. To expand coverage to roughly 70%, GitHub introduced a machine learning model that predicts suitable issues by analyzing issue titles and bodies, using both classical and deep learning methods such as neural networks implemented in TensorFlow. The model employs a weakly-supervised approach with a heavily imbalanced training set, and prioritizes precision over recall to avoid false positives. Data processing and inference run daily, ensuring recommendations remain current. Future plans include improving classifier models, enhancing repository recommendations, and allowing maintainers to manage ML-based suggestions, while also exploring personalized recommendations for contributors.

GitHub provides a streamlined way for individuals to begin contributing to open source projects by helping them find "good first issues" to work on. Users can explore topics of interest, such as machine learning, or find beginner-friendly issues in specific projects by accessing relevant links on GitHub. Personalized recommendations are also available based on users' past activities on the platform. GitHub employs machine learning algorithms to maintain the relevance of these recommendations, and users are encouraged to provide feedback for improvement. The platform's curated lists and innovative features aim to make starting with open source contributions accessible and engaging.

Derrick Stolee, a Git contributor since 2017, discusses the introduction of the experimental git sparse-checkout command in Git 2.25.0, designed to enhance performance for large repositories by allowing users to restrict their working directory to only necessary files. This feature is particularly beneficial for developers dealing with monorepos, like those using the "microservices in a monorepo" pattern, as it speeds up Git operations by minimizing the number of files to be processed. Previously, using sparse-checkout required complex manual operations, but the new command simplifies this process. Stolee provides examples demonstrating how developers can optimize their workflow by limiting the files they interact with, using a sample Git repository for illustration. The post also explores the technical improvements in the sparse-checkout feature, such as the adoption of "cone mode" for better scalability, and highlights the potential of pairing sparse-checkout with the partial clone feature to further accelerate workflows. Future plans for the feature include adding more commands and improving the cloning experience, with ongoing evaluations and updates anticipated as the feature matures.

GitHub is enhancing its Student Developer Pack by adding 22 new partners, bringing the total value of included tools and training to over $200,000 per student. This initiative aims to provide students worldwide with access to a wide array of resources, including web accessibility testing, educational electronics, SaaS business tools, code visualization, and translation management, among others. The new partners emphasize the importance of sharing work and the Pack's potential to deepen students' understanding of software development. Available to verified students aged 13 and above globally, the Pack encourages students to leverage these resources for various projects and offers a platform for both students and companies to suggest new tools to be included.

Game Off, a month-long game jam that began on November 1, engaged 4,000 participants to develop games inspired by the theme "leaps and bounds," culminating in a voting process that determined the winners across various categories. The first place overall went to "Sealed Bite," a pixelated platformer by @securas, which also topped the Gameplay and Graphics categories. "Retrochase," a game featuring a teleporting futuristic ninja, secured second place, and "BLAWK," a unique Windows game, won first in the Innovation category. "The Gems," an action-adventure point-and-click game, triumphed in the Audio category, while "Noneuclilypads," a turn-based puzzle roguelike, excelled in Theme Interpretation. Participants were encouraged to explore over 200 diverse games created for the event, showcasing a wide array of genres and platforms, with appreciation extended to all contributors and anticipation for the next Game Off.

GitHub for mobile was announced at GitHub Universe 2019, offering users the ability to engage in development tasks such as reviewing code and collaborating with team members via a fully native app experience on both iOS and Android devices. Following an initial beta release for iOS, the Android version began testing, with interested users able to sign up for the waitlist to participate. The app is compatible with Android devices running version 5.1 or later and supports features like multiple screen sizes and dark mode, allowing users to work flexibly and efficiently from any location.

Git 2.25 introduces several notable updates and features, including enhanced support for partial clones and a new git sparse-checkout command aimed at simplifying sparse checkout management by allowing users to set, list, enable, and disable sparse checkouts more easily. This version also includes the introduction of "cone mode" for improved performance with sparse checkouts, enhancement of git apply's compatibility with merge.conflictStyle, and updated support for function signature detection, now including the Elixir language. Additionally, Git 2.25 refines the git log --graph output, improves handling of directory-level rename detection, and introduces the --pathspec-from-file option for commands like git commit to handle numerous pathspecs efficiently. These advancements highlight Git's continued evolution toward more efficient and user-friendly functionality, while also addressing previous limitations and bugs.