Techlist.io - Korean Tech Blog Curator

figmaMar 18, 20252 min readCurated summary

How Teams Tap Into the Power of Design with Figma Slides | Figma Blog

Figma Slides has moved out of open beta with new features designed to combine presentation simplicity with Figma’s high-fidelity design and collaboration capabilities. The platform aims to help designers and cross-functional teams create polished, brand-consistent decks while giving them more control over storytelling and presentation workflows. ## Figma Slides Leaves Open Beta - Available to users with Full, Dev, or Collab seats. - New features include: - Improved video controls, including blur effects, scrubbing, and playback-speed adjustments. - Components for maintaining consistency across brand assets and customer logos. - Object-level slide and fade animations. - `.PPTX` import and export for working with external collaborators. - Existing Figma capabilities—such as auto layout, custom fonts, brand libraries, and embeddable prototypes—are available within the slides experience. - Figma positions Slides as a shared storytelling workspace rather than merely a traditional presentation tool. ## Fine-Tune Designs with Templates and Components - Teams can build decks from templates connected to their existing Figma design systems and brand libraries. - Merkle uses curated templates for UI/UX presentations, research, data visualization, and client-specific work. - Reusable components reduce design decisions and help teams produce higher-quality decks more consistently. - Decagon redesigned its sales deck around its brand identity, using custom visuals, layouts, and an appendix of adaptable slides. - Custom templates let teams tailor presentations to specific customers, priorities, and business goals while preserving brand consistency. ## Control the Narrative - Figma Slides supports audience-specific storytelling by making it easy to rearrange slides and review the full deck in grid view. - The article emphasizes distilling a presentation into a clear central idea rather than adding unnecessary visual elements. - Templates can adapt to different contexts, such as fundraising stages, audiences, products, and founder profiles. - The ideal structure may vary: some decks lead with insights, while others introduce the team early to establish credibility. Figma Slides is best suited to teams that want presentation speed without giving up design precision. Its combination of reusable systems, collaborative editing, flexible storytelling, and `.PPTX` compatibility makes it a practical option for internal reviews, sales presentations, and client-facing decks.

Read original

Figma Design Systems User Experience Collaboration+3

figmaMar 17, 20252 min readCurated summary

Karri Saarinen: 10 Rules for Crafting Products That Stand Out | Figma Blog

Linear CEO Karri Saarinen argues that exceptional products stand out through deliberate craft, not merely speed or feature delivery. Quality must be embedded in leadership, team structure, collaboration, and product decisions. The goal is not perfection, but a consistently high quality bar that earns users’ attention and trust. ## Quality Starts with Leadership - Leaders must explicitly prioritize craft and create the conditions for teams to pursue it. - Declaring quality important is insufficient; companies need to explain how it supports the business. - Design can make complex products more approachable, as Saarinen observed at Coinbase. ## Build Small, High-Quality Teams - Smaller teams reduce excessive debate and preserve execution quality. - Hire people who already demonstrate strong craft rather than focusing only on measuring output. - High standards should be established through the quality of the team’s work. ## Eliminate Handoffs - Product managers, designers, and engineers should share responsibility instead of working in isolated stages. - Removing “handoff to development” prevents teams from treating quality as someone else’s responsibility. - Understanding implementation constraints helps everyone contribute to the final result. ## Avoid Rigid Product Silos - Too many specialized teams can create inconsistent quality and fragmented cultures. - Rotating responsibilities keeps ideas fresh and prevents teams from becoming narrowly focused. - Users care about the overall product experience, not the organization chart. ## Treat the Specification as the Starting Point - Meeting requirements is only the minimum standard, not the definition of a great product. - Teams should add care and judgment beyond what the specification explicitly demands. - Craft is what turns a technically functional product into a memorable one. ## Quality Is Not Perfection - Products should pass a clear quality bar before reaching customers, even if they are not flawless. - Beta users can help teams iterate from a solid initial release toward greater polish. - Defects are measurable, but excellence is often judged qualitatively through user perception and organic praise. ## Design with Strong Opinions - Products work best when designed for a specific audience and purpose. - Flexible products that serve everyone moderately well often fail to excel at anything. - A clear point of view strengthens the product’s value proposition and helps earn user trust. The practical recommendation is to make quality a company-wide operating principle: hire carefully, keep teams connected, exceed the basic specification, and design decisively for a particular user.

Read original

Figma User Experience Product Design Software Development+2

googleMar 17, 2025Original article

Generating synthetic data with differentially private LLM inference (opens in new tab)

Researchers at Google have developed an inference-only method for generating differentially private (DP) synthetic data that avoids the high costs and data requirements associated with private fine-tuning. By prompting off-the-shelf large language models (LLMs) with sensitive examples in parallel and aggregating their outputs, the approach can generate thousands of high-quality synthetic data points while maintaining rigorous privacy guarantees. This method allows synthetic data to serve as a secure interface for model development, enabling teams to collaborate without requiring specialized knowledge of differential privacy. ## Differentially Private Prediction and Aggregation The core of this method relies on "private prediction," where privacy is applied to the model's output rather than the model itself. * Sensitive data points are distributed across multiple independent prompts, ensuring that no single individual's record can significantly influence the final output. * The LLM generates next-token predictions for each prompt in parallel, which are then aggregated to mask individual contributions. * The researchers designed a DP token sampling algorithm that treats the standard LLM "softmax" sampling process as a version of the exponential mechanism, a mathematical framework used to select the best option from a set while maintaining privacy. ## Enhancing Efficiency via KV Caching Previous attempts at private prediction were computationally expensive because they required a fresh batch of sensitive examples for every single token generated. * A new privacy analysis allows the system to reuse a fixed batch of sensitive examples across an entire generation sequence. * By maintaining the same context for each generation step, the system becomes compatible with standard inference optimization techniques like KV (Key-Value) caching. * This improvement enables the generation of synthetic data at a scale two to three orders of magnitude larger than prior methods. ## Optimizing Privacy Spend with Public Drafters To preserve the "privacy budget"—the limited amount of information that can be released before privacy is compromised—the method introduces a public drafter model. * The drafter model predicts the next token based solely on previously generated synthetic text, without ever seeing the sensitive data. * Using the sparse vector technique, the system only consumes the privacy budget when the public drafter’s suggestion disagrees with the private aggregate of the sensitive data. * This is particularly useful for structured data, where the drafter can handle formatting and syntax tokens, saving the privacy budget for the actual content. By leveraging off-the-shelf models like Gemma, this approach provides a scalable way to transform sensitive datasets into useful synthetic versions. These synthetic datasets are high-quality enough to replace real data in downstream machine learning tasks, such as in-context learning or fine-tuning models like BERT, without the risk of leaking individual user information.

AI Large Language Models Machine Learning Differential Privacy+5

discordMar 16, 20253 min readCurated summary

Modern Image Formats at Discord: Supporting WebP and AVIF

Discord modernized its image pipeline to support animated WebP and AVIF attachments, embeds, and emojis. The change improves animation quality while significantly reducing file sizes compared with GIFs. Discord chose WebP as its primary delivery format because it offers broader platform support and more predictable performance, while converting AVIF sources—including tone-mapping HDR images to SDR when necessary. ## Benefits of WebP and AVIF - **Better transparency:** Both formats support full alpha channels, allowing smooth edges and animations that blend naturally with Discord’s light and dark themes. - **Greater color depth:** - GIF supports only 256 colors. - WebP supports roughly 16 million colors through 8-bit color channels. - AVIF supports up to 12 bits per channel. - **Improved visual quality:** Higher color fidelity reduces banding, produces more accurate colors, and better supports photographic and HDR content. - **Smaller animations:** Modern video-style compression provides: - Inter-frame compression - Advanced prediction - Configurable quality levels - Millisecond-precise frame timing - In the example provided, a GIF was **302 KB**, while the equivalent WebP animation was **43 KB**. ## Cross-Platform Image Processing - Discord processes and resizes every attached or embedded image to provide consistent behavior across: - Web browsers - Windows, macOS, and Linux - iOS - Android - Supporting new formats required coordinating inspection, conversion, decoding, and rendering throughout the media infrastructure. - Discord selected WebP as its preferred transformation target because it offers: - Near-universal platform support - Faster encoding and decoding than AVIF - Mature tools and ecosystem support - More predictable performance ## Tradeoffs Between WebP and AVIF - WebP is less capable than AVIF for: - 10- or 12-bit color - HDR content - Maximum compression efficiency - AVIF can produce smaller files for some content, but its platform support and performance are less consistent. - Discord converts AVIF images to WebP while preserving as much visual quality as possible. - HDR AVIF sources are tone-mapped to SDR because WebP is limited to 8-bit color depth. ## iOS Decoding Challenges - Discord uses SDWebImage to render images on iOS. - Its default WebP implementation relies on Apple’s ImageIO framework, available in iOS 14 and later. - Testing revealed that ImageIO decoded animated WebP files with increasingly inconsistent timing, especially for large or frame-heavy animations. - In a 10-second 480p30 comparison: - `libwebp` decoded the animation in real time. - ImageIO fell more than three seconds behind, with the delay increasing throughout playback. Discord’s move to WebP and AVIF demonstrates how modern formats can deliver higher-quality animations with substantially smaller files. For broad compatibility and reliable playback, WebP serves as the practical default, while AVIF remains useful as a high-efficiency source format.

Read original

iOS Image Processing Webp Avif+4

discordMar 16, 20252 min readCurated summary

The Game Developer Playbook: Three Incredible Game-Focused Communities

Discord’s Game Developer Playbook argues that successful communities should be inspired by other servers rather than follow rigid “best practices.” Engagement matters more than raw membership, and Server Discovery can help teams find useful ideas. The post highlights Fortnite, Rocket League, and Deep Rock Galactic as strong examples, with the provided section focusing on Fortnite’s organization and immersive community campaigns. ## Designing Around Community Needs - Discord servers evolve as their communities and marketing goals change. - Effective layouts should be easy for new members to understand and should clearly communicate the server’s purpose. - Server Discovery is recommended for studying both gaming and non-gaming communities. - Engagement and conversation rates are more meaningful measures of success than member count. ## Fortnite’s Server Structure At the time discussed, Fortnite’s server had more than 750,000 members while maintaining healthy engagement. Its categories were organized around common player needs: - **News:** Aggregates Fortnite updates, separating Discord-specific announcements from automated posts drawn from social media. - **LFG:** Divides matchmaking channels by playlist and game mode, allowing players to find teammates and join related voice channels. - **Discussion:** Uses broad, practical topics without overwhelming new members with too many channels. - **Bug Information and Reporting:** Provides known-issue information, reporting instructions, and game-mode-specific channels. Submission templates help standardize feedback. - **Community Help:** Enables members to support one another, while bots and integrations such as Zendesk or Helpshift can answer common questions and route support requests. ## Discord as an Extension of the Game Fortnite treated its Discord server as part of the game’s universe rather than merely a communication tool. During the Chapter Two Season Two “Top Secret” campaign: - Server artwork was updated with mysterious Agency branding. - A “redacted” role was secretly assigned to members who came close to correctly predicting details about the new season. - A custom Agency bot announced when selected members had been “activated.” - Activated members appeared separately in the member list, creating curiosity and fear of missing out. - The campaign connected Discord activity with Fortnite’s existing lore and marketing narrative. ## Practical Takeaway A strong game community server should provide clear utility—news, matchmaking, discussion, bug reporting, and support—while also creating experiences that feel unique to the game. Thoughtful structure combined with lore, events, and custom automation can make Discord feel like an extension of the game itself.

Read original

Discord Game Development Webhooks Community Management+1

discordMar 16, 2025Original article

Announcing Discord’s Social SDK, Helping Power Your Game’s Social Experiences (opens in new tab)

Discord has launched the Discord Social SDK, a free toolkit that allows game developers to integrate Discord's social infrastructure directly into their titles to drive player engagement and discovery. The SDK enables features like friends lists and messaging for all players, regardless of whether they have a Discord account, while offering deeper integration for those who choose to link their profiles. By tapping into Discord’s ecosystem of over 200 million monthly active users, the tool aims to help developers overcome discovery challenges in a market where a small number of franchises dominate total playtime. **Solving Discovery Through Social Connectivity** * With 20,000 games launching annually, Discord aims to leverage the fact that 50% of its users discover a new game on the platform every month. * Internal data suggests that playing with at least one friend increases gameplay session lengths sevenfold. * The SDK facilitates "friend-to-friend influence," where 28% of users launch a game within an hour of watching a friend stream it on the platform. **Core SDK Integration Features** * **Unified Friends List:** Synchronizes in-game and Discord friends lists, allowing players to maintain connections both inside and outside the game environment. * **Deep-linked Game Invites:** Enables players to send invites from their in-game list that allow Discord friends to join a specific party, lobby, or session with a single click. * **Rich Presence:** Displays real-time gaming activity on Discord profiles across PC, console, and mobile, serving as a passive discovery tool for a player's social circle. * **Flexible Account Requirements:** Developers can provide a unified social experience to all players without requiring a Discord login, though account linking unlocks more persistent social features. **Advanced Communication Tools in Beta** * **Cross-Platform Messaging:** A closed beta feature that allows players to continue game-related conversations across desktop, console, and mobile devices. * **Linked Channels:** Enables developers to sync in-game chat with specific Discord server channels, providing persistent messaging for guilds and squads. * **Integrated Voice Chat:** Grants developers access to Discord’s proprietary high-quality audio technology to power in-game voice communications. **Technical Specifications and Partnerships** * The SDK is currently available for C++, Unreal Engine, and Unity developers. * Initial support covers Windows 11+ and macOS, with console and mobile support listed as coming soon. * Early adoption partners include major studios such as Theorycraft Games, Facepunch Studios, 1047 Games, and Scopely. By providing these social tools for free, Discord is positioning itself as the foundational social layer for the gaming industry. Developers looking to capitalize on existing social graphs and improve player retention should consider integrating the SDK to bridge the gap between their game and the "digital living rooms" where players already spend their time.

Voice Chat Discord Social Sdk Rich Presence Unity+4

figmaMar 13, 20253 min readCurated summary

Dylan Field and Garry Tan on design, AI, and the power of “locking in” | Figma Blog

AI is expanding what designers and product builders can explore, but it has not eliminated the need for human judgment, context, or craft. Dylan Field argues that design will become even more important as teams use AI to generate ideas and prototypes faster, while still needing expertise to turn them into thoughtful, polished products. The central challenge is closing the gap between making something work and making it work well. ## AI Expands the Design “Idea Maze” - Current AI systems primarily function as tools that augment people in specific tasks, rather than as general intelligence. - They lower the barrier to participating in design while also raising the ceiling on what experienced creators can accomplish. - AI allows teams to explore more branches of an “idea maze,” generating greater breadth during ideation. - However, meaningful progress still requires depth: teams must investigate, refine, and evaluate promising directions. ## The Value of Rapid Feedback and “Vibe Coding” - Terms such as “getting locked in,” “I’m cooking,” and “vibe coding” describe the flow state created by rapid experimentation. - Faster feedback loops help people move ideas from their heads onto the screen more fluidly. - Figma’s emphasis on play reflects the goal of making creative expression accessible and enjoyable, even for non-experts. - AI tools are increasingly effective at helping users start and prototype quickly. - The unresolved problem is helping users move from an exciting prototype to a finished, reliable product—an issue shared by both design and code-generation tools. ## Design Is More Than Functionality - Founders and teams increasingly recognize design as a source of product value. - The important question is no longer only whether software works, but how it works. - User experience, clarity, quality, and the overall interaction determine whether a product feels successful. ## Why Human Designers Still Matter - AI has developed along partly separate tracks: diffusion models address visual creation, while language models focus on reasoning and code generation. - It remains unclear how effectively these approaches can be combined into systems capable of true design. - Field describes design as “art as it applies to problem solving,” requiring more than producing an image or implementing a short specification. - Designers contribute context involving culture, brand, product experience, and the broader problem being addressed. - As software creation becomes more automated, the ability to supply judgment and context may make design an even more critical role. AI is best understood as a force multiplier for exploration and iteration, not a replacement for design expertise. Teams should use it to accelerate experimentation while preserving the human work required to select, shape, and finish products well.

Read original

Figma AI Large Language Models Software Design+4

figmaMar 12, 20252 min readCurated summary

How eBay Built a Bridge Between Brand and Product with Figma | Figma Blog

eBay rebuilt its design-system documentation to unite brand, product, accessibility, and engineering guidance in one evolving experience. Its new Evo system is centered on the 300-plus-page Playbook, which replaces scattered, manually maintained files with a centralized source of truth. By making Figma the starting point for content and automating validation and publishing, eBay reduced documentation updates from days to minutes and encouraged teams to treat documentation as part of their normal workflow. ## Rebuilding from the Ground Up - Before Evo, documentation was spread across separate Figma files, developer resources, and accessibility guidelines. - Designers manually updated static files and submitted tickets for changes to be published. - eBay’s OneExperience team chose to rebuild the process internally rather than patch the old system or rely on an external agency. - The goal was a system integrated with everyday design and development work, rather than documentation maintained as a separate task. ## Bringing Brand and Product Together - The new Playbook combines brand guidance, design-system documentation, technology resources, and accessibility information in one destination. - It acts as a centralized source of truth for both internal teams and agency partners. - Component pages include design, development, and accessibility views. - Component availability and implementation status are updated from library metadata instead of manually maintained tables. ## Component Status Across Platforms - eBay created an internal Component Status API to track components across its libraries. - The API records component names, versions, availability, and implementation status across: - Published Figma libraries - Native platform libraries - Open-source web-component libraries - Figma components store relevant metadata in their descriptions. - Developers can verify whether a component exists in their framework, whether it matches the Figma version, and whether the Playbook documentation is current. ## Automating Documentation Updates - eBay built a custom Figma plugin and exporter that powers the documentation workflow. - Components, guidelines, and accessibility notes originate in Figma. - Updates are linted, validated, and published directly from Figma without manual CMS work. - Changes that previously took days can now appear on the Playbook site in under two minutes. - Faster publishing makes documentation easier to iterate on and helps teams view it as a living part of their workflow rather than an administrative burden.

Read original

Figma Design Systems Accessibility API+3

discordMar 10, 20253 min readCurated summary

Discord Patch Notes: March 11, 2025

Discord’s March 11, 2025 patch focuses on reliability, media processing, sharing, and broad bug fixes across desktop and mobile. Major improvements include safer API deployments, more accurate HDR-to-SDR conversion, faster and more reliable Spotify embeds, a redesigned Android sharing flow, and an Electron upgrade. The release also addresses numerous issues affecting chat, profiles, search, notifications, emoji pickers, and mobile rendering. ## Reliability and Platform Improvements - Discord changed its API deployment strategy to isolate critical services and reduce the impact of reliability incidents. - Desktop clients were upgraded from Electron 33.0.0 to 33.4.0. - HDR-to-SDR tone mapping was improved for more accurate colors in HDR images and videos converted for SDR display. - Spotify embeds now have approximately: - 83% lower p95 latency - 88% fewer failures - Android’s share interface was redesigned to resemble Discord’s message-forwarding experience. - Fixes may continue rolling out across platforms after the patch’s publication. ## General Interface and Account Fixes - Fixed Soundboard previews that could loop indefinitely after an upload was canceled. - Improved iOS performance when opening search and collapsing long channel topics. - Restored functionality for favorited GIFs and corrected poll avatars to use server-specific profiles. - Fixed iOS animation rendering failures and inconsistent Rich Presence display in member lists. - Improved mobile modal, emoji picker, GIF picker, and keyboard behavior. - Corrected several profile, server profile, custom status, pop-out, and profile preview issues. - Fixed server creation, friend request notifications, invite-link dropdowns, and Add Friends interface problems. - Improved desktop User Management sizing, Server Settings search, and member-search placeholder text. - Corrected mobile and desktop alignment, padding, hover states, button sizing, and masked channel-topic links. - Fixed duplicate known-device entries during two-factor authentication setup. - Message reaction notifications should now respect Do Not Disturb on mobile. ## Chat and Messaging - Empty Group DMs no longer display undefined users. - Fixed Server Guide popups that could obscure message editing. - Improved truncation and reliability for typing indicators. - Fixed iOS reply views where message text or input could become hidden. - Android users can now rename Group DMs. - Corrected Android reply themes and Emoji Confetti menu rendering. - Closing an unread DM on mobile no longer leaves an empty unread space in the server list. - Deleted messages can once again be acknowledged, preventing stale notification indicators. - Corrected Android permission interactions between sending messages and creating threads. - Fixed 24px chat font sizing, mobile DM alignment, and missing “Mark as Read” or “Reply” notification actions. ## Activities and Apps - The patch also begins addressing Activity behavior on iOS, including an issue involving Activity open and close sounds; the supplied post ends before the full Activities and Apps section is provided.

Read original

Android iOS Electron Spotify+2

figmaMar 10, 20253 min readCurated summary

From Screen to Zine: Meet the Makers Using Figma for Digital DIY | Figma Blog

Figma is helping a new generation of creators revive zine culture by combining digital speed with the handmade spirit of print. The article highlights designers, musicians, and community organizers using Figma to document underground scenes, produce physical artifacts, and collaborate remotely. Its flexibility, plugins, and collaborative canvas make it especially useful for highly iterative, visually experimental work. ## Figma and the revival of zines - Zines have historically given fringe communities and subcultures a way to publish outside mainstream media. - The article profiles several creators using Figma for digital DIY projects: - Dance Data’s 97-page *Underground Fanzine*, created by Jesse Pimenta and Cheyce Batchelor. - Figma’s AAPI employee resource group, which created *AZINE*. - Dominique Saiegh, who designs cassette and CD packaging. - Sebastián Ortega, who creates assets for online radio station SutroFM. - Tara Ridgedell and Gonzalo Guerrero, who run remote risograph workshops through Secret Riso Club. - These projects use digital tools to build real-world connections and preserve ephemeral music and cultural communities. ## The history and spirit of zine culture - Modern zines began as science-fiction fan magazines in the 1930s and 1940s. - Their history continued through: - Beat-generation small presses in the 1950s and 1960s. - Punk music zines and copy-shop publishing in the 1970s and 1980s. - Riot Grrrl and broader zine culture in the 1990s. - E-zines, blogs, and websites in the 2000s. - A renewed interest in printed zines, festivals, and libraries from the 2010s onward. - Dance Data’s publication deliberately recreates the scrappy, Xerox-inspired aesthetic associated with punk-era zines. - Printing the zine gives readers a lasting physical record of an underground music scene whose parties and collectives are often temporary. ## Designing a 1990s-inspired publication in Figma - Pimenta and Batchelor produced the zine in approximately three months. - Figma’s canvas provided a broad overview of the publication, while Auto Layout helped organize and reorder content efficiently. - Figma’s speed supported an iterative process in which every interview spread could have a distinct visual identity. - Pimenta drew inspiration from rave flyers, computer newsletters, technology magazines, and Archive.org’s *The Magazine Rack*. - The desired aesthetic combined hacker culture, cyberpunk imagery, retro computer graphics, and old-school clip art. ## Plugins for a photocopied, experimental look - **Noise & Texture** added gritty, speckled effects resembling 1990s photocopies. - **Photopea** enabled image manipulation, including noise and threshold filters. - **Skew Skew** distorted typography into irregular shapes that suggested warped Xerox reproductions. - These tools, combined with retro rave references, helped the digital design resemble handmade print work. ## Preparing digital work for print - Pimenta had to export individual pages at four times their intended size, convert them into print-ready PDFs, and then resize the final document. - Exporting was slow because the publication contained thousands of small assets. - The project illustrates both Figma’s usefulness for complex editorial design and the practical challenges of converting a large digital file into a physical zine.

Read original

Figma Figma Plugins Graphic Design Pdf+2

discordMar 6, 20253 min readCurated summary

Supercharging Discord Mobile: Our Journey to a Faster App

Discord’s mobile performance work focuses on making power-user workflows faster and more reliable, especially on lower-end Android devices. The team improved memory usage and frame rates through virtualization, native UI optimizations, recycling, and better media formats. When React Native could not meet performance needs, Discord chose targeted native implementations despite the added maintenance burden. ## React Native and Performance Priorities - Discord uses React for desktop and React Native for mobile. - Android moved to React Native in 2022 after improvements in device capabilities and the Hermes JavaScript engine. - The transition initially affected startup performance, but median startup times were cut in half during 2023. - Recent work targets heavily used features and users with large servers, busy chats, and extensive emoji collections. ## Virtualized Server List - The old server list kept every server view and icon—including animated icons—in memory. - The rewritten virtualized list retains only information for servers currently visible on screen. - The change also enabled improved drag-and-drop animations. - Users with more than 100 servers saw: - 14% lower startup memory usage - 10% faster overall startup time ## Native Chat Performance Improvements - Discord profiled its already-native chat list and found opportunities to improve loading, scrolling, and memory use. - Application Components and Media Mosaics now use recycling mechanisms. - Infrequently used elements, including spoilers and upload overlays, are loaded lazily. - Message containers and reactions were made faster to measure. - Recycling pools are pre-filled with common chat elements before a chat is opened. - Results included up to 60% fewer slow frames and approximately 12% lower chat-list memory usage. ## Rebuilding the Android Emoji Picker - Large servers can provide hundreds of custom emojis, many of them animated. - The previous React Native picker produced blank frames—and sometimes crashes—on budget Android devices during rapid scrolling. - Discord rebuilt the picker natively for Android rather than applying incremental optimizations. - Stress tests involving emojis from 200 servers eliminated blank frames and delivered smooth scrolling on low-end hardware. - Discord accepted the cost of maintaining separate Android and iOS implementations, while planning to reassess the decision with React Native’s New Architecture. ## Animated WebP Emojis - Discord’s media proxy added animated WebP support. - Replacing GIFs with WebP enabled animated emojis to render at up to 60 frames per second on low-end devices. - The team plans to publish more details about its animated-image improvements. ## Eliminating ChannelList Blanking - Discord initially replaced its homegrown React Native FastList with Shopify’s FlashList. - FlashList simplified development because components did not require precisely accurate dimensions. - In practice, FlashList blanked more frequently on low-end Android devices. - Discord restored and refactored FastList, eliminating blanking even on a $50 phone. Discord’s approach combines broad React Native development with native solutions where performance is critical. Virtualization, recycling, lazy loading, and efficient image formats provide general improvements, while platform-specific rewrites remain justified for demanding experiences such as emoji browsing.

Read original

Android JavaScript React Native Webp+4

discordMar 6, 20252 min readCurated summary

December Staff Picks: It’s Giving Games

Discord’s December Staff Picks explores the games employees have been playing recently and the gifted titles that shaped their lives. The contributors highlight a mix of MMORPGs, JRPGs, cozy multiplayer games, card games, and narrative adventures, emphasizing how games create lasting connections with friends and family. ## Scott’s Current Games - Scott, a member of Discord’s Quests team, recently introduced his husband to **Final Fantasy XIV**, a long-running personal obsession. - He is also playing: - **Metaphor: ReFantazio**, reflecting his enthusiasm for JRPGs and Atlus games. - **Marvel Rivals**, while trying to resist spending heavily on cosmetics. ## Games That Changed Scott’s Life - Scott identifies **NieR: Automata** and **Persona 5**, gifted by his best friend in 2018, as his most meaningful gifts. - **Persona 5** rekindled his interest in JRPGs and led him to explore more Atlus titles. - **NieR: Automata** stood out for its excellent gameplay, complex lore, tragic story, and emotional impact. ## Emi’s Cozy Multiplayer Lineup - Emi has been playing relaxed, social games such as: - **Kallax** - **WEBFISHING** - **Jigsaw Puzzle Dreams** - She is also progressing through **Yakuza Kiwami**. - Many of these games serve as a backdrop for chatting with friends. - Emi has recently played **Jackbox Party Pack 3** and **Jackbox Party Pack 6** with friends and family, using voice chat to stay connected across different states. ## Pokémon Blue and Family Memories - Emi’s most memorable gifted game is **Pokémon Blue**, which she and her brothers received alongside their Game Boys. - They spent time outdoors battling and trading Pokémon with link cables. - The game became an early foundation for shared gaming experiences with her family, a tradition that continues today. ## Cody’s Current Games - Cody has shifted from **Marvel SNAP** to the daily routine of **Pokémon TCG Pocket**. - He appreciates being able to open packs quickly without committing to a full match. - When playing competitively, he relies heavily on **Marowak EX**. - He is also playing **Apollo Justice: Ace Attorney Trilogy** and working through the wider Ace Attorney series. - The games’ exaggerated characters, unexpected twists, and bizarre humor have produced reactions ranging from laughter to shock. ## Continuing Through Ace Attorney - Cody plans to play the remaining entries, including: - **The Great Ace Attorney** - **Ace Attorney Investigations: Miles Edgeworth** - He estimates that he is nearly halfway through the full series. The post presents gaming as more than entertainment: gifted games can inspire lifelong interests, while multiplayer and family games preserve relationships across time and distance.

Read original

Video Games Multiplayer Gaming Steam Jrpg+2

discordMar 4, 20253 min readCurated summary

Discord Patch Notes: February 3, 2025

Discord’s February 3, 2025 patch focuses on faster startup, more reliable infrastructure, improved notifications, and extensive usability fixes across desktop and mobile. Major changes include a roughly 20% improvement in median Android launch times, a Rust rewrite of the Media Proxy service that reduced aggregate latency by about 22%, and substantially improved push-notification delivery. The release also adds desktop DM Apps, redesigns Privacy & Safety settings, and addresses numerous platform-specific bugs. ## Performance and Reliability - Android app initialization improvements reduced launch times by approximately 20% on the median Android device. - Discord rewrote its Media Proxy service from Go to Rust and modernized its infrastructure: - Aggregate latency improved by about 22%. - Latency at the 90th percentile improved by more than 40%. - Service reliability also improved. - Push-notification reliability increased across server sizes: - Servers with fewer than 2,500 users now achieve a 99.98% delivery rate within 30 seconds. - Discord can notify approximately 20 million users on large servers such as Midjourney within 15 minutes of an `@everyone` mention. - Android 13 and earlier received fixes for layout failures caused by device rotation and incorrect screen-size calculations. ## New Features and Interface Updates - Desktop users can now launch Apps directly in DMs. - A quick-launch bar displays recently used Apps at the top of the DM list. - Privacy & Safety settings were redesigned with: - A new tab structure. - Better grouping and clearer descriptions. - Improvements intended to support future Trust settings. - macOS desktop notifications now use native operating-system APIs: - Message sounds respect Focus and Do Not Disturb modes. - Multiple notifications from the same channel can be grouped. - Desktop search now autocompletes `from:me` and `mentions:me`. - Linux now supports Soundshare and Wayland screen sharing, including video and audio support on Ubuntu 22 and later. - Mobile users can edit server invite links while inviting someone. - Push notifications now include user nicknames. ## General Bug Fixes - Fixed UI and layout problems involving: - Forum channel tag editing. - Overlapping channel-topic modals. - Incorrect friend-request layouts at narrow desktop widths. - Misaligned images, padding, borders, and animations across Shop, Quests, profiles, and reaction pickers. - Resolved Android issues involving duplicated channel-name letters, hidden confirmation dialogs, incorrect toast positioning, keyboard overlap, and navigation-bar padding. - Fixed mobile and iOS issues affecting: - Search-result counts and back-swipe navigation. - Text-entry scrolling during channel searches. - Excessive zooming. - Animated Emojis stopping after interaction. - Nitro Home back-swipe navigation. - Corrected problems with Privacy & Safety tabs being cropped or nonfunctional. - Improved reliability for the Claimed Quests interface, especially for users with many completed Quests. - Fixed profile customization issues related to Nitro Credit. - Corrected custom status duration behavior. - Resolved issues with mutual-friend and mutual-server views getting stuck loading. - Fixed Emoji picker server filters and enabled editing of Emoji forum tags whose associated Emoji had been deleted. - Corrected several Shop, tablet banner, profile-effect, and mobile DM layout issues. Overall, the patch combines measurable backend performance gains with a broad collection of cross-platform fixes. Users should see faster Android launches, more dependable media uploads and notifications, improved desktop and mobile navigation, and fewer layout and interface failures.

Read original

Android Rust Go Linux+3

microsoftMar 3, 20252 min readCurated summary

How Microsoft Engineers Build AI: Learn about scalable RAG-enabled AI Apps

Microsoft’s new *How Microsoft Engineers Build AI* video series explains how its teams develop AI applications at scale. The first episode focuses on retrieval-augmented generation (RAG), using Copilot for Azure’s Ask Learn plugin as a practical example. It shows how RAG can combine proprietary data with large language models to deliver accurate, contextually relevant answers. ## Building AI Applications with RAG - RAG is presented as a practical way to improve AI applications without relying solely on model fine-tuning. - It retrieves relevant information from a knowledge base and provides that context to an LLM when generating responses. - The approach is useful for applications that need current, domain-specific, or proprietary information. ## The Ask Learn Plugin - Microsoft engineers explain how they built the Ask Learn RAG plugin for Copilot for Azure. - The plugin helps Azure developers find answers quickly within their existing workflow. - The project involved product managers and engineering leaders sharing development challenges, design decisions, and best practices. ## Challenges in Developing Reliable RAG - Selecting the right source content is essential for producing useful answers. - Data must be preprocessed effectively before it can be retrieved. - RAG systems require careful performance evaluation to measure accuracy and relevance. - Keeping responses accurate and up to date requires ongoing improvements to content and retrieval methods. ## Broader Microsoft Applications - The episode discusses RAG implementations across: - Copilot in Azure - Microsoft Security Copilot - Dynamics 365 Business Central - These examples demonstrate how RAG can support different products and business scenarios. The episode is intended as a practical introduction for developers building RAG-based applications, covering prototyping, data management, evaluation, and common pitfalls. Developers can explore the series alongside Microsoft Learn resources and Azure AI development tools such as Visual Studio and GitHub Copilot.

Read original

AI Large Language Models Machine Learning Github Copilot+4

figmaMar 3, 20252 min readCurated summary

Why Everyone Can (and Should) Be a Great Storyteller | Figma Blog

The article argues that storytelling and writing are not specialized talents reserved for professional writers; they are skills anyone can develop. Clear communication helps turn technical ideas, product plans, and design decisions into shared understanding and action. The difference between “good” and “bad” writers is largely practice, feedback, and willingness to revise—not innate ability. ## Why Storytelling Matters at Work - Strong products, designs, and ideas depend on people who can explain: - Why something should exist - Who it serves - How it improves people’s lives - Writing helps teams bridge the gap between imagination and execution. - At Figma, engineers write technical proposals, designers explain their decisions, and product managers create PRDs to align teams. - Companies such as Microsoft and Slack also use writing to encourage deep thinking, clear communication, and buy-in. ## Challenging the “Bad Writer” Myth - Many people mistake their rough first drafts for the polished writing they read elsewhere. - Every writer begins with imperfect work; improvement comes through: - Feedback - Editing - Deliberate practice - Writing is described as organized thinking: putting ideas on paper can clarify problems and lead to better decisions. - The ability to think clearly is a foundation for becoming a better writer. ## Overcoming the Blank Page - Speaking an idea aloud or explaining it to someone else can reveal a clearer, more natural structure. - Reading widely—including books, magazines, product packaging, and online discussions—helps develop writing ability. - Leslie Lamport’s observation captures the central lesson: without writing, people may only think they are thinking. The practical recommendation is to treat writing as a learnable muscle: read often, write imperfect drafts, seek feedback, and use storytelling to make ideas understandable and actionable.

Read original

Design Thinking Product Management Technical Writing Storytelling+2