# Summary of code walkthrough of Hyperfy v2 ## Summary (Artisinal) > **Model Used:** maximus > **Prompt:** Summarize and takeaways > **Link to Source:** [Twitter](https://x.com/i/broadcasts/1BdxYEygmLgxX) > **Summary:** Ash from the core development team sharing initial release of Hyperfy v2 code ### Informal review of the code released today: * As background: Hyperfy is an engine/platform for immersive and interactive apps * Apps behave like browser plugins sandboxed from each other using are secure SCS (private compartments) * Apps will be configurable and the code will also be editable * Is a refined version of javascript to allow simpler coding without relying on more complex frameworks * Goal is to be able to create whole self-contained worlds and modular apps that can be readily shared and used * As easy to start as build apps should start with a 3d model, attach scripts and configuration on it ### Currently 2 main git repos opened up by Hyperfy * Reference Engine (reimagining of the v1) * There’s a post that describes things that can be done such as mad instancing, physics * Coding in browser enabled here; eliminating the need to do coding/development like it was needed for v1 * Intended as a test build not fully hostable (ex. Uses Firebolt) * Hyperfy Self-hostable Engine (v2 repo) * Crafted to be be self-hostable * Does not have all the features for Reference Engine (but will be updated soon to have everything the reference engine has) ### Questions: starting from 7:28pm** * Can we back up worlds? Yes, technically create fully backed up worlds embedded in NFT * 7:32 Will all v1 emotes work in v2… for example you can create an object and combine with an emote * 7:36 = For someone who doesn’t want to host what options will they have? Eventually there will be packaged providers that can provide the back-end * Can we use blockchain? Yes, will be chain agnostic, chains will be supported by apps * 8:29pm Q: Pathfinding for agents? Yes, better than v1. Can use pathing as well as “follow” commands * 8:33 Q: Avatar backpacks? Yes, planned to improve data portability * 8:36 Q: Policing malicious apps? Users will be responsible for their pages, no central authority since things are self hosted * 8:38 Q: App ratings? We thought of it but it isn’t necessary yet, not our place to police/host a market place * Will streaming work from v2? A: Will require porting apps from v1 primarily supporting HLS and m3u8 * Does flying work? A: Yes PhysX integrated can really fly! * How big can a world be? A: Very big, limited by javascript floating point * How to control avatars to sit or aim? 1) That is an object app with an emote bound to it + 2) A core app is needed to change perspectives to do things like FPS perspectives ### Ash spinning up a of v2 repo pull * Meant to be simpl npm install with very basic dependencies (meant for easy maintenance) * Basically a git clone npm etc and it spins up an Sqlite database to hold data * Intent is to run on low compute servers such as 512mb (easy self hosting) * 8:16 talking about how engine in it’s purest form is basically a javascript handler for DOM/colliders/LOD to create immersive worlds rich with plugins ### Key Next Steps (7:39pm – 7:51 + 8:03) * 1. Persistence for worlds to be fully self-hostable * 2. Identity service layer will be created (so people can save their Avatar accessible across worlds) * 3. Code editor will be moved into * 4. World import /export capability (mirror) * 5. Versioning app scripts (making it so apps work forever) * 6. Docs plzzz * 7: 40 = Requests to the public * Try it out, give feedback * If you are keen and understand how the 3js in working :: We have apps from the reference engine AND v1 that needs things ported over to the core engine ### Far Future: * Backpack files across worlds * Let people whitelist certain types of apps to be usable in app ## Summary (LLM) > **Model Used:** OpenAi o3 > **Prompt:** Summarize transcript into Platform | Code | Features | Questions and Next Steps > **Link to Source:** [Twitter](https://x.com/i/broadcasts/1BdxYEygmLgxX) > **Summary:** Ash from the core development team sharing initial release of Hyperfy v2 code ### Overview of the Hyperfy Platform * Primary Objective: To simplify virtual world-building and interactivity, targeting accessibility for both developers and non-developers. * Core Innovations: * In-browser coding capability, eliminating SDK installation requirements. * Persistent, self-hostable worlds with drag-and-drop features. * Export functionality for interoperable assets in a single file. * Introduction of cryptographically signed apps for secure sharing and integration. * Streamlined resource requirements allowing for cost-efficient hosting (e.g., servers with 512MB RAM). * Potential Applications: * Embedding apps in NFTs for innovative use cases. * Creating open ecosystems akin to Minecraft servers, with enhanced portability. * Modular marketplace concept for assets and apps. ### Code Development Experience* Development Flow: * Developers can start with basic commands like npm install to spin up a local environment. * World-building involves creating directories containing SQLite databases and asset folders. * Drag-and-drop mechanics streamline 3D asset integration, with minimal dependencies for faster deployment and testing. * Flexibility for Customization: * Developers can embed specific functionalities (e.g., emotes, interactive objects) using pre-set configurations. * Physics and interactive object creation are simplified through customizable scripts. * Ease of Deployment: * Worlds are scalable and can support hundreds of concurrent users. * Features include modularity for switching between world configurations and seamless server transitions. ### Showcase of Key Features * Interactive World Dynamics: * Real-time collaboration allows users to edit or modify elements in live environments. * Apps can take over user controls for immersive experiences (e.g., sitting animations, custom interactions). * Enhanced App Integration: * Apps function as modular elements, providing flexibility for adding functionality like physics-based objects or Blockchain integrations. * Sandbox environments isolate app behaviors, ensuring world stability. * Performance Highlights: * Efficient handling of assets and physics-based simulations (e.g., multi-player synced environments). * Supports lightweight hosting solutions, enabling low-cost scalability. * Customization Capabilities: * Developers can create unique controllers for specific use cases, such as first-person or racing simulations. * Compatibility with existing assets and flexibility to integrate advanced features like AI agents or live media streaming. ### Open Questions * How will media streaming evolve in V2 to accommodate user-hosted solutions? * What enhancements are planned for avatar interaction and customization (e.g., integrating dynamic emotes)? * How will Hyperfy encourage third-party contributions while maintaining platform integrity? ### Next Steps * Finalize integration of critical features like persistence, app import/export, and modular configurations. * Establish tools and support for developers, particularly in areas like identity management and advanced scripting. * Explore collaborative opportunities for creating ecosystems, such as marketplaces or pre-configured app bundles. * Focus on user feedback to refine the platform’s usability and scalability. {%hackmd Syq4xkPwkg %}