text stringlengths 0 2.02k |
|---|
• Dino as a "second shot" or "another round" at creating a better JavaScript runtime |
• Dino is an effort to move the server-side JavaScript platform forward with radical changes |
• Node has maintained backwards compatibility, but is slow to adopt new changes |
• JavaScript has evolved significantly, with async await syntax and ES modules syntax being major changes |
• Dino's TypeScript aspect is a key feature, but also a point of controversy as browsers do not support TypeScript |
• The decision to support TypeScript out of the box may have been a mistake |
• Dino is trying to bring server-side JavaScript closer to browser JavaScript |
• Dino's goals and philosophy are to be web-compatible and bring modern features to server-side development |
• Dino is written in Rust, unlike Node, which is written in C++ |
• Rust's cargo system for linking different libraries is a key advantage |
• Complexity of linking libraries in C++ leads to complexity and difficulty |
• Rust's ability to easily link third-party libraries makes it easier to manage dependencies |
• Deno's bindings system (ops) organizes JavaScript-native language interactions |
• Everything in Deno is organized as promises, reducing callback chaos |
• WebSockets and async await paradigm solve back pressure problems |
• Package management story in Deno aims to use web APIs, avoiding invented APIs |
• Node.js module system limitations |
• Introduction of ECMAScript modules and imports/exports |
• Dino module system, using HTTP to link remote code |
• Advantages of Dino system: immediate feedback, no need to install packages ahead of time, minimal boilerplate |
• Comparison to NPM and traditional module systems |
• Benefits for small scripts and single-file projects |
• Reducing boilerplate with Dino |
• Dino's use of URLs for dependency management, reducing dependence on specific servers |
• Versioning with Dino, using URLs to specify versions |
• Compatibility with existing NPM packages, potential issues with non-ESM packages |
• Security features of Deno, specifically secure by default |
• Pulling in random packages over HTTP can be a security risk due to unaudited dependencies and potential execution of arbitrary code |
• Node.js has no built-in security sandboxing, making it vulnerable to attacks if an NPM package is compromised |
• Dino is a server-side JavaScript system that aims to mitigate these risks by defaulting to a secure sandbox and allowing privileges to be granted via command line flags |
• Dino has a centralized system for calling from JavaScript into Rust, which provides centralized gating for security and allows for privileges to be controlled |
• Dino allows for privileges to be granted via command line flags, including allow write, allow run, allow read, allow plugin, allow net, and allow end |
• Security is a top priority for Dino, and the team is conservative about adopting new file formats, opting for command line flags instead of configuration files |
• Discussion of the Dino tool and its security features |
• Explanation of the concept of "shift left" in software development |
• Introduction of a new feature in Dino, "dash dash prompt", which provides an interactive prompt for users to grant or deny system access |
• Overview of other features and tools in Dino, including code formatter, linter, test runner, and documentation generator |
• Discussion of the inspiration behind Dino's design and features from other programming languages and ecosystems |
• Dino's ecosystem is still developing and lacks some packages available in Node.js |
• Dino's standard library is similar to Node.js and provides many similar tools |
• Dino's HTTP2 web server is still a work in progress, with recent improvements |
• Native web servers using Rust are being explored for better performance |
• Dino 1.9 includes an unstable native web server with good latency and throughput |
• Performance analysis is ongoing, with stabilization planned in the next few months |
• Discussion of Node and dynamic programming languages |
• Importance of JavaScript and its future in the web |
• Comparison of dynamic programming languages (JavaScript, Python, Ruby, Perl) |
• Empowerment of scripting languages for fast software development |
• Bet on JavaScript as the most suitable language for current needs |
• Replacement of utility scripts with Dino, a JavaScript-based alternative |
• Comparison of Make and Rake, with Rake being a wrapper for Make |
• Bash programming language is not well-liked and is considered hard to use |
• Accessing documentation for syntax is common and easier than learning a new language |
• JavaScript is considered a better choice due to its ubiquity and versatility |
• Node.js was created and sold to Joyent, and the creator went on to work at Google |
• A collaborator on Node.js worked on a Windows port, which was a significant undertaking |
• Node.js is now well-supported on Windows and runs fast |
• The speaker discusses the founding of Strongloop by Burt, which was later sold to IBM |
• The speaker and Burt left their respective companies to start a new project, which eventually became Dino |
• The speaker reflects on the challenges of building a software system like Node, which requires a large team and funding |
• Dino requires a funding model to scale, unlike open-source software like Node |
• The company's growth began with the speaker and Burt working on the project for a year, followed by hiring a first engineer in 2019 |
• The speaker discusses the choice of the MIT license and the challenges of building an open-source business |
• The company considers using the open core model, where open-source software is free but enterprise features are charged for |
• Payment hook concerns with Dino |
• Alternative funding model for Dino |
• Dino Deploy: a cloud-based, serverless JavaScript runtime |
• Architecture of Dino Deploy: Anycast IP, multiple data centers, and V8 isolation |
• Comparison to AWS Lambda and Cloudflare Workers |
• Lightweight, secure sandbox architecture |
• Commercially applicable ability to use Dino Deploy |
• Relationship between Dino Deploy and Deno |
• Discussing potential commercial applications of the company's infrastructure |
• Mentioning the "Deploy" application and its limitations |
• Teasing potential future applications, including Electron and GUI frameworks |
• Explaining Deno's design philosophy and lack of a specific Deno API |
• Describing the Deno Deploy product and its current state (open beta, not yet production-ready) |
• Discussing plans for converting Deno Deploy into a commercial product |
• Mentioning potential competition from Dino and the possibility of competing with Node.js |
• Encouraging competition with Dino and welcoming new competitors |
• Investing in Rust and JavaScript infrastructure |
• Serverless solutions and database co-location |
• Dino Deploy as a minimal viable product |
• Opportunities for developers to contribute to Dino's growth |
• Contributing to the standard library as a way to make a significant impact |
• Contributing to the standard library through adding modules or utilities |
• Dino and its repository on GitHub |
• Contributing to Dino, including creating issues and discussing new ideas |
• Dino's relationship to other languages and systems, including Rust, TypeScript, and Node.js |
• Porting existing Node.js applications to Dino |
• Tooling and workflows for Dino development |
• Community and resources for Dino, including a Discord channel |
[0.16 --> 7.38] Hey, welcome back. You are listening to the Change Law. We feature the hackers, the leaders, and the innovators in the world of software. |
[7.86 --> 18.80] Today, we're joined by Ryan Dahl, Node.js creator, and now the creator of Dino, a simple, modern, and secure runtime for JavaScript and TypeScript that uses V8 and is built in Rust. |
[19.10 --> 26.78] We talk with Ryan about the massive success of Node and how it impacted his life, how he eventually created Dino, and what he's doing differently this time around. |
[26.78 --> 30.42] We also talk about the Dino company and what's in store for Dino Deploy. |
[30.88 --> 33.98] Big thanks to our partners, Linode, Fastly, and LaunchDarkly. |
[34.20 --> 39.36] We love Linode. They keep it fast and simple. Check them out at linode.com slash changelog. |
[39.68 --> 43.94] Our bandwidth is provided by Fastly. Learn more at fastly.com. |
[44.22 --> 48.44] And get your feature flags powered by LaunchDarkly. Get a demo at launchdarkly.com. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.