I. Overview
On August 8, Google announced the launch of the AI code editor IDX, which aims to provide a browser-based artificial intelligence development environment for building full-stack networks and multi-platform applications. Google didn't build a new IDE (Integrated Development Environment) when they created IDX, but used VS Code as the basis for their projects. Currently, IDX supports frameworks such as Angular, Flutter, Next.js, React, Svelte, and Vue, as well as languages such as JavaScript and Dart, and will support Python, Go, and other languages in the future.
It is understood that IDX is a complete browser development experience built on Google Cloud and supported by Codey. The IDX project is also based on the Code OSS editor. At present, IDX has not been opened to a large area, and only a few designated testers can participate in the experience in person. However, the IDX team revealed that at the upcoming Google Cloud Next conference, there may be more news about Codey's existing achievements and future development plans to meet with developers. The IDX team stated:
We spend a lot of time writing code, and recent advances in AI have created a huge opportunity space to use our time more efficiently. Through the IDX project, we are exploring how Google's innovations in AI, including the Codey and PaLM 2 models powering Studio Bot in Android Studio, Duet in Google Cloud, and more, can help developers write faster and faster code. Writing quality.
2. Google releases AI code editor IDX
The IDX team blogged that today, building an app from scratch—especially one that works well on mobile, web, and desktop platforms—is no less difficult than building a Rube Goldberg machine. machine, a mechanical assembly designed to be overly complex). Developers need to navigate an endless sea of complexity, glue various technology stacks together, and struggle to find a way to correctly boot, compile, test, deploy and monitor applications.
Although Google has been committed to reducing the difficulty of developing multi-platform applications for many years, and has successively launched achievements such as Angular, Flutter, Google Cloud and even Firebase, it seems that it can do better. After all, the current multi-platform application development is far from fast and smooth. So a few months ago, a few small partners at Google got together and started experimenting, and an early result of this experiment was the IDX project ( http://idx.dev/ ).
One of the important reasons why the team decided to build the IDX project is to listen to the opinions of the broad developer community and understand which elements can help everyone improve work efficiency.
3. IDX Features
Specifically, the IDX project currently mainly implements the following functions:
Work anytime, anywhere
At the heart of the IDX project is helping developers develop anywhere, on any device, with a full-fidelity local development experience. Each workspace in the IDX project is fully functional as a Linux-based virtual machine with universal access hosted in the developer's adjacent cloud data center.
Import existing applications and create new ones
IDX projects allow developers to import existing projects from GitHub and pick up where they left off at any time. Developers can also create new projects using pre-made templates for various popular frameworks, including Angular, Flutter, Next.js, React, Svelte, Vue, as well as JavaScript and Dart, with support for Python and Go languages coming soon. Additionally, the IDX team is actively working to provide optimal support for more project types and frameworks.
Implement app preview across platforms
Today, building a successful app means optimizing the design and behavior of the app across platforms, and previewing the effect of the app in a "what you see is what you get" way. To lower the barrier to entry for all of this, the IDX project offers a built-in web preview and will soon have a fully configured Android emulator and an embedded iOS emulator. All of these can be used directly in the browser.
Combined with AI technology
Currently, the AI capabilities of the IDX project are in their early stages, but already have smart code completion, an assistive chatbot, and contextual code actions like "add a comment" and "explain this code."
Web Publishing with Firebase Hosting
A common pain point of putting an application into production is the deployment process. The IDX project integrates with Firebase Hosting to ease the entire operation, and with just a few clicks, you can deploy a sharable preview of your web app, or deploy it to production using a fast, secure global hosting platform. Since Firebase Hosting supports a dynamic backend based on Cloud Functions, it works well with full-stack frameworks like Next.js.
The IDX team stated that the current IDX project has just started, and there is still a long way to go before the final end-to-end development process improvement. It is hoped that developers can register to be the first batch of IDX project experiencers. The problems discovered together will guide the direction of subsequent iterations of the project and what new features are needed to match the workflow of the application team.
As for the next step, the IDX team said that it will continue to work hard to add new features and solve problems reported by developers. “We are already working on new collaboration features as we know how important this is in today’s hybrid world. Also, deeper framework integrations and personalized/contextualized AI options are in the pipeline. Looking forward to hearing Everyone's request for more features!"
Codey, the AI model behind IDX
At the Google I/O 2023 conference, Google officially released Codey. It's a new AI-driven tool that writes and understands code. The new tool is seen as Google's answer to GitHub Copilot, the result of an alliance with Replit.
Codey is based on Google's next-generation large language model PaLM 2, and uses Google's own product code and a large number of legally licensed source codes as training materials. More importantly, Codey is still learning and developing, and continues to draw new strengths from various projects in the Google service ecosystem.
Codey supports more than 20 programming languages, including Go, Google Standard SQL, Java, JavaScript, Python, and TypeScript. Developers can access Codey through extensions for Visual Studio Code, JetBrains IDE, Google Shell editor, and Google Cloud's hosted workstation service. Developers can communicate with the model directly in the IDE's chat box (such as the Android Studio Bot), or write comments in the text file to instruct it to generate the relevant code. It supports a variety of coding tasks, helping developers work faster and close the skills gap by:
- Code Completion: Codey suggests the next few lines based on the context of the code entered in the prompt.
- Code generation: Codey generates codes based on the developer's natural language prompts.
- Code Chat: Codey allows developers to chat with a bot to get help with debugging, documentation, learning new concepts, and other code-related issues.
Codey was trained specifically to handle coding-related prompt words, and Google also trained the model to handle general queries about Google Cloud.
Reference link: