I’ve been reading a thread on here and it hits on something that’s been bugging me while looking into a project I stumbled across recently called Frog, it’s an interesting attempt to rethink what “no-code” could mean in the AI era.
A commenter on here made a great point: traditional “no-code” tools (Wix, Squarespace, Shopify, Gumroad, etc.) succeeded because they picked a narrow, well-defined slice of functionality and made it dead simple. They didn’t try to be general-purpose. Once a no-code system starts handling arbitrary logic, state, or data flow, it quietly turns into a visual programming language, and simplicity dies.
But what seems different now is that AI-driven apps are going after an entirely new cohort, not people who want to wire together APIs or drag and drop logic, but people who just want to describe what they want built in natural language and let the system infer the rest. It’s a completely different UX paradigm, less about hiding the code, more about automating the act of thinking in code.
That’s where I think something like Frog gets interesting. Should projects like this really aim to be traditional no-code tools, or should they focus on being accessibility and efficiency amplifiers, leveraging AI to bridge intent and implementation without collapsing into the same “visual programming” trap?
A few questions I’d love to get the community’s perspective on:
1. Do you view AI-driven tools (like Replit’s Ghostwriter, v0.dev, or Frog) as a genuine continuation of the no-code movement, or something completely separate?
2. How much abstraction is too much before users lose agency or predictability in what’s being built?
3. Historically, no-code succeeded by constraining scope, is there any realistic path for AI tools to stay simple and general-purpose at the same time?
reieicucv•4mo ago
A commenter on here made a great point: traditional “no-code” tools (Wix, Squarespace, Shopify, Gumroad, etc.) succeeded because they picked a narrow, well-defined slice of functionality and made it dead simple. They didn’t try to be general-purpose. Once a no-code system starts handling arbitrary logic, state, or data flow, it quietly turns into a visual programming language, and simplicity dies.
But what seems different now is that AI-driven apps are going after an entirely new cohort, not people who want to wire together APIs or drag and drop logic, but people who just want to describe what they want built in natural language and let the system infer the rest. It’s a completely different UX paradigm, less about hiding the code, more about automating the act of thinking in code.
That’s where I think something like Frog gets interesting. Should projects like this really aim to be traditional no-code tools, or should they focus on being accessibility and efficiency amplifiers, leveraging AI to bridge intent and implementation without collapsing into the same “visual programming” trap?
A few questions I’d love to get the community’s perspective on:
1. Do you view AI-driven tools (like Replit’s Ghostwriter, v0.dev, or Frog) as a genuine continuation of the no-code movement, or something completely separate?
2. How much abstraction is too much before users lose agency or predictability in what’s being built?
3. Historically, no-code succeeded by constraining scope, is there any realistic path for AI tools to stay simple and general-purpose at the same time?