One AI Chat Is Not a Research Workspace
Originally published on omid.dev . I was deep in a research thread that was not going to become a weekend post. The topic started small: juniors asking an AI to format code that Prettier already owns, or to invent a debounce helper the repo already has. It got larger fast. Sometimes the person is not a developer at all — they have an idea, they paste a warning into ChatGPT, the "build" goes…
One AI chat session is not equivalent to a research workspace, according to a report from omid.dev. The article details how the author, who was immersed in a research thread, found that junior developers were using AI tools to handle tasks they should have learned themselves, leading to a loss of essential knowledge and understanding.
The researcher observed that AI could create more complexity than the user's mental model allowed, and that the conversation itself became overly reliant on scrollback, making it difficult to keep track of discussions. The report argues that a chat is a scratchpad, which works well for moving an idea forward but fails as a place to keep notes for a research project.
This is because chats are linear, lack the ability to hold contradictory viewpoints, and ultimately cannot retain the messy, contradictory, and duplicated thoughts that are crucial for early research. The author suggests creating a separate workspace, organized by folders like "inbox," "research," "concepts," "examples," "counterarguments," and "drafts."
This workspace should allow for messy, contradictory, and duplicated thoughts, as organization is a later step. The idea is to capture raw thoughts without judging them, allowing for later categorization and refinement.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
