Login
You're viewing the functional.cafe public feed.
  • Nov 17, 2026, 11:01 AM

    @aud to add more craziness: metr actually measured, and they found that while programmers self-report expectations of high efficiency gains, none of these turn out to be real when measuring their actual time spent on a task with or without LLMs.

    In response to that metr …

    … what do you guess? Any respectable answer would be to publish that self-reported efficiency gains are still bullshit. But metr …

    … stopped measuring and now only does "research" on self-reported questionnaires.

    @lina

    💬 0🔄 1⭐ 8