• Artwork@lemmy.world
    link
    fedilink
    English
    arrow-up
    2
    ·
    4 hours ago

    Based on the handwriting style, Nicolardi thinks that PHerc. 1667 dates to the second or third century B.C.E., which would make it one of the oldest scrolls in the library. The title and author have been lost to history, but experts suspect it wasn’t written by Philodemus. Perhaps, they theorize, Chrysippus was the author.

    Chrysippus was a foundational figure in Stoicism, but modern scholars only know of his writings from references in other works. “To have access to a source text rather than quotes and summaries, which can be modified or interpreted by other writers, is very important,” Thomas Coward, a classicist at the University of Bristol in England who isn’t part of the Vesuvius Challenge, tells New Scientist’s Hayley Bennett.

    Source1: https://lemmy.world/post/48737520
    Source2: Smithsonian Magazine [2026-06-26]

  • jpreston2005@lemmy.world
    link
    fedilink
    arrow-up
    17
    arrow-down
    1
    ·
    2 days ago

    I don’t mind if AI is used for cool shit like this, I bet if we limited it to these types of applications we wouldn’t need 20 data centers positioned all around my city, destroying the environment with noise pollution, wasting something like 15 million gallons of water EVERY DAY during a severe drought, all while getting discounted electricity that the rate hikes on regular citizens pay for.

    • valar@lemmy.ca
      link
      fedilink
      arrow-up
      29
      ·
      2 days ago

      People should keep in mind these scientific advances come from purpose-built machine learning algorithms and not like, ChatGPT or Grok. It’s really not the same thing.

      • Artwork@lemmy.world
        link
        fedilink
        English
        arrow-up
        2
        ·
        edit-2
        4 hours ago

        Thank you…

        The term AI existed long prior, and there are numerous types of it, including LLM, the recent, yet it still not the AI since it doesn’t think but outputs the most probable output depending on the model trained on dataset and algorithm, and the prompt. The recent use of it, including vibe-condig and “art” generation based on stolen actual art and knowledge from now unknown and defaced people by the common “AI” vendors who even sell it… darkened it all so indescribably much…
        And not a single person is responsible for the output of these models… with no copyright nor law… That is, a pure dark sorrow of noise and cries… of authentic and effortful hard work by real, now unknown, artists, developers, people… inside these models…

        Thankfully, this usage, for supporting a human to do a human work, accountably, responsibly, and carefully, is heartfelt appreciated indeed…

        If you know, may I ask you provide machine-learning systems used in actually accountable, authentic science like that to read more and compare about? This would help as a balm on heart…