Live data from Hacker News

Viewing profile — IsaacL

IsaacL

HN member
Joined
Thu, Jun 04, 2009, 8:52 PM UTC
HN karma
2,169
Public activity
694 items

About IsaacL

http://isaacklewis.com

isaac.kaspar.lewis@gmail.com

Recent public activity

  1. comment
    Comment #47204980

    It was obvious to me as well that your comment was not AI-generated. I thought I'd reply as I like reading in-depth discussions online, so I don't want to see a future where everyo…

  2. comment
    Comment #47204976

    There's too many incidental details, too much substance and complexity for this to be an AI-generated comment. (I know this is an 8-day old thread, but accusing a long comment of b…

  3. comment
    Comment #43075841

    I commented about this on X. As a developer, "infinite scroll to get more digits" sounds really cool. It sounds conceptually similar to lazily-evaluated sequences in languages like…

  4. comment
    Comment #43075779

    This relates to what I wrote in reply to the original tweet thread. Performing arithmetic on arbitrarily complex mathematical functions is an interesting area of research but not u…

  5. comment
    Comment #42016095

    I still maintain that this (originally from ancient China) is the clearest proof, and gives the best insight into why the Pythagorean Theorem holds. https://cdn.britannica.com/43/7…

  6. comment
    Comment #38415165

    I agree with you politically and spiritually. However, Patrick Mackenzie is a person of the highest calibre (see, e.g., this: https://worksinprogress.co/issue/the-story-of-vaccinat…

  7. comment
    Comment #38226260

    This is a terrible article that rambles on for far too long without offering an actually polite way to tell/request someone to do something. These are the best approaches: 1. "Plea…

  8. comment
    Comment #36417765

    Quotes from the Memo: "We will distribute tools to leverage the power of these white box models, and create a developer community around our trademark. This approach is an ideologi…

  9. comment
    Comment #36417569

    I didn't notice that line, but my immediate thought on reading the memo was that they were going hard after European capital. Seemingly both governmental and from VC firms in Europ…

  10. comment
    Comment #36375100

    In memory of the olden days when Hacker News focused on discussions for entrepreneurs, here's a question: I have no doubt that hundreds of millions of BS jobs exist, but isn't that…

  11. comment
    Comment #36036839

    I asked a follow-up question: Where did Person get the spoon from? ChatGPT's answer: I apologize for the oversight. Here is the corrected version, including the step to get the spo…

  12. comment
    Comment #36036784

    I ran an interesting session with ChatGPT. I don't think the following proves that it has a "world model", but it does seem to be able to emulate having a world model to an impress…

  13. comment
    Comment #35712248

    This looks like a very interesting paper that takes the rare approach of actually trying to understand what all the cool new language models are doing at a fundamental level. Does …

  14. comment
    Comment #35670083

    I pretty much agree. The "scientific" approach the author pushes for in the article -- running experiments with multiple similar prompts on problems where you desire a short specif…

  15. comment
    Comment #35523928

    Funnily enough, I was reading Minsky's book recently. I second the recommendation. I think he's missing many technical details*, but the basic approach seems to be correct. *(For e…

  16. comment
    Comment #35204764

    In the same vein, I was playing around with its ability to annotate code with explanations of the changing constraints on different variables (I thought this was a good proxy for d…

  17. comment
    Comment #35200246

    If they're willing to learn, GPT is a very powerful tool. https://www.youtube.com/watch?v=VznoKyh6AXs ["How to learn to code FAST using ChatGPT (it's a game changer seriously)"]

  18. comment
    Comment #35197883

    Check out my comments higher up in the thread (eg https://news.ycombinator.com/item?id=35197613 ), I really do believe that GPT4+ will be primarily useful as augmenters for capable…

  19. comment
    Comment #35197816

    I've been able to give it arbitrary blocks of code and have it explain how they work. (Asking this makes GPT more effective when I ask it make further changes. One reason I do this…

  20. comment
    Comment #35197613

    > Reading this article I get the same sense about software engineering, if you haven't solved the problem, you don't really understand the code the AI is generating and so you don'…

  21. comment
    Comment #34963469

    Fascinating story! It reminds me of William Binney and ThinThread ( https://en.wikipedia.org/wiki/ThinThread ). (ThinThread was a system developed by a small team at the NSA that b…

  22. comment
    Comment #34957883

    I'm very interested in stories like this. > I wrote software to automate all of this. It took a half an hour, with NO manual intervention. It was strangled in the crib. Can you say…

  23. comment
    Comment #33856587

    I messed up the last few entries in the thread, the complete thread is here: https://mobile.twitter.com/isaac_lewis/status/15994780455296...

  24. story
  25. comment
    Comment #33306040

    I'd say people who work as solo entrepreneurs or independent contractors are usually either "sociopaths" (of a particular kind) or some kind of sociopath/loser mixture, depending o…