Live data from Hacker News

Viewing profile — gwerbin

gwerbin

HN member
Joined
Thu, Oct 12, 2017, 7:55 PM UTC
HN karma
910
Public activity
315 items

About gwerbin

No profile information was provided.

Recent public activity

  1. comment
    Comment #49225759

    I believe this is exactly what is happening. US DoD, and whoever else is buying. I have heard several experience reports from users of GPT 5.6 Sol and Fable 5 that the models are t…

  2. comment
    Comment #49213241

    On the other hand, the ancients assumed there had to be big continents in the southern and western hemispheres to "balance" the land in their own quadrant. They turned out to be ri…

  3. comment
    Comment #49213213

    What are the "gridded" regions in the middle? Is that some kind of measurement artifact?

  4. comment
    Comment #49210810

    Ultimately the strings are being pulled by someone who stands to gain, but it's not like Trump, Hegseth, Rubio, &co are playing 5D chess here. There are a lot of stupid blundering …

  5. comment
    Comment #49210734

    At this point it's fair to doubt just about any claim or number that comes out of the federal government. I'm sure there are still thousands of honest researchers and professionals…

  6. comment
    Comment #49204123

    Totally valid, and the Unix-style filesystem sprawl is definitely a pain sometimes. On the other hand, when an application does follow standards, you always know where something is…

  7. comment
    Comment #49204071

    No but rarely are they as well compensated as software developers. Also I think engineering team quality is bimodal, essentially a mixture of "good" teams "bad" teams. As with taxi…

  8. comment
    Comment #49196863

    Sometimes I feel like the software field is just begging to be "disrupted" by AI. So much bad management and useless wheel-spinning, all with salaries far above what is typical for…

  9. comment
    Comment #49196594

    One nice thing about English is you can just make up words with plausible etymological roots in Latin/French or Old English and often people will know what you mean. In this case t…

  10. comment
    Comment #49196522

    Yeah but what happens when it's a mix of config and local state like logs and the app database?

  11. comment
    Comment #49191867

    That's the point

  12. comment
    Comment #49191865

    Moderation by well-intentioned humans is totally different from moderation by a large corporation with a profit motive and political pressures.

  13. comment
    Comment #49182794

    I suspect some people are out there who actually like the single directory style. Obviously it's easier for the software developer, but I suspect some users like it too because eve…

  14. comment
    Comment #49182769

    I think underuse/under-marketing of XDG_STATE_HOME is a contributing factor. Developers see XDG_DATA_HOME and XDG_CONFIG_HOME and, following the spec, don't feel like their app's f…

  15. comment
    Comment #49157012

    Incidentally I think I was more productive with Opus 4.6 than with any subsequent Anthropic model.

  16. comment
    Comment #49156405

    This is the most common "not exploitable" CVE I have to deal with at work all the time. Things like buffer overruns when passing headers, but my server shouldn't be accessible to t…

  17. comment
    Comment #49147112

    I don't follow your reasoning at all. It's a shift and what people talk about and write about. It can't be anything more than that, by construction. What I see is shifting from a f…

  18. comment
    Comment #49141022

    I remember this was true of MMORPGs in the 00s. Not saying that justifies the state of things of course.

  19. comment
  20. comment
    Comment #49129756

    Agent racing sounds fun, where each agent is given a suite of increasingly hard but verifiable tasks, and the agent that completes them all correctly first, wins. You can even have…

  21. comment
    Comment #49129717

    As someone who's not a security professional I actually found it both interesting and informative. Experts usually are not the target audience for content marketing after all.

  22. comment
    Comment #49101830

    I tried this on my work M4 with 48 GB of memory and it's not as useful as I had hoped. If you set a large context window you end up using all your memory. Prefill is slow. And the …

  23. comment
    Comment #49054327

    Using it effectively is not evidence that someone is happy about it. I use it effectively for all kinds of things and I hate quite a lot of it. The anthropomorophization and cutesi…

  24. comment
    Comment #49037982

    I have never heard of this role and I'm not aware of anyone who follows it. The document you linked says that it's only used "in certain styles". So it sounds like a stylistic choi…

  25. comment
    Comment #49023850

    The rubber ducking potential is the first thing that got me interested in AI as a useful tool. You do have to be very careful of when the AI might be confidently wrong, or leading …