Hacker News: queuebert

New comment by queuebert in "Tesla killing Solar Roof is leaving installers with six-figure losses"

queuebert — Mon, 07 Sep 2026 20:21:02 +0000

Is it? Modern roofing is pretty complex and is begging for innovation, even with materials that don't generate electricity.

New comment by queuebert in "Caltech Mathathon – first hackathon ever devoted to research level mathematics"

queuebert — Mon, 07 Sep 2026 20:17:06 +0000

Serious question: Are you sure solving abstract mathematical problems with AI is responsible use? You will likely put mathematicians out of jobs, and I doubt solving the Collatz conjecture is urgent or will save lives. It also robs a future Fields medalist of the pride of doing all by themselves.

All things AI seems to assume that more and faster is better, but there is no justification of that assumption. As a biological counterexample, a tree grown quickly will likely not be as healthy or strong as one grown slowly.

New comment by queuebert in "Mother convicted of: her 5 yo walks short way to pond alone in Virginia"

queuebert — Mon, 07 Sep 2026 04:47:14 +0000

It's not backward, it's too forward. No one got arrested 100 yrs ago for letting their kid walk around alone.

New comment by queuebert in "Research acceleration: The view inside OpenAI"

queuebert — Mon, 07 Sep 2026 04:45:06 +0000

/loop ?

New comment by queuebert in "Research acceleration: The view inside OpenAI"

queuebert — Mon, 07 Sep 2026 04:44:16 +0000

Yes, because no employee with million-dollar comp has ever been untrustworthy in the history of business.

New comment by queuebert in "Dwarf Fortress is getting the mother of all magic updates"

queuebert — Tue, 01 Sep 2026 18:34:52 +0000

That's impressive. I thought the NetHack wiki would be older, but it dates to 2005, although I suspect there was an earlier version somewhere else.

New comment by queuebert in "All 253 Patterns from Christopher Alexander's a Pattern Language Summarized"

queuebert — Thu, 23 Jul 2026 04:27:00 +0000

I highly recommend getting the full book. This should be required reading for every housing developer. If you ever felt like a house or community was uncomfortable or made you subconsciously unhappy, you might find the reason in this book.

New comment by queuebert in "Goes-19 weather satellite enters Safe Hold mode"

queuebert — Thu, 16 Jul 2026 18:18:11 +0000

I'm sure that was 100% secure, too. /s (I did the same thing back then.)

New comment by queuebert in "Goes-19 weather satellite enters Safe Hold mode"

queuebert — Thu, 16 Jul 2026 18:13:10 +0000

The link OP submitted appears to be a webpage displaying a screenshot of another web page, and the image aspect ratio has been altered. It's so comically bad it had to be on purpose, or someone is doing their web dev in MS Word.

Edit: I think actually it's a screenshot of a screenshot even, and this appears to be the entire design of spaceweather.gov. What in the holy heck is going on there? This has to be a top 10 worst website designs of all time.

New comment by queuebert in "Midjourney Medical"

queuebert — Thu, 18 Jun 2026 15:39:09 +0000

Statistically overdetection leads to poorer outcomes because interventions have a risk as well. That's why everyone doesn't get a yearly full body CT scan, for example. The current guidelines are based on optimizing for maximum good, and believe it or not some things are best not known about because the risk of dying from it is about the same as the risk of the treatment.

New comment by queuebert in "SpaceX, Other Mega IPOs Denied Fast Index Entry by S&P"

queuebert — Fri, 12 Jun 2026 18:38:16 +0000

Markets are forward looking, pricing in future cash flows. No one should be thinking about SpaceX's present value when buying their stock. If you do that you'll always be too late to the party.

That said, I don't agree on the valuation, but I can see where some people could be coming from.

New comment by queuebert in "Claude Fable 5: mid-tier results on coding tasks"

queuebert — Fri, 12 Jun 2026 18:27:45 +0000

It's a joke about how fast it compiles. whoosh

New comment by queuebert in "Claude Fable 5: mid-tier results on coding tasks"

queuebert — Fri, 12 Jun 2026 15:51:28 +0000

Why is that?

New comment by queuebert in "Claude Fable 5: mid-tier results on coding tasks"

queuebert — Fri, 12 Jun 2026 14:28:08 +0000

> ... applying the same steps and rules n times

I do this too, with a document written for this purpose.

> ... a significant chunk of those runs involved the agent waiting for the compiler, linters, type checks, and test suites, as well as updating journals.

That is a good point. I'm mostly using C, which seemingly compiles in O(1) time, so I could imagine a large C++ or Rust codebase taking much longer to iterate simply due to compilation times.

New comment by queuebert in "Claude Fable 5: mid-tier results on coding tasks"

queuebert — Thu, 11 Jun 2026 22:03:31 +0000

I don't understand how some of y'all use these things. I get garbage unless I give them very specific concrete tasks with as much context as possible. Anything that takes more than 30 min is usually a waste because the scope was too large.

New comment by queuebert in "Google to pay SpaceX $920M a month for compute capacity at xAI data centers"

queuebert — Sun, 07 Jun 2026 03:41:23 +0000

Worth noting the cost of dealing with Treasury's absolute dumpster fire of a website, though.

New comment by queuebert in "SpaceX, Other Mega IPOs Denied Fast Index Entry by S&P"

queuebert — Fri, 05 Jun 2026 18:06:14 +0000

I didn't say that at all. I said it was achievable for SpaceX eventually. It's not a $1T company yet. Reading comprehension, people.

New comment by queuebert in "SpaceX, Other Mega IPOs Denied Fast Index Entry by S&P"

queuebert — Fri, 05 Jun 2026 02:31:12 +0000

Historically a $1T market cap with a PE of 20.0 would be achievable with a $50B/yr profit. That seems easily achievable eventually for SpaceX, as it has actual hardware and services and IP.

New comment by queuebert in "Uber cuts 23% of people division as new president takes over"

queuebert — Wed, 03 Jun 2026 20:24:24 +0000

"What would you say you do here?"

"A lot of things."

You still gave zero concrete information about why HR is necessary. Which is exactly the problem here.

New comment by queuebert in "Uber cuts 23% of people division as new president takes over"

queuebert — Wed, 03 Jun 2026 17:37:39 +0000

As we enter this era of far more qualified candidates than jobs, HR will die eventually the equivalent of index funds is for hiring. Just as most money managers didn't beat the market and lost out to Bogle's low-cost index funds, people will figure out that HR doesn't do any better than any other random criteria for hiring and firing employees, since most of the applicants for most jobs will be able to do the job sufficiently well. Probably the answer is some sort of AI, but I bet you could do just as well rolling dice.

If most of us are honestly with ourselves, we'd realize the marginal return on difficult hiring decisions is extremely small.

As for the CYA aspect of HR, an AI can definitely do that cheaper and more callously.