Hacker News: mckn1ght

New comment by mckn1ght in "Solar panels that plug into your wall"

mckn1ght — Tue, 13 Oct 2026 03:39:11 +0000

Same boat here. I installed my own ground mounted array at double the capacity for half the quoted price from the local installer, even with an electrician coming in to do last mile hookup. Learned a ton. Got a couple fun new tools. Already have my next spot picked out.

New comment by mckn1ght in "Trump administration is suspending Microsoft from a green card program"

mckn1ght — Thu, 08 Oct 2026 17:16:11 +0000

Is it though? Markets may be perfect, but humans are not. There’s plenty of laziness, incompetence and corruption, even in capitalism.

I guess it’s the least worst option we’ve found. That doesn’t mean it doesn’t need any guardrails at all though.

New comment by mckn1ght in "Coding Is Not Solved"

mckn1ght — Mon, 28 Sep 2026 17:45:35 +0000

I came here to say something similar. We won’t need to understand the implementation. But we will need to understand the requirements. The tests, or some higher level DSL they’re (deterministically! not via LLMs) generated from, will still need to be human verified.

New comment by mckn1ght in "Rusty thoughts on "Parse, don't validate""

mckn1ght — Sun, 27 Sep 2026 19:40:09 +0000

Personally I don’t like using nil to represent specific states in my domain. I try to only judiciously use it to represent the lack of knowledge about some state, like if I need to reach out to some external service I don’t control, that may or may not give me an answer for my query (and conversely, I don’t like putting “unknown” cases into my enums).

New comment by mckn1ght in "How to write an effective software design document"

mckn1ght — Mon, 14 Sep 2026 16:20:16 +0000

I’d say don’t bother trying to parse it alone. Have a meeting where the person that created it presents it, and drill into anything weird.

If you can back them into a corner and force them to admit they don’t even understand something they supposedly wrote, use that as an opportunity to remind people it’s not acceptable to generate design documents that you don’t understand and then push them off onto other people.

New comment by mckn1ght in "Waymo pulls over, calls cops on juvenile riders who had 'ghost gun""

mckn1ght — Sat, 12 Sep 2026 18:54:26 +0000

That’s possible to do without also sending all the captured data to police.

NTSB issues investigative update on B-767 runway excursion accident in Miami

mckn1ght — Thu, 10 Sep 2026 21:30:55 +0000

Article URL: https://www.ntsb.gov:443/news/press-releases/Pages/NR20260909.aspx

Comments URL: https://news.ycombinator.com/item?id=49650418

Points: 150

# Comments: 287

New comment by mckn1ght in "New type of dice guarantees no tie when deciding who goes first"

mckn1ght — Fri, 04 Sep 2026 03:45:56 +0000

According to the wikipedia article linked at https://news.ycombinator.com/item?id=49559416 this discovery is from around the same time as that video came out. This submission should be tagged (2023).

ETA: Oh I just looked at the linked article again and it was just published. Weird.

New comment by mckn1ght in "Debian votes to allow "responsible use of generative AI""

mckn1ght — Sat, 29 Aug 2026 17:54:45 +0000

Using LLMs is basically rolling a huge number of weighted dice.

New comment by mckn1ght in "How to read more books"

mckn1ght — Mon, 13 Jul 2026 08:05:40 +0000

Two things here. First, you have to know what to ask an LLM. A book on a subject is already organized with all the information that could be returned by the same prompts that would lead to it. The prompting effort was essentially already done by the author and publisher.

Second, you’re getting a more compressed and lossy form of that information from an LLM, unless you go to such efforts that it’s be easier to just read the book. It’s akin to a conversation where a friend asks you about a book in finer and finer detail until you either recite the book for them word for word or direct them to just go read the book.

So, I don’t think LLMs are replacements for reading books. They’re complimentary.

New comment by mckn1ght in "My favorite keyboards"

mckn1ght — Thu, 02 Jul 2026 19:07:01 +0000

I went from “regular” keyboard to kinesis freestyle 2 to kinesis advantage 360.

The switch to the freestyle 2 was basically seamless. Switching to the advantage 360 was brutal, but I can tell you that i have not only fully recovered my typing speed but have even improved on my historical ability.

I used monkeytype.com for a couple years before the jump to the advantage 360 and still to this day, and you can see the deep trough my speed went through while i figured out the ortholinear layout and thumb clusters, and the new all time high records i’ve recently set.

New comment by mckn1ght in "MTG Bench: Testing how well LLMs can play Magic"

mckn1ght — Fri, 12 Jun 2026 05:25:13 +0000

Someone made a video based on the paper, if you want to see the cards being used and a little more explanation: https://www.youtube.com/watch?v=pdmODVYPDLA

New comment by mckn1ght in "World Capitals Voronoi"

mckn1ght — Thu, 11 Jun 2026 07:57:36 +0000

I was wondering what kind of metric could be used to visualize a nation’s ability to project power. Maybe some ratio involving the furthest distance from the capital city to the nation’s border?

I’m curious why the sqrt of the population in the denominator?

New comment by mckn1ght in "RIP software hackathons. Long live the hardware hackathon"

mckn1ght — Wed, 10 Jun 2026 23:29:43 +0000

I’m not following here. Marathons are very much actual races. They even give out 1st/2nd/3rd place awards and everything.

New comment by mckn1ght in "Dopamine Fracking"

mckn1ght — Mon, 08 Jun 2026 07:05:18 +0000

There’s a lot of possibility in between hunting and eating fast food. Buy some healthy food at the grocery store and pack a lunch to bring with you.

New comment by mckn1ght in "Conventional Commits encourages focus on the wrong things"

mckn1ght — Fri, 05 Jun 2026 20:03:02 +0000

I like using conventional commits but I’ve often wondered if some sort of tagging/labeling using git-notes wouldn’t be better: https://git-scm.com/docs/git-notes

I’m just unsure that the short title is the ideal place to put this kind of tagging info: the kind of fix, and optionally, the relevant component(s). I find sometimes that can take up the majority of the title.

A forge could consume the git-notes and decorate a commit/pr accordingly. Heck, GitHub PRs already have a labeling system in place, just have to add some glue.

New comment by mckn1ght in "Artificial intelligence is not conscious"

mckn1ght — Thu, 04 Jun 2026 17:13:13 +0000

Being able to play music doesn’t imply consciousness. It implies intelligence. We’ve had player pianos for ages. It’s an ability, not a phenomenology.

Being able to appreciate and enjoy music is closer to consciousness. Now how would we go about proving that an LLM does so, versus merely generating sentences that imply it does?

New comment by mckn1ght in "Citing 'severe' math deficits, UC faculty demand a return to SAT tests for STEM"

mckn1ght — Fri, 29 May 2026 04:06:13 +0000

You do still actually need the credential process though, in order to demonstrate that a person has in fact received that education.

New comment by mckn1ght in "Citing 'severe' math deficits, UC faculty demand a return to SAT tests for STEM"

mckn1ght — Fri, 29 May 2026 03:48:46 +0000

Those countries are usually quite strict about who they will subsidize though. Not a bad thing IMO.

New comment by mckn1ght in "Citing 'severe' math deficits, UC faculty demand a return to SAT tests for STEM"

mckn1ght — Fri, 29 May 2026 03:45:59 +0000

Especially when “It is difficult to get a man to understand something, when his salary depends upon his not understanding it." – Upton Sinclair