Hacker News: kjs3

New comment by kjs3 in "Astronomers Discover a Giant New Dwarf Planet Candidate Hiding Beyond Pluto"

kjs3 — Thu, 29 Oct 2026 03:12:24 +0000

Planet X is supposed to be Neptune sized. This thing is less than 1000 miles across. 'Giant' for a Kupier belt object; not so much for planets.

New comment by kjs3 in "AI-assisted PCB design for KiCad 10"

kjs3 — Wed, 28 Oct 2026 23:02:28 +0000

Unfortunately, that doesn't seem to be a popular ethic, particularly around here. These are the disciples of MVP and 'move fast, break things' and all the rest.

New comment by kjs3 in "Behind the scenes at the NeXT Computer office"

kjs3 — Wed, 28 Oct 2026 19:45:05 +0000

I remember NeWS being significantly slower than NeXTStep (and on a faster processor). X11 could be slow, too (VAXStation QVSS, anyone?), but flew on decent hardware. The tragedy is that with 5+ orders of magnitude more processor, 3+ orders of magnitude more memory and a GPU, I still think "crap this is laggy".

New comment by kjs3 in "Behind the scenes at the NeXT Computer office"

kjs3 — Wed, 28 Oct 2026 19:35:11 +0000

This might surprise some people, but there are pretty cool tech companies that aren't in the Bay Area.

New comment by kjs3 in "Behind the scenes at the NeXT Computer office"

kjs3 — Wed, 28 Oct 2026 19:33:42 +0000

I think the same thing, then I price a Power 11 (or Power 10) machine. Ouch. I even kinda liked AIX, even though it is definitely it's own thing. We had such hope for OpenPower but it never came to anything.

New comment by kjs3 in "Police Are Pulling People over Because AI Deems Their Driving "Suspicious""

kjs3 — Wed, 28 Oct 2026 05:20:28 +0000

I've certainly had such nudge-nudge, wink-wink interactions around reatil system stupidity, but this isn't a general solution and not every person/demographic would have such a happy ending. Indeed, you're lucky a separate anti-shoplifting system didn't flag you for 'enhanced screening' or some such taking the friendly clerk out of the equation and handing you off to the rental security drones. Then you'll just be the guy on the surveillance camera taking a part of the shelf and walking out with it (and no they probably won't care you put a part back). The more automation, the less 'human in the loop', the less opportunity for common sense to inject itself into the scenario.

New comment by kjs3 in "Police Are Pulling People over Because AI Deems Their Driving "Suspicious""

kjs3 — Tue, 27 Oct 2026 21:21:31 +0000

That's true, but it falls under the thing I keep saying: people saying "the AI said so" with no thought or checks as an excuse for pretty much anything. And AI provided probable cause is terrifying. Minority Report was not aspirational.

New comment by kjs3 in "Why Micron Is Suddenly Ditching 4F² DRAM as Rivals Double Down"

kjs3 — Tue, 27 Oct 2026 21:11:07 +0000

Wouldn't be the first once hot memory tech that ended up in the scrapheap. See: RDRAM.

New comment by kjs3 in "Why Micron Is Suddenly Ditching 4F² DRAM as Rivals Double Down"

kjs3 — Tue, 27 Oct 2026 21:09:18 +0000

Well, yes, but wouldn't this fall under Regulation S-K, Item 303? They don't have to provide forward revenue guidance, but if they're exiting a line of business don't they have to disclose future trends and risks leading to that decision in their 10Q/10K?

New comment by kjs3 in "Big Fish: Exploiting a 37-Year-Old Kernel Bug in OpenBSD"

kjs3 — Mon, 26 Oct 2026 22:39:08 +0000

You'd be surprised how many working VAXen are out there. The later desktops (the 3100s and 4000s) are pretty reliable, and you can put enough memory in them they don't (always) swap to death.

New comment by kjs3 in "E Ink Kaleido 4"

kjs3 — Mon, 26 Oct 2026 22:13:29 +0000

I wish I had just some yellow or orange for highlighting.

Yes, I would really love this as well. They make tri- and quad-color e-ink displays, but I've only ever seen the 'badge' size. I'd buy a full sized reader with just that right now.

New comment by kjs3 in "The result of the C++26 DIS ballot"

kjs3 — Mon, 26 Oct 2026 18:30:39 +0000

Cfront seems so quaint in retrospect.

New comment by kjs3 in "The Curse of Soldered RAM"

kjs3 — Mon, 26 Oct 2026 12:33:14 +0000

PATA & SATA drives don't have low level format, but SCSI did long after 1992.

New comment by kjs3 in "eMudhra to exit global SSL/TLS certificate business"

kjs3 — Thu, 22 Oct 2026 18:49:00 +0000

The Ministry of Electronics and Information Technology (MeitY) on Wednesday said it has taken “serious note” of the reported security incident

Will they send a "strongly worded letter"? That's the best way to make sure such an incident never occurs again. /s

New comment by kjs3 in "eMudhra to exit global SSL/TLS certificate business"

kjs3 — Thu, 22 Oct 2026 18:35:42 +0000

Anyone relying on them with anything related to security or trust is playing Russian roulette at this point.

In any legal dispute involving them, Exhibit A will be "look at how they ran their SSL/TLS business". That won't go well.

New comment by kjs3 in "Several Linux kernel vulnerabilities accidentally posted to public mailing list"

kjs3 — Thu, 22 Oct 2026 18:30:50 +0000

The boat long ago sailed on Mach/Hurd; there are probably better options in microkernel-land. There's Zircon, but Google laid off a bunch of the Fuchsia folks and has been pretty quite lately. QNX is still a thing, but closed source yadda yadda. Best hope there is probably the L4 family (seL4, et. al).

New comment by kjs3 in "Several Linux kernel vulnerabilities accidentally posted to public mailing list"

kjs3 — Thu, 22 Oct 2026 13:53:39 +0000

Like when graphics stack and drivers were "imported" into the WinNT kernel as an optimization for the user-space version. And thus began 30 years of "another week, another graphics subsystem crash/vuln".

New comment by kjs3 in "WorldTUI: I trained a 360M model to pretend to be a terminal"

kjs3 — Thu, 22 Oct 2026 13:46:01 +0000

That...kinda brilliant. Demented, but kinda brilliant.

New comment by kjs3 in "Florida's HOAs and condo associations are making the housing crisis worse"

kjs3 — Thu, 22 Oct 2026 01:56:00 +0000

I'm betting you spend a lot of your life telling other people your inability to understand them is their fault.

New comment by kjs3 in "Today's PHP security release, and how we're strengthening security"

kjs3 — Wed, 21 Oct 2026 18:29:48 +0000

This is probably the future, tho. Being first to disclose a vuln is a Big Deal in some quarters. And now, the velocity of AI driven pentest swarms mean every researcher who finds something worthwhile has to think "how close is the next droog's swarm to finding what I found" and the temptation to jump the gun only gets stronger.