Hacker News: flir

New comment by flir in "Making chicken transparent using Dorito dust"

flir — Sat, 17 Oct 2026 19:03:32 +0000

You made that one up.

I can see right through you.

New comment by flir in "Bark: The Toll-Free BBS for 90s Teens"

flir — Fri, 16 Oct 2026 16:00:24 +0000

I wrote a program in C that took two Blue Wave packets and compacted them into one. Probably my first "in anger" C program. I was very proud of it.

New comment by flir in "Using AI to reverse engineer hardware"

flir — Fri, 16 Oct 2026 04:53:02 +0000

> but it may just end up killing entire industries.

as did the internet before it

> but it may just end up killing entire industries.

how many record companies have been put out of business by piracy?

New comment by flir in "Wanted to download an ISO, ended up writing a kernel exploit"

flir — Thu, 15 Oct 2026 13:35:21 +0000

Definitely the simplest explanation.

But I quite like the idea that they see us as a source of polite training data first and as customers second. It aligns with my priors.

New comment by flir in "David Rudnick on "borderline impossible" process of designing Arsenal typeface"

flir — Thu, 15 Oct 2026 09:05:25 +0000

White on red, with a stripe of red through the white.

Obviously a Spurs supporter.

New comment by flir in "Who Is Planting Backdoors in Ledger Crypto Wallets?"

flir — Wed, 14 Oct 2026 12:17:00 +0000

Hm. Computation with paper and pencil?

(I agree in general, I'm just exploring the statement).

New comment by flir in "Muse Wonder"

flir — Tue, 13 Oct 2026 11:05:19 +0000

The winner is the first company to RELEASE THE HYPNODRONES

(Markets are a human construct, not a law of nature. There's no reason to think they'll reward intelligence).

New comment by flir in "AI Is a Tool. Using It Well Takes Skill."

flir — Mon, 12 Oct 2026 23:14:36 +0000

I'm thinking about that plus time-to-market which would be what, a decade?

New comment by flir in "AI is a tool and using it well takes skill"

flir — Mon, 12 Oct 2026 22:47:17 +0000

Is it worth mentioning that the Flynn Effect seems to have been running backwards lately?

(The drug thing, maybe you've got a point. But in just 20 years?)

New comment by flir in "QGIS enacts moratorium on AI assisted contributions"

flir — Mon, 12 Oct 2026 20:20:43 +0000

Then I encourage you to kick down that fence.

New comment by flir in "Codex CLI installs a persistent, self-updating background daemon by default"

flir — Mon, 12 Oct 2026 20:05:52 +0000

Both CLIs are getting worse. I think it's because they keep bolting in unconsidered "features", not because of the code quality.

But maybe being able to code up a feature at the drop of a hat has turbo-charged their ability to make bad UX decisions.

New comment by flir in "QGIS: No A contributions permitted"

flir — Mon, 12 Oct 2026 11:41:54 +0000

Go for it. I bet you could do it in a week.

I'll fork it in about 30 seconds.

And neither of us will get any traction, because the QGIS team has a history and proven track record within the GIS community, and neither of us are attempting to solve a problem that community has.

New comment by flir in "Programmable pigments: Photochromic nanoparticle inks for rewritable tattoos"

flir — Mon, 12 Oct 2026 04:48:54 +0000

Regardless of the actual tech here, "he turned his tattoos off" has "the door dilated" vibes. Very cool.

New comment by flir in "A sustainable web career, for when all this blows over"

flir — Sun, 11 Oct 2026 18:16:01 +0000

That's fair.

New comment by flir in "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"

flir — Fri, 09 Oct 2026 08:31:52 +0000

> This won’t last forever

It probably will. Moore's Law is still churning away in the background.

Frontier models might get more expensive, but that's a moving target. For any particular capability point, the models will only get cheaper.

New comment by flir in "“Math 2.0” will need to value mathematical progress more holistically"

flir — Thu, 08 Oct 2026 08:51:00 +0000

But these days hardware is software. One microcontroller replaces so much electronics. Software is eating the world, as the man said.

New comment by flir in "A sustainable web career, for when all this blows over"

flir — Tue, 06 Oct 2026 21:29:56 +0000

Not everyone, but it's one possible path for a company when the VC money burns out. It's more likely than these companies just dragging the model to the trashcan and declaring bankruptcy IMO.

Take a look at this: https://en.wikipedia.org/wiki/Koomey%27s_law

2.6 years is about 16x in a decade. That's another path out - inference on GPUs gets so cheap they become profitable at current prices.

New comment by flir in "A sustainable web career, for when all this blows over"

flir — Tue, 06 Oct 2026 21:13:13 +0000

You think LLMs are on a par with NFTs?

Huh.

New comment by flir in "A sustainable web career, for when all this blows over"

flir — Tue, 06 Oct 2026 21:11:35 +0000

Koomey's law is a pretty safe bet.

When Taalas burned Llama 3.1 to the HC1 they saw at least an order of magnitude more tokens per joule. That saving is sitting right there, today, ready for anyone who is willing to say "this is the model we're freezing on for a couple of years".

New comment by flir in "A sustainable web career, for when all this blows over"

flir — Tue, 06 Oct 2026 20:49:46 +0000

After the trough of disillusionment comes the plateau of productivity.

All those previous silver bullets are still with us. This one's not going away either.