Hacker News: srdjanr

New comment by srdjanr in "IBM didn't want Microsoft to use the Tab key to move between dialog fields"

srdjanr — Tue, 05 May 2026 18:21:47 +0000

It definitely is visible, OS login screens usually warn you if caps lock is on when typing password

New comment by srdjanr in "Three Inverse Laws of AI"

srdjanr — Tue, 05 May 2026 16:54:09 +0000

The harm is in actually believing AI has wants, intentions, feelings, etc.

Saying that I killed a process won't make me more likely to believe that a process is human-like, because it's quite obviously not.

But because AI does sound like a human, anthropomorphising it will reinforce that belief.

New comment by srdjanr in "AI didn't delete your database, you did"

srdjanr — Tue, 05 May 2026 16:43:34 +0000

It's a tradeoff.

LLMs stopping and asking more would make them less useful. I'd much rather let an agent run for 1 hour, than it wanting my input every 15 mins, even if results are somewhat worse.

The real solution for security is a proper sandbox.

New comment by srdjanr in "What I'm Hearing About Cognitive Debt (So Far)"

srdjanr — Tue, 05 May 2026 09:56:35 +0000

Of course there are examples of software killing or being capable of killing (operating machinery or medical devices), but that doesn't apply to probably 90+% of software

New comment by srdjanr in "I am worried about Bun"

srdjanr — Mon, 04 May 2026 18:29:31 +0000

Genuine question: why not just wait?

If Bun stays great, you saved yourself some time for switching, and got to keep using Bun.

If Bun worsens, you spend the same time for switching, just moved a bit later, and got to use Bun for a little longer.

New comment by srdjanr in "Security through obscurity is not bad"

srdjanr — Mon, 04 May 2026 08:12:10 +0000

The author gave a few examples where compiled/minified code is public (Javascript and games) or automated vuln exploits (Wordpress example). That does explain nature of threat well enough for me.

There's a whole spectrum between 9 year old and a motivated state actor, and obfuscation is effective for a big part of the spectrum.

New comment by srdjanr in "Why has there been so little progress on Alzheimer's disease?"

srdjanr — Sun, 26 Apr 2026 10:07:00 +0000

> Nonfamilial early-onset AD can develop in people who are in their 30s or 40s, but this is extremely rare, and mostly people in their 50s or early 60s are affected.

Depending on what you consider "old", this might not change much

New comment by srdjanr in "Backblaze has stopped backing up OneDrive and Dropbox folders and maybe others"

srdjanr — Tue, 14 Apr 2026 17:30:06 +0000

It's a reference to https://news.ycombinator.com/item?id=8863

New comment by srdjanr in "The economics of software teams: Why most engineering orgs are flying blind"

srdjanr — Mon, 13 Apr 2026 08:03:55 +0000

Why is "ignorant" a synonym for "bad" (as a moral judgement, like "bad person")?

It just means you don't know something, which is usually a relatively bad situation for you, but it doesn't make you a bad person.

If you think otherwise, that's on you.

New comment by srdjanr in "The economics of software teams: Why most engineering orgs are flying blind"

srdjanr — Mon, 13 Apr 2026 07:55:15 +0000

Well flying blind is unsafe guessing (ignoring modern instruments), that's a fact. But only "flying" and "blind" together. No one thinks this makes the word "flying" has a negative connotation here, and same with "blind".

Like "drinking" and "driving". On their own, they're both neutral, but "drinking and driving" is really bad.

New comment by srdjanr in "Tofolli gates are all you need"

srdjanr — Sun, 12 Apr 2026 08:07:41 +0000

> We’re far from the ultimate physical limit, but reversibility still provides practical efficiency gains today.

I'd love to hear more about this. Where it's used today and how big are the gains?

New comment by srdjanr in "We've raised $17M to build what comes after Git"

srdjanr — Fri, 10 Apr 2026 08:40:03 +0000

The point was only that Linus didn't build git in 8 days and alone.

New comment by srdjanr in "Cloudflare targets 2029 for full post-quantum security"

srdjanr — Tue, 07 Apr 2026 21:41:35 +0000

AFAIK, PQ certificates are significantly longer than current ones. I don't know exact numbers though.

New comment by srdjanr in "Employers use your personal data to figure out the lowest salary you'll accept"

srdjanr — Mon, 06 Apr 2026 09:17:44 +0000

That's true for some jobs, but I'd be very surprised if anyone enjoys cleaning shit, for example

New comment by srdjanr in "Cross-Model Void Convergence: GPT-5.2 and Claude Opus 4.6 Deterministic Silence"

srdjanr — Sun, 22 Mar 2026 09:28:40 +0000

I don't really understand what's the point here, other than a somewhat inserting playing with LLMs. What does this tell us that's in any way applicable or points to further research? Genuinely asking

New comment by srdjanr in "The three pillars of JavaScript bloat"

srdjanr — Sun, 22 Mar 2026 09:09:20 +0000

Regarding tiny packages, I don't think they affect the size of shipped bundle at all. They only bloat your local dev environment.

New comment by srdjanr in "The three pillars of JavaScript bloat"

srdjanr — Sun, 22 Mar 2026 09:01:10 +0000

Bloat is mostly added by package authors, not website authors. And they can't know who's running it and can't look at the metrics. I doubt many website authors directly use isEven or polyfills.

New comment by srdjanr in "JSLinux Now Supports x86_64"

srdjanr — Mon, 09 Mar 2026 19:45:16 +0000

What's wrong with a well protected VM? Especially compared to something where the security selling point is "no one uses it" (according to your argument; I don't know how secure this actually is)

New comment by srdjanr in "Does that use a lot of energy?"

srdjanr — Wed, 04 Mar 2026 21:03:57 +0000

Sure, by your own argument, you should somehow increase the price of people telling other people what to avoid spending money on

New comment by srdjanr in "An AI Agent Published a Hit Piece on Me – The Operator Came Forward"

srdjanr — Fri, 20 Feb 2026 12:00:49 +0000

Regarding safety, no benchmark showed 0% misalignment. The best we had was "safest model so far" marketing speech.

Regarding predicting the future (in general, but also around AI), I'm not sure why would anyone think anything is certain, or why would you trust anyone who thinks that.

Humanity is a complex system which doesn't always have predictable output given some input (like AI advancing). And here even the input is very uncertain (we may reach "AGI" in 2 years or in 100).