Hacker News: lmm

New comment by lmm in "Uber shuts operations in Nigeria and Uganda with immediate effect"

lmm — Thu, 03 Sep 2026 03:16:50 +0000

> at $156B market cap someone sure is making money.

So 50x Pets.com or 2x Enron.

> judging a mega-corporation on lemonade-stand level accounting is a fool's errand.

Disagree. Unit economics matter and can help you cut through a lot of hype and nonsense. "Where's the money coming from?" is always a good question to ask. Sure, sometimes a megacorp loses money on each transaction and makes it up on float or whatever, but that's at least worth knowing if that's what's going on.

New comment by lmm in "Bug Blindness"

lmm — Wed, 02 Sep 2026 13:15:34 +0000

> What's bad about them? There is a sensible default and users (in particular those with poor eyesight) will adjust them to their needs.

16pt is a sensible default on some screens and very much not on others.

> users (in particular those with poor eyesight) will adjust them to their needs

In the real world they won't though, much as we might both wish otherwise.

New comment by lmm in "Bug Blindness"

lmm — Sun, 30 Aug 2026 11:45:51 +0000

> What's wrong with the following CSS snippet for example?

It doesn't achieve a reasonable readable font size which was presumably the goal. Your padding won't be the right size for all screens either. (And personally I hate excessive line height).

> respect the browser's default font size

The browser's defaults are also bad. But they shouldn't be. The browser is in a position to actually fix this; the webdev isn't.

New comment by lmm in "California lawmakers unanimously pass Linux exemption from age-verification law"

lmm — Sun, 30 Aug 2026 07:43:48 +0000

What a pointlessly provocative and unproductive thing to say.

New comment by lmm in "Bug Blindness"

lmm — Sun, 30 Aug 2026 07:31:21 +0000

I am in the same boat as GP, and I wonder if it's one of the reasons I migrated hard to ebooks and online magazines/newspapers as soon as that became possible and never looked back.

New comment by lmm in "Bug Blindness"

lmm — Sun, 30 Aug 2026 07:29:16 +0000

There is no way to specify a good size in CSS - a size that looks good on current computers will be too big on old computers, and a size that looks good on old computers looks too small on current ones. Browsers and users are the ones who can fix this.

New comment by lmm in "Our decision on Cursor following its acquisition by SpaceX"

lmm — Sat, 29 Aug 2026 07:22:58 +0000

Only any who hypocritically want protection for their own models (are there any?). The morality of scraping everyone is arguable; it's when you object to being scraped back that you reveal yourself as a scoundrel.

New comment by lmm in ""IT WoRKs BeTter in the App ""

lmm — Fri, 28 Aug 2026 13:42:13 +0000

This isn't about the app being unfinished, it's about Google deliberately half-assing support for open standards. They want you to be stuck in their calendar ecosystem so they make sure there's more friction adding a calendar from outside than adding someone else's Google calendar.

New comment by lmm in "Software development with AI is starting to feel like cooking steak"

lmm — Sat, 08 Aug 2026 01:16:29 +0000

It has other uses. But the point is that saying "we" when one is not, in fact, a sovereign, is extremely pretentious.

New comment by lmm in "Taste Is All That's Left"

lmm — Fri, 07 Aug 2026 03:43:07 +0000

> I literally can't imagine generating a script with an LLM without testing it at all.

Then you're extremely unimaginative as well as unusually fastidious.

Certainly someone - several someones - are generating lots of scripts and not testing them, given the PRs I'm seeing.

New comment by lmm in "Software development with AI is starting to feel like cooking steak"

lmm — Fri, 07 Aug 2026 03:37:53 +0000

Talking about the "royal we" is a correct, established (if somewhat sarcastic) way to complain about this kind of usage. It highlights the presumptiveness of the author in their use of "we".

New comment by lmm in "Software development with AI is starting to feel like cooking steak"

lmm — Thu, 06 Aug 2026 19:06:52 +0000

No, that is exactly GP's complaint. Their whole point is that OP has written things like "we throw everything at a model and hope it gives us what we imagined, without ever having to understand how any of it actually works" when the honest statement would be "I throw everything at a model and hope it gives me what I imagined, without ever having to understand how any of it actually works".

New comment by lmm in "MPs demand answers on Fujitsu's inclusion in lucrative frameworks"

lmm — Mon, 03 Aug 2026 11:38:12 +0000

> IT projects that cost billions are incomprehensible to me, outside of frontier work

Getting a large bureaucracy to agree on a question where any answer will disadvantage some members of it is frontier work. No-one has a good general solution. Very few people even have approximate solutions for a few special cases.

> is there no space in the market for a good software consultancy? Or are they just outcompeted for getting the contract?

You can't grow big enough to bid on these big projects and stay good.

New comment by lmm in "Codex Security"

lmm — Wed, 29 Jul 2026 07:39:05 +0000

> Credit card entry, I think you should know a few basics like don't put it in MongoDB?? Or nah? It's just like any other user data?

> How about a background check - can anyone take a user-entered DL and randomly Google stuff to see what they find?

> Can I store your SSN in plain text in a text file? Why not?

You wouldn't be touching any of those unless you work for a handful of providers where that's their whole business. Usually you add the dependency, use their widget, and that's it.

New comment by lmm in "Nobody knows what a used GPU cluster is worth"

lmm — Thu, 23 Jul 2026 23:55:35 +0000

https://en.wikipedia.org/wiki/Memorylessness

New comment by lmm in "Nobody knows what a used GPU cluster is worth"

lmm — Thu, 23 Jul 2026 23:54:21 +0000

> It’s a contract

Many kinds of contracts are illegal, including those that are anticompetitive in certain ways.

> Should I be allowed to pay a landscaper less if I agree to not do part of the work myself?

No?

New comment by lmm in "Nobody knows what a used GPU cluster is worth"

lmm — Thu, 23 Jul 2026 02:29:52 +0000

Trade ins are fine. A contractual commitment to not buy hardware sounds like illegal restraint of trade.

New comment by lmm in "Nobody knows what a used GPU cluster is worth"

lmm — Thu, 23 Jul 2026 02:28:02 +0000

What are these orgs that have unused meeting rooms? Everywhere I've ever worked has had a chronic meeting room shortage.

New comment by lmm in "Nobody knows what a used GPU cluster is worth"

lmm — Thu, 23 Jul 2026 02:25:42 +0000

Most hardware has an exponential (memoryless) failure distribution in practice. The bathtub curve is a myth.

New comment by lmm in "OpenAI and Hugging Face address security incident during model evaluation"

lmm — Wed, 22 Jul 2026 07:46:17 +0000

That's been a common claim, but I don't think I've seen anyone provide actual evidence.